diff options
1099 files changed, 28692 insertions, 14468 deletions
diff --git a/Documentation/admin-guide/pm/intel_pstate.rst b/Documentation/admin-guide/pm/intel_pstate.rst index d2b6fda3d67b..ab2fe0eda1d7 100644 --- a/Documentation/admin-guide/pm/intel_pstate.rst +++ b/Documentation/admin-guide/pm/intel_pstate.rst @@ -145,7 +145,7 @@ feature enabled.] In this mode ``intel_pstate`` registers utilization update callbacks with the CPU scheduler in order to run a P-state selection algorithm, either -``powersave`` or ``performance``, depending on the ``scaling_cur_freq`` policy +``powersave`` or ``performance``, depending on the ``scaling_governor`` policy setting in ``sysfs``. The current CPU frequency information to be made available from the ``scaling_cur_freq`` policy attribute in ``sysfs`` is periodically updated by those utilization update callbacks too. diff --git a/Documentation/admin-guide/pm/sleep-states.rst b/Documentation/admin-guide/pm/sleep-states.rst index 1e5c0f00cb2f..dbf5acd49f35 100644 --- a/Documentation/admin-guide/pm/sleep-states.rst +++ b/Documentation/admin-guide/pm/sleep-states.rst @@ -15,7 +15,7 @@ Sleep States That Can Be Supported ================================== Depending on its configuration and the capabilities of the platform it runs on, -the Linux kernel can support up to four system sleep states, includig +the Linux kernel can support up to four system sleep states, including hibernation and up to three variants of system suspend. The sleep states that can be supported by the kernel are listed below. diff --git a/Documentation/bpf/bpf_devel_QA.txt b/Documentation/bpf/bpf_devel_QA.txt index 1a0b704e1a38..da57601153a0 100644 --- a/Documentation/bpf/bpf_devel_QA.txt +++ b/Documentation/bpf/bpf_devel_QA.txt @@ -557,6 +557,14 @@ A: Although LLVM IR generation and optimization try to stay architecture pulls in some header files containing file scope host assembly codes. - You can add "-fno-jump-tables" to work around the switch table issue. - Otherwise, you can use bpf target. + Otherwise, you can use bpf target. Additionally, you _must_ use bpf target + when: + + - Your program uses data structures with pointer or long / unsigned long + types that interface with BPF helpers or context data structures. Access + into these structures is verified by the BPF verifier and may result + in verification failures if the native architecture is not aligned with + the BPF architecture, e.g. 64-bit. An example of this is + BPF_PROG_TYPE_SK_MSG require '-target bpf' Happy BPF hacking! diff --git a/Documentation/device-mapper/thin-provisioning.txt b/Documentation/device-mapper/thin-provisioning.txt index 4bcd4b7f79f9..3d01948ea061 100644 --- a/Documentation/device-mapper/thin-provisioning.txt +++ b/Documentation/device-mapper/thin-provisioning.txt @@ -264,7 +264,10 @@ i) Constructor data device, but just remove the mapping. read_only: Don't allow any changes to be made to the pool - metadata. + metadata. This mode is only available after the + thin-pool has been created and first used in full + read/write mode. It cannot be specified on initial + thin-pool creation. error_if_no_space: Error IOs, instead of queueing, if no space. diff --git a/Documentation/devicetree/bindings/ata/ahci-platform.txt b/Documentation/devicetree/bindings/ata/ahci-platform.txt index f4006d3c9fdf..c760ecb81381 100644 --- a/Documentation/devicetree/bindings/ata/ahci-platform.txt +++ b/Documentation/devicetree/bindings/ata/ahci-platform.txt @@ -30,7 +30,6 @@ compatible: Optional properties: - dma-coherent : Present if dma operations are coherent - clocks : a list of phandle + clock specifier pairs -- resets : a list of phandle + reset specifier pairs - target-supply : regulator for SATA target power - phys : reference to the SATA PHY node - phy-names : must be "sata-phy" diff --git a/Documentation/devicetree/bindings/display/bridge/tda998x.txt b/Documentation/devicetree/bindings/display/bridge/tda998x.txt index 24cc2466185a..1a4eaca40d94 100644 --- a/Documentation/devicetree/bindings/display/bridge/tda998x.txt +++ b/Documentation/devicetree/bindings/display/bridge/tda998x.txt @@ -27,6 +27,9 @@ Optional properties: in question is used. The implementation allows one or two DAIs. If two DAIs are defined, they must be of different type. + - nxp,calib-gpios: calibration GPIO, which must correspond with the + gpio used for the TDA998x interrupt pin. + [1] Documentation/sound/alsa/soc/DAI.txt [2] include/dt-bindings/display/tda998x.h diff --git a/Documentation/devicetree/bindings/display/exynos/exynos5433-decon.txt b/Documentation/devicetree/bindings/display/exynos/exynos5433-decon.txt index fc2588292a68..775193e1c641 100644 --- a/Documentation/devicetree/bindings/display/exynos/exynos5433-decon.txt +++ b/Documentation/devicetree/bindings/display/exynos/exynos5433-decon.txt @@ -19,7 +19,8 @@ Required properties: clock-names property. - clock-names: list of clock names sorted in the same order as the clocks property. Must contain "pclk", "aclk_decon", "aclk_smmu_decon0x", - "aclk_xiu_decon0x", "pclk_smmu_decon0x", clk_decon_vclk", + "aclk_xiu_decon0x", "pclk_smmu_decon0x", "aclk_smmu_decon1x", + "aclk_xiu_decon1x", "pclk_smmu_decon1x", clk_decon_vclk", "sclk_decon_eclk" - ports: contains a port which is connected to mic node. address-cells and size-cells must 1 and 0, respectively. @@ -34,10 +35,14 @@ decon: decon@13800000 { clocks = <&cmu_disp CLK_ACLK_DECON>, <&cmu_disp CLK_ACLK_SMMU_DECON0X>, <&cmu_disp CLK_ACLK_XIU_DECON0X>, <&cmu_disp CLK_PCLK_SMMU_DECON0X>, + <&cmu_disp CLK_ACLK_SMMU_DECON1X>, + <&cmu_disp CLK_ACLK_XIU_DECON1X>, + <&cmu_disp CLK_PCLK_SMMU_DECON1X>, <&cmu_disp CLK_SCLK_DECON_VCLK>, <&cmu_disp CLK_SCLK_DECON_ECLK>; clock-names = "aclk_decon", "aclk_smmu_decon0x", "aclk_xiu_decon0x", - "pclk_smmu_decon0x", "sclk_decon_vclk", "sclk_decon_eclk"; + "pclk_smmu_decon0x", "aclk_smmu_decon1x", "aclk_xiu_decon1x", + "pclk_smmu_decon1x", "sclk_decon_vclk", "sclk_decon_eclk"; interrupt-names = "vsync", "lcd_sys"; interrupts = <0 202 0>, <0 203 0>; diff --git a/Documentation/devicetree/bindings/display/panel/panel-common.txt b/Documentation/devicetree/bindings/display/panel/panel-common.txt index 557fa765adcb..5d2519af4bb5 100644 --- a/Documentation/devicetree/bindings/display/panel/panel-common.txt +++ b/Documentation/devicetree/bindings/display/panel/panel-common.txt @@ -38,7 +38,7 @@ Display Timings require specific display timings. The panel-timing subnode expresses those timings as specified in the timing subnode section of the display timing bindings defined in - Documentation/devicetree/bindings/display/display-timing.txt. + Documentation/devicetree/bindings/display/panel/display-timing.txt. Connectivity diff --git a/Documentation/devicetree/bindings/dma/renesas,rcar-dmac.txt b/Documentation/devicetree/bindings/dma/renesas,rcar-dmac.txt index aadfb236d53a..61315eaa7660 100644 --- a/Documentation/devicetree/bindings/dma/renesas,rcar-dmac.txt +++ b/Documentation/devicetree/bindings/dma/renesas,rcar-dmac.txt @@ -26,6 +26,7 @@ Required Properties: - "renesas,dmac-r8a7794" (R-Car E2) - "renesas,dmac-r8a7795" (R-Car H3) - "renesas,dmac-r8a7796" (R-Car M3-W) + - "renesas,dmac-r8a77965" (R-Car M3-N) - "renesas,dmac-r8a77970" (R-Car V3M) - "renesas,dmac-r8a77980" (R-Car V3H) diff --git a/Documentation/devicetree/bindings/input/atmel,maxtouch.txt b/Documentation/devicetree/bindings/input/atmel,maxtouch.txt index 23e3abc3fdef..c88919480d37 100644 --- a/Documentation/devicetree/bindings/input/atmel,maxtouch.txt +++ b/Documentation/devicetree/bindings/input/atmel,maxtouch.txt @@ -4,6 +4,13 @@ Required properties: - compatible: atmel,maxtouch + The following compatibles have been used in various products but are + deprecated: + atmel,qt602240_ts + atmel,atmel_mxt_ts + atmel,atmel_mxt_tp + atmel,mXT224 + - reg: The I2C address of the device - interrupts: The sink for the touchpad's IRQ output diff --git a/Documentation/devicetree/bindings/net/can/rcar_canfd.txt b/Documentation/devicetree/bindings/net/can/rcar_canfd.txt index 93c3a6ae32f9..ac71daa46195 100644 --- a/Documentation/devicetree/bindings/net/can/rcar_canfd.txt +++ b/Documentation/devicetree/bindings/net/can/rcar_canfd.txt @@ -5,7 +5,9 @@ Required properties: - compatible: Must contain one or more of the following: - "renesas,rcar-gen3-canfd" for R-Car Gen3 compatible controller. - "renesas,r8a7795-canfd" for R8A7795 (R-Car H3) compatible controller. - - "renesas,r8a7796-canfd" for R8A7796 (R-Car M3) compatible controller. + - "renesas,r8a7796-canfd" for R8A7796 (R-Car M3-W) compatible controller. + - "renesas,r8a77970-canfd" for R8A77970 (R-Car V3M) compatible controller. + - "renesas,r8a77980-canfd" for R8A77980 (R-Car V3H) compatible controller. When compatible with the generic version, nodes must list the SoC-specific version corresponding to the platform first, followed by the diff --git a/Documentation/devicetree/bindings/net/renesas,ravb.txt b/Documentation/devicetree/bindings/net/renesas,ravb.txt index c306f55d335b..890526dbfc26 100644 --- a/Documentation/devicetree/bindings/net/renesas,ravb.txt +++ b/Documentation/devicetree/bindings/net/renesas,ravb.txt @@ -18,6 +18,7 @@ Required properties: - "renesas,etheravb-r8a7795" for the R8A7795 SoC. - "renesas,etheravb-r8a7796" for the R8A7796 SoC. + - "renesas,etheravb-r8a77965" for the R8A77965 SoC. - "renesas,etheravb-r8a77970" for the R8A77970 SoC. - "renesas,etheravb-r8a77980" for the R8A77980 SoC. - "renesas,etheravb-r8a77995" for the R8A77995 SoC. diff --git a/Documentation/devicetree/bindings/pinctrl/allwinner,sunxi-pinctrl.txt b/Documentation/devicetree/bindings/pinctrl/allwinner,sunxi-pinctrl.txt index ed5eb547afc8..64bc5c2a76da 100644 --- a/Documentation/devicetree/bindings/pinctrl/allwinner,sunxi-pinctrl.txt +++ b/Documentation/devicetree/bindings/pinctrl/allwinner,sunxi-pinctrl.txt @@ -56,9 +56,9 @@ pins it needs, and how they should be configured, with regard to muxer configuration, drive strength and pullups. If one of these options is not set, its actual value will be unspecified. -This driver supports the generic pin multiplexing and configuration -bindings. For details on each properties, you can refer to -./pinctrl-bindings.txt. +Allwinner A1X Pin Controller supports the generic pin multiplexing and +configuration bindings. For details on each properties, you can refer to + ./pinctrl-bindings.txt. Required sub-node properties: - pins diff --git a/Documentation/devicetree/bindings/serial/renesas,sci-serial.txt b/Documentation/devicetree/bindings/serial/renesas,sci-serial.txt index a006ea4d065f..106808b55b6d 100644 --- a/Documentation/devicetree/bindings/serial/renesas,sci-serial.txt +++ b/Documentation/devicetree/bindings/serial/renesas,sci-serial.txt @@ -43,6 +43,8 @@ Required properties: - "renesas,hscif-r8a7795" for R8A7795 (R-Car H3) HSCIF compatible UART. - "renesas,scif-r8a7796" for R8A7796 (R-Car M3-W) SCIF compatible UART. - "renesas,hscif-r8a7796" for R8A7796 (R-Car M3-W) HSCIF compatible UART. + - "renesas,scif-r8a77965" for R8A77965 (R-Car M3-N) SCIF compatible UART. + - "renesas,hscif-r8a77965" for R8A77965 (R-Car M3-N) HSCIF compatible UART. - "renesas,scif-r8a77970" for R8A77970 (R-Car V3M) SCIF compatible UART. - "renesas,hscif-r8a77970" for R8A77970 (R-Car V3M) HSCIF compatible UART. - "renesas,scif-r8a77980" for R8A77980 (R-Car V3H) SCIF compatible UART. diff --git a/Documentation/devicetree/bindings/vendor-prefixes.txt b/Documentation/devicetree/bindings/vendor-prefixes.txt index b5f978a4cac6..a38d8bfae19c 100644 --- a/Documentation/devicetree/bindings/vendor-prefixes.txt +++ b/Documentation/devicetree/bindings/vendor-prefixes.txt @@ -182,6 +182,7 @@ karo Ka-Ro electronics GmbH keithkoep Keith & Koep GmbH keymile Keymile GmbH khadas Khadas +kiebackpeter Kieback & Peter GmbH kinetic Kinetic Technologies kingnovel Kingnovel Technology Co., Ltd. kosagi Sutajio Ko-Usagi PTE Ltd. diff --git a/Documentation/devicetree/overlay-notes.txt b/Documentation/devicetree/overlay-notes.txt index a4feb6dde8cd..725fb8d255c1 100644 --- a/Documentation/devicetree/overlay-notes.txt +++ b/Documentation/devicetree/overlay-notes.txt @@ -98,6 +98,14 @@ Finally, if you need to remove all overlays in one-go, just call of_overlay_remove_all() which will remove every single one in the correct order. +In addition, there is the option to register notifiers that get called on +overlay operations. See of_overlay_notifier_register/unregister and +enum of_overlay_notify_action for details. + +Note that a notifier callback is not supposed to store pointers to a device +tree node or its content beyond OF_OVERLAY_POST_REMOVE corresponding to the +respective node it received. + Overlay DTS Format ------------------ diff --git a/Documentation/doc-guide/parse-headers.rst b/Documentation/doc-guide/parse-headers.rst index 96a0423d5dba..82a3e43b6864 100644 --- a/Documentation/doc-guide/parse-headers.rst +++ b/Documentation/doc-guide/parse-headers.rst @@ -177,14 +177,14 @@ BUGS **** -Report bugs to Mauro Carvalho Chehab <mchehab@s-opensource.com> +Report bugs to Mauro Carvalho Chehab <mchehab@kernel.org> COPYRIGHT ********* -Copyright (c) 2016 by Mauro Carvalho Chehab <mchehab@s-opensource.com>. +Copyright (c) 2016 by Mauro Carvalho Chehab <mchehab+samsung@kernel.org>. License GPLv2: GNU GPL version 2 <http://gnu.org/licenses/gpl.html>. diff --git a/Documentation/media/uapi/rc/keytable.c.rst b/Documentation/media/uapi/rc/keytable.c.rst index e6ce1e3f5a78..217237f93b37 100644 --- a/Documentation/media/uapi/rc/keytable.c.rst +++ b/Documentation/media/uapi/rc/keytable.c.rst @@ -7,7 +7,7 @@ file: uapi/v4l/keytable.c /* keytable.c - This program allows checking/replacing keys at IR - Copyright (C) 2006-2009 Mauro Carvalho Chehab <mchehab@infradead.org> + Copyright (C) 2006-2009 Mauro Carvalho Chehab <mchehab@kernel.org> This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by diff --git a/Documentation/media/uapi/v4l/v4l2grab.c.rst b/Documentation/media/uapi/v4l/v4l2grab.c.rst index 5aabd0b7b089..f0d0ab6abd41 100644 --- a/Documentation/media/uapi/v4l/v4l2grab.c.rst +++ b/Documentation/media/uapi/v4l/v4l2grab.c.rst @@ -6,7 +6,7 @@ file: media/v4l/v4l2grab.c .. code-block:: c /* V4L2 video picture grabber - Copyright (C) 2009 Mauro Carvalho Chehab <mchehab@infradead.org> + Copyright (C) 2009 Mauro Carvalho Chehab <mchehab@kernel.org> This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by diff --git a/Documentation/sphinx/parse-headers.pl b/Documentation/sphinx/parse-headers.pl index a958d8b5e99d..d410f47567e9 100755 --- a/Documentation/sphinx/parse-headers.pl +++ b/Documentation/sphinx/parse-headers.pl @@ -387,11 +387,11 @@ tree for more details. =head1 BUGS -Report bugs to Mauro Carvalho Chehab <mchehab@s-opensource.com> +Report bugs to Mauro Carvalho Chehab <mchehab@kernel.org> =head1 COPYRIGHT -Copyright (c) 2016 by Mauro Carvalho Chehab <mchehab@s-opensource.com>. +Copyright (c) 2016 by Mauro Carvalho Chehab <mchehab+samsung@kernel.org>. License GPLv2: GNU GPL version 2 <http://gnu.org/licenses/gpl.html>. diff --git a/Documentation/translations/zh_CN/video4linux/v4l2-framework.txt b/Documentation/translations/zh_CN/video4linux/v4l2-framework.txt index 698660b7f21f..c77c0f060864 100644 --- a/Documentation/translations/zh_CN/video4linux/v4l2-framework.txt +++ b/Documentation/translations/zh_CN/video4linux/v4l2-framework.txt @@ -6,7 +6,7 @@ communicating in English you can also ask the Chinese maintainer for help. Contact the Chinese maintainer if this translation is outdated or if there is a problem with the translation. -Maintainer: Mauro Carvalho Chehab <mchehab@infradead.org> +Maintainer: Mauro Carvalho Chehab <mchehab@kernel.org> Chinese maintainer: Fu Wei <tekkamanninja@gmail.com> --------------------------------------------------------------------- Documentation/video4linux/v4l2-framework.txt 的中文翻译 @@ -14,7 +14,7 @@ Documentation/video4linux/v4l2-framework.txt 的中文翻译 如果想评论或更新本文的内容，请直接联系原文档的维护者。如果你使用英文 交流有困难的话，也可以向中文版维护者求助。如果本翻译更新不及时或者翻 译存在问题，请联系中文版维护者。 -英文版维护者： Mauro Carvalho Chehab <mchehab@infradead.org> +英文版维护者： Mauro Carvalho Chehab <mchehab@kernel.org> 中文版维护者： 傅炜 Fu Wei <tekkamanninja@gmail.com> 中文版翻译者： 傅炜 Fu Wei <tekkamanninja@gmail.com> 中文版校译者： 傅炜 Fu Wei <tekkamanninja@gmail.com> diff --git a/MAINTAINERS b/MAINTAINERS index ec7b1d890bf6..c7fe1c331bb1 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -137,9 +137,9 @@ Maintainers List (try to look for most precise areas first) ----------------------------------- 3C59X NETWORK DRIVER -M: Steffen Klassert <klassert@mathematik.tu-chemnitz.de> +M: Steffen Klassert <klassert@kernel.org> L: netdev@vger.kernel.org -S: Maintained +S: Odd Fixes F: Documentation/networking/vortex.txt F: drivers/net/ethernet/3com/3c59x.c @@ -2556,7 +2556,6 @@ F: Documentation/devicetree/bindings/sound/axentia,* F: sound/soc/atmel/tse850-pcm5142.c AZ6007 DVB DRIVER -M: Mauro Carvalho Chehab <mchehab@s-opensource.com> M: Mauro Carvalho Chehab <mchehab@kernel.org> L: linux-media@vger.kernel.org W: https://linuxtv.org @@ -3085,7 +3084,6 @@ F: include/linux/btrfs* F: include/uapi/linux/btrfs* BTTV VIDEO4LINUX DRIVER -M: Mauro Carvalho Chehab <mchehab@s-opensource.com> M: Mauro Carvalho Chehab <mchehab@kernel.org> L: linux-media@vger.kernel.org W: https://linuxtv.org @@ -3695,7 +3693,6 @@ F: drivers/cpufreq/arm_big_little_dt.c CPU POWER MONITORING SUBSYSTEM M: Thomas Renninger <trenn@suse.com> -M: Shuah Khan <shuahkh@osg.samsung.com> M: Shuah Khan <shuah@kernel.org> L: linux-pm@vger.kernel.org S: Maintained @@ -3814,7 +3811,6 @@ S: Maintained F: drivers/media/dvb-frontends/cx24120* CX88 VIDEO4LINUX DRIVER -M: Mauro Carvalho Chehab <mchehab@s-opensource.com> M: Mauro Carvalho Chehab <mchehab@kernel.org> L: linux-media@vger.kernel.org W: https://linuxtv.org @@ -4675,7 +4671,7 @@ F: Documentation/devicetree/bindings/display/exynos/ DRM DRIVERS FOR FREESCALE DCU M: Stefan Agner <stefan@agner.ch> -M: Alison Wang <alison.wang@freescale.com> +M: Alison Wang <alison.wang@nxp.com> L: dri-devel@lists.freedesktop.org S: Supported F: drivers/gpu/drm/fsl-dcu/ @@ -5072,7 +5068,6 @@ F: drivers/edac/thunderx_edac* EDAC-CORE M: Borislav Petkov <bp@alien8.de> -M: Mauro Carvalho Chehab <mchehab@s-opensource.com> M: Mauro Carvalho Chehab <mchehab@kernel.org> L: linux-edac@vger.kernel.org T: git git://git.kernel.org/pub/scm/linux/kernel/git/bp/bp.git for-next @@ -5101,7 +5096,6 @@ S: Maintained F: drivers/edac/fsl_ddr_edac.* EDAC-GHES -M: Mauro Carvalho Chehab <mchehab@s-opensource.com> M: Mauro Carvalho Chehab <mchehab@kernel.org> L: linux-edac@vger.kernel.org S: Maintained @@ -5118,21 +5112,18 @@ S: Maintained F: drivers/edac/i5000_edac.c EDAC-I5400 -M: Mauro Carvalho Chehab <mchehab@s-opensource.com> M: Mauro Carvalho Chehab <mchehab@kernel.org> L: linux-edac@vger.kernel.org S: Maintained F: drivers/edac/i5400_edac.c EDAC-I7300 -M: Mauro Carvalho Chehab <mchehab@s-opensource.com> M: Mauro Carvalho Chehab <mchehab@kernel.org> L: linux-edac@vger.kernel.org S: Maintained F: drivers/edac/i7300_edac.c EDAC-I7CORE -M: Mauro Carvalho Chehab <mchehab@s-opensource.com> M: Mauro Carvalho Chehab <mchehab@kernel.org> L: linux-edac@vger.kernel.org S: Maintained @@ -5182,7 +5173,6 @@ S: Maintained F: drivers/edac/r82600_edac.c EDAC-SBRIDGE -M: Mauro Carvalho Chehab <mchehab@s-opensource.com> M: Mauro Carvalho Chehab <mchehab@kernel.org> L: linux-edac@vger.kernel.org S: Maintained @@ -5241,7 +5231,6 @@ S: Maintained F: drivers/net/ethernet/ibm/ehea/ EM28XX VIDEO4LINUX DRIVER -M: Mauro Carvalho Chehab <mchehab@s-opensource.com> M: Mauro Carvalho Chehab <mchehab@kernel.org> L: linux-media@vger.kernel.org W: https://linuxtv.org @@ -7696,9 +7685,11 @@ L: linux-kbuild@vger.kernel.org S: Maintained F: Documentation/kbuild/ F: Makefile -F: scripts/Makefile.* +F: scripts/Kbuild* +F: scripts/Makefile* F: scripts/basic/ F: scripts/mk* +F: scripts/mod/ F: scripts/package/ KERNEL JANITORS @@ -7723,7 +7714,6 @@ F: include/linux/sunrpc/ F: include/uapi/linux/sunrpc/ KERNEL SELFTEST FRAMEWORK -M: Shuah Khan <shuahkh@osg.samsung.com> M: Shuah Khan <shuah@kernel.org> L: linux-kselftest@vger.kernel.org T: git git://git.kernel.org/pub/scm/linux/kernel/git/shuah/linux-kselftest.git @@ -8890,7 +8880,6 @@ F: Documentation/devicetree/bindings/media/nvidia,tegra-vde.txt F: drivers/staging/media/tegra-vde/ MEDIA INPUT INFRASTRUCTURE (V4L/DVB) -M: Mauro Carvalho Chehab <mchehab@s-opensource.com> M: Mauro Carvalho Chehab <mchehab@kernel.org> P: LinuxTV.org Project L: linux-media@vger.kernel.org @@ -9744,6 +9733,7 @@ W: https://fedorahosted.org/dropwatch/ F: net/core/drop_monitor.c NETWORKING DRIVERS +M: "David S. Miller" <davem@davemloft.net> L: netdev@vger.kernel.org W: http://www.linuxfoundation.org/en/Net Q: http://patchwork.ozlabs.org/project/netdev/list/ @@ -9900,7 +9890,7 @@ F: include/linux/platform_data/nxp-nci.h F: Documentation/devicetree/bindings/net/nfc/ NFS, SUNRPC, AND LOCKD CLIENTS -M: Trond Myklebust <trond.myklebust@primarydata.com> +M: Trond Myklebust <trond.myklebust@hammerspace.com> M: Anna Schumaker <anna.schumaker@netapp.com> L: linux-nfs@vger.kernel.org W: http://client.linux-nfs.org @@ -12278,7 +12268,6 @@ S: Odd Fixes F: drivers/media/i2c/saa6588* SAA7134 VIDEO4LINUX DRIVER -M: Mauro Carvalho Chehab <mchehab@s-opensource.com> M: Mauro Carvalho Chehab <mchehab@kernel.org> L: linux-media@vger.kernel.org W: https://linuxtv.org @@ -12517,6 +12506,7 @@ F: drivers/scsi/st_*.h SCTP PROTOCOL M: Vlad Yasevich <vyasevich@gmail.com> M: Neil Horman <nhorman@tuxdriver.com> +M: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com> L: linux-sctp@vger.kernel.org W: http://lksctp.sourceforge.net S: Maintained @@ -12782,7 +12772,6 @@ S: Maintained F: drivers/media/radio/si4713/radio-usb-si4713.c SIANO DVB DRIVER -M: Mauro Carvalho Chehab <mchehab@s-opensource.com> M: Mauro Carvalho Chehab <mchehab@kernel.org> L: linux-media@vger.kernel.org W: https://linuxtv.org @@ -13773,7 +13762,6 @@ S: Maintained F: drivers/media/i2c/tda9840* TEA5761 TUNER DRIVER -M: Mauro Carvalho Chehab <mchehab@s-opensource.com> M: Mauro Carvalho Chehab <mchehab@kernel.org> L: linux-media@vger.kernel.org W: https://linuxtv.org @@ -13782,7 +13770,6 @@ S: Odd fixes F: drivers/media/tuners/tea5761.* TEA5767 TUNER DRIVER -M: Mauro Carvalho Chehab <mchehab@s-opensource.com> M: Mauro Carvalho Chehab <mchehab@kernel.org> L: linux-media@vger.kernel.org W: https://linuxtv.org @@ -13872,7 +13859,6 @@ S: Supported F: drivers/iommu/tegra* TEGRA KBC DRIVER -M: Rakesh Iyer <riyer@nvidia.com> M: Laxman Dewangan <ldewangan@nvidia.com> S: Supported F: drivers/input/keyboard/tegra-kbc.c @@ -14199,7 +14185,6 @@ F: Documentation/networking/tlan.txt F: drivers/net/ethernet/ti/tlan.* TM6000 VIDEO4LINUX DRIVER -M: Mauro Carvalho Chehab <mchehab@s-opensource.com> M: Mauro Carvalho Chehab <mchehab@kernel.org> L: linux-media@vger.kernel.org W: https://linuxtv.org @@ -14682,7 +14667,6 @@ F: drivers/usb/common/usb-otg-fsm.c USB OVER IP DRIVER M: Valentina Manea <valentina.manea.m@gmail.com> -M: Shuah Khan <shuahkh@osg.samsung.com> M: Shuah Khan <shuah@kernel.org> L: linux-usb@vger.kernel.org S: Maintained @@ -15426,7 +15410,6 @@ S: Maintained F: arch/x86/entry/vdso/ XC2028/3028 TUNER DRIVER -M: Mauro Carvalho Chehab <mchehab@s-opensource.com> M: Mauro Carvalho Chehab <mchehab@kernel.org> L: linux-media@vger.kernel.org W: https://linuxtv.org @@ -2,8 +2,8 @@ VERSION = 4 PATCHLEVEL = 17 SUBLEVEL = 0 -EXTRAVERSION = -rc3 -NAME = Fearless Coyote +EXTRAVERSION = -rc5 +NAME = Merciless Moray # *DOCUMENTATION* # To see a list of typical targets execute "make help" diff --git a/arch/Kconfig b/arch/Kconfig index 8e0d665c8d53..75dd23acf133 100644 --- a/arch/Kconfig +++ b/arch/Kconfig @@ -464,6 +464,10 @@ config GCC_PLUGIN_LATENT_ENTROPY config GCC_PLUGIN_STRUCTLEAK bool "Force initialization of variables containing userspace addresses" depends on GCC_PLUGINS + # Currently STRUCTLEAK inserts initialization out of live scope of + # variables from KASAN point of view. This leads to KASAN false + # positive reports. Prohibit this combination for now. + depends on !KASAN_EXTRA help This plugin zero-initializes any structures containing a __user attribute. This can prevent some classes of information diff --git a/arch/arm/boot/dts/imx35.dtsi b/arch/arm/boot/dts/imx35.dtsi index bf343195697e..54111ed218b1 100644 --- a/arch/arm/boot/dts/imx35.dtsi +++ b/arch/arm/boot/dts/imx35.dtsi @@ -303,7 +303,7 @@ }; can1: can@53fe4000 { - compatible = "fsl,imx35-flexcan"; + compatible = "fsl,imx35-flexcan", "fsl,imx25-flexcan"; reg = <0x53fe4000 0x1000>; clocks = <&clks 33>, <&clks 33>; clock-names = "ipg", "per"; @@ -312,7 +312,7 @@ }; can2: can@53fe8000 { - compatible = "fsl,imx35-flexcan"; + compatible = "fsl,imx35-flexcan", "fsl,imx25-flexcan"; reg = <0x53fe8000 0x1000>; clocks = <&clks 34>, <&clks 34>; clock-names = "ipg", "per"; diff --git a/arch/arm/boot/dts/imx53.dtsi b/arch/arm/boot/dts/imx53.dtsi index 7d647d043f52..3d65c0192f69 100644 --- a/arch/arm/boot/dts/imx53.dtsi +++ b/arch/arm/boot/dts/imx53.dtsi @@ -551,7 +551,7 @@ }; can1: can@53fc8000 { - compatible = "fsl,imx53-flexcan"; + compatible = "fsl,imx53-flexcan", "fsl,imx25-flexcan"; reg = <0x53fc8000 0x4000>; interrupts = <82>; clocks = <&clks IMX5_CLK_CAN1_IPG_GATE>, @@ -561,7 +561,7 @@ }; can2: can@53fcc000 { - compatible = "fsl,imx53-flexcan"; + compatible = "fsl,imx53-flexcan", "fsl,imx25-flexcan"; reg = <0x53fcc000 0x4000>; interrupts = <83>; clocks = <&clks IMX5_CLK_CAN2_IPG_GATE>, diff --git a/arch/arm64/include/asm/cputype.h b/arch/arm64/include/asm/cputype.h index 30014a9f8f2b..ea690b3562af 100644 --- a/arch/arm64/include/asm/cputype.h +++ b/arch/arm64/include/asm/cputype.h @@ -75,6 +75,7 @@ #define ARM_CPU_IMP_CAVIUM 0x43 #define ARM_CPU_IMP_BRCM 0x42 #define ARM_CPU_IMP_QCOM 0x51 +#define ARM_CPU_IMP_NVIDIA 0x4E #define ARM_CPU_PART_AEM_V8 0xD0F #define ARM_CPU_PART_FOUNDATION 0xD00 @@ -99,6 +100,9 @@ #define QCOM_CPU_PART_FALKOR 0xC00 #define QCOM_CPU_PART_KRYO 0x200 +#define NVIDIA_CPU_PART_DENVER 0x003 +#define NVIDIA_CPU_PART_CARMEL 0x004 + #define MIDR_CORTEX_A53 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_CORTEX_A53) #define MIDR_CORTEX_A57 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_CORTEX_A57) #define MIDR_CORTEX_A72 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_CORTEX_A72) @@ -114,6 +118,8 @@ #define MIDR_QCOM_FALKOR_V1 MIDR_CPU_MODEL(ARM_CPU_IMP_QCOM, QCOM_CPU_PART_FALKOR_V1) #define MIDR_QCOM_FALKOR MIDR_CPU_MODEL(ARM_CPU_IMP_QCOM, QCOM_CPU_PART_FALKOR) #define MIDR_QCOM_KRYO MIDR_CPU_MODEL(ARM_CPU_IMP_QCOM, QCOM_CPU_PART_KRYO) +#define MIDR_NVIDIA_DENVER MIDR_CPU_MODEL(ARM_CPU_IMP_NVIDIA, NVIDIA_CPU_PART_DENVER) +#define MIDR_NVIDIA_CARMEL MIDR_CPU_MODEL(ARM_CPU_IMP_NVIDIA, NVIDIA_CPU_PART_CARMEL) #ifndef __ASSEMBLY__ diff --git a/arch/arm64/include/asm/kvm_emulate.h b/arch/arm64/include/asm/kvm_emulate.h index 23b33e8ea03a..1dab3a984608 100644 --- a/arch/arm64/include/asm/kvm_emulate.h +++ b/arch/arm64/include/asm/kvm_emulate.h @@ -333,7 +333,7 @@ static inline void kvm_vcpu_set_be(struct kvm_vcpu *vcpu) } else { u64 sctlr = vcpu_read_sys_reg(vcpu, SCTLR_EL1); sctlr |= (1 << 25); - vcpu_write_sys_reg(vcpu, SCTLR_EL1, sctlr); + vcpu_write_sys_reg(vcpu, sctlr, SCTLR_EL1); } } diff --git a/arch/arm64/kernel/cpu_errata.c b/arch/arm64/kernel/cpu_errata.c index a900befadfe8..e4a1182deff7 100644 --- a/arch/arm64/kernel/cpu_errata.c +++ b/arch/arm64/kernel/cpu_errata.c @@ -316,6 +316,7 @@ static const struct midr_range arm64_bp_harden_smccc_cpus[] = { MIDR_ALL_VERSIONS(MIDR_CAVIUM_THUNDERX2), MIDR_ALL_VERSIONS(MIDR_QCOM_FALKOR_V1), MIDR_ALL_VERSIONS(MIDR_QCOM_FALKOR), + MIDR_ALL_VERSIONS(MIDR_NVIDIA_DENVER), {}, }; diff --git a/arch/arm64/kvm/hyp/vgic-v2-cpuif-proxy.c b/arch/arm64/kvm/hyp/vgic-v2-cpuif-proxy.c index 86801b6055d6..39be799d0417 100644 --- a/arch/arm64/kvm/hyp/vgic-v2-cpuif-proxy.c +++ b/arch/arm64/kvm/hyp/vgic-v2-cpuif-proxy.c @@ -18,11 +18,20 @@ #include <linux/compiler.h> #include <linux/irqchip/arm-gic.h> #include <linux/kvm_host.h> +#include <linux/swab.h> #include <asm/kvm_emulate.h> #include <asm/kvm_hyp.h> #include <asm/kvm_mmu.h> +static bool __hyp_text __is_be(struct kvm_vcpu *vcpu) +{ + if (vcpu_mode_is_32bit(vcpu)) + return !!(read_sysreg_el2(spsr) & COMPAT_PSR_E_BIT); + + return !!(read_sysreg(SCTLR_EL1) & SCTLR_ELx_EE); +} + /* * __vgic_v2_perform_cpuif_access -- perform a GICV access on behalf of the * guest. @@ -64,14 +73,19 @@ int __hyp_text __vgic_v2_perform_cpuif_access(struct kvm_vcpu *vcpu) addr += fault_ipa - vgic->vgic_cpu_base; if (kvm_vcpu_dabt_iswrite(vcpu)) { - u32 data = vcpu_data_guest_to_host(vcpu, - vcpu_get_reg(vcpu, rd), - sizeof(u32)); + u32 data = vcpu_get_reg(vcpu, rd); + if (__is_be(vcpu)) { + /* guest pre-swabbed data, undo this for writel() */ + data = swab32(data); + } writel_relaxed(data, addr); } else { u32 data = readl_relaxed(addr); - vcpu_set_reg(vcpu, rd, vcpu_data_host_to_guest(vcpu, data, - sizeof(u32))); + if (__is_be(vcpu)) { + /* guest expects swabbed data */ + data = swab32(data); + } + vcpu_set_reg(vcpu, rd, data); } return 1; diff --git a/arch/arm64/mm/init.c b/arch/arm64/mm/init.c index 9f3c47acf8ff..1b18b4722420 100644 --- a/arch/arm64/mm/init.c +++ b/arch/arm64/mm/init.c @@ -646,8 +646,10 @@ static int keep_initrd __initdata; void __init free_initrd_mem(unsigned long start, unsigned long end) { - if (!keep_initrd) + if (!keep_initrd) { free_reserved_area((void *)start, (void *)end, 0, "initrd"); + memblock_free(__virt_to_phys(start), end - start); + } } static int __init keepinitrd_setup(char *__unused) diff --git a/arch/hexagon/include/asm/io.h b/arch/hexagon/include/asm/io.h index 9e8621d94ee9..e17262ad125e 100644 --- a/arch/hexagon/include/asm/io.h +++ b/arch/hexagon/include/asm/io.h @@ -216,6 +216,12 @@ static inline void memcpy_toio(volatile void __iomem *dst, const void *src, memcpy((void *) dst, src, count); } +static inline void memset_io(volatile void __iomem *addr, int value, + size_t size) +{ + memset((void __force *)addr, value, size); +} + #define PCI_IO_ADDR (volatile void __iomem *) /* diff --git a/arch/hexagon/lib/checksum.c b/arch/hexagon/lib/checksum.c index 617506d1a559..7cd0a2259269 100644 --- a/arch/hexagon/lib/checksum.c +++ b/arch/hexagon/lib/checksum.c @@ -199,3 +199,4 @@ csum_partial_copy_nocheck(const void *src, void *dst, int len, __wsum sum) memcpy(dst, src, len); return csum_partial(dst, len, sum); } +EXPORT_SYMBOL(csum_partial_copy_nocheck); diff --git a/arch/parisc/Makefile b/arch/parisc/Makefile index e2364ff59180..34ac503e28ad 100644 --- a/arch/parisc/Makefile +++ b/arch/parisc/Makefile @@ -123,6 +123,9 @@ INSTALL_TARGETS = zinstall install PHONY += bzImage $(BOOT_TARGETS) $(INSTALL_TARGETS) +# Default kernel to build +all: bzImage + zImage: vmlinuz Image: vmlinux diff --git a/arch/parisc/kernel/drivers.c b/arch/parisc/kernel/drivers.c index 3b8507f71050..ee5a78a151a6 100644 --- a/arch/parisc/kernel/drivers.c +++ b/arch/parisc/kernel/drivers.c @@ -448,7 +448,8 @@ static int match_by_id(struct device * dev, void * data) * Checks all the children of @parent for a matching @id. If none * found, it allocates a new device and returns it. */ -static struct parisc_device * alloc_tree_node(struct device *parent, char id) +static struct parisc_device * __init alloc_tree_node( + struct device *parent, char id) { struct match_id_data d = { .id = id, @@ -825,8 +826,8 @@ static void walk_lower_bus(struct parisc_device *dev) * devices which are not physically connected (such as extra serial & * keyboard ports). This problem is not yet solved. */ -static void walk_native_bus(unsigned long io_io_low, unsigned long io_io_high, - struct device *parent) +static void __init walk_native_bus(unsigned long io_io_low, + unsigned long io_io_high, struct device *parent) { int i, devices_found = 0; unsigned long hpa = io_io_low; diff --git a/arch/parisc/kernel/pci.c b/arch/parisc/kernel/pci.c index 13ee3569959a..ae684ac6efb6 100644 --- a/arch/parisc/kernel/pci.c +++ b/arch/parisc/kernel/pci.c @@ -174,7 +174,7 @@ void pcibios_set_master(struct pci_dev *dev) * pcibios_init_bridge() initializes cache line and default latency * for pci controllers and pci-pci bridges */ -void __init pcibios_init_bridge(struct pci_dev *dev) +void __ref pcibios_init_bridge(struct pci_dev *dev) { unsigned short bridge_ctl, bridge_ctl_new; diff --git a/arch/parisc/kernel/time.c b/arch/parisc/kernel/time.c index c3830400ca28..a1e772f909cb 100644 --- a/arch/parisc/kernel/time.c +++ b/arch/parisc/kernel/time.c @@ -205,7 +205,7 @@ static int __init rtc_init(void) device_initcall(rtc_init); #endif -void read_persistent_clock(struct timespec *ts) +void read_persistent_clock64(struct timespec64 *ts) { static struct pdc_tod tod_data; if (pdc_tod_read(&tod_data) == 0) { diff --git a/arch/parisc/kernel/traps.c b/arch/parisc/kernel/traps.c index 68e671a11987..71d31274d782 100644 --- a/arch/parisc/kernel/traps.c +++ b/arch/parisc/kernel/traps.c @@ -837,6 +837,17 @@ void __init initialize_ivt(const void *iva) if (pdc_instr(&instr) == PDC_OK) ivap[0] = instr; + /* + * Rules for the checksum of the HPMC handler: + * 1. The IVA does not point to PDC/PDH space (ie: the OS has installed + * its own IVA). + * 2. The word at IVA + 32 is nonzero. + * 3. If Length (IVA + 60) is not zero, then Length (IVA + 60) and + * Address (IVA + 56) are word-aligned. + * 4. The checksum of the 8 words starting at IVA + 32 plus the sum of + * the Length/4 words starting at Address is zero. + */ + /* Compute Checksum for HPMC handler */ length = os_hpmc_size; ivap[7] = length; diff --git a/arch/parisc/mm/init.c b/arch/parisc/mm/init.c index cab32ee824d2..2607d2d33405 100644 --- a/arch/parisc/mm/init.c +++ b/arch/parisc/mm/init.c @@ -516,7 +516,7 @@ static void __init map_pages(unsigned long start_vaddr, } } -void free_initmem(void) +void __ref free_initmem(void) { unsigned long init_begin = (unsigned long)__init_begin; unsigned long init_end = (unsigned long)__init_end; diff --git a/arch/powerpc/include/asm/ftrace.h b/arch/powerpc/include/asm/ftrace.h index 9abddde372ab..b2dabd06659d 100644 --- a/arch/powerpc/include/asm/ftrace.h +++ b/arch/powerpc/include/asm/ftrace.h @@ -69,17 +69,30 @@ struct dyn_arch_ftrace { #endif #if defined(CONFIG_FTRACE_SYSCALLS) && !defined(__ASSEMBLY__) -#ifdef PPC64_ELF_ABI_v1 +/* + * Some syscall entry functions on powerpc start with "ppc_" (fork and clone, + * for instance) or ppc32_/ppc64_. We should also match the sys_ variant with + * those. + */ #define ARCH_HAS_SYSCALL_MATCH_SYM_NAME +#ifdef PPC64_ELF_ABI_v1 +static inline bool arch_syscall_match_sym_name(const char *sym, const char *name) +{ + /* We need to skip past the initial dot, and the __se_sys alias */ + return !strcmp(sym + 1, name) || + (!strncmp(sym, ".__se_sys", 9) && !strcmp(sym + 6, name)) || + (!strncmp(sym, ".ppc_", 5) && !strcmp(sym + 5, name + 4)) || + (!strncmp(sym, ".ppc32_", 7) && !strcmp(sym + 7, name + 4)) || + (!strncmp(sym, ".ppc64_", 7) && !strcmp(sym + 7, name + 4)); +} +#else static inline bool arch_syscall_match_sym_name(const char *sym, const char *name) { - /* - * Compare the symbol name with the system call name. Skip the .sys or .SyS - * prefix from the symbol name and the sys prefix from the system call name and - * just match the rest. This is only needed on ppc64 since symbol names on - * 32bit do not start with a period so the generic function will work. - */ - return !strcmp(sym + 4, name + 3); + return !strcmp(sym, name) || + (!strncmp(sym, "__se_sys", 8) && !strcmp(sym + 5, name)) || + (!strncmp(sym, "ppc_", 4) && !strcmp(sym + 4, name + 4)) || + (!strncmp(sym, "ppc32_", 6) && !strcmp(sym + 6, name + 4)) || + (!strncmp(sym, "ppc64_", 6) && !strcmp(sym + 6, name + 4)); } #endif #endif /* CONFIG_FTRACE_SYSCALLS && !__ASSEMBLY__ */ diff --git a/arch/powerpc/include/asm/paca.h b/arch/powerpc/include/asm/paca.h index 4185f1c96125..3f109a3e3edb 100644 --- a/arch/powerpc/include/asm/paca.h +++ b/arch/powerpc/include/asm/paca.h @@ -165,7 +165,6 @@ struct paca_struct { u64 saved_msr; /* MSR saved here by enter_rtas */ u16 trap_save; /* Used when bad stack is encountered */ u8 irq_soft_mask; /* mask for irq soft masking */ - u8 soft_enabled; /* irq soft-enable flag */ u8 irq_happened; /* irq happened while soft-disabled */ u8 io_sync; /* writel() needs spin_unlock sync */ u8 irq_work_pending; /* IRQ_WORK interrupt while soft-disable */ diff --git a/arch/powerpc/include/asm/topology.h b/arch/powerpc/include/asm/topology.h index 9f421641a35c..16b077801a5f 100644 --- a/arch/powerpc/include/asm/topology.h +++ b/arch/powerpc/include/asm/topology.h @@ -91,6 +91,7 @@ extern int start_topology_update(void); extern int stop_topology_update(void); extern int prrn_is_enabled(void); extern int find_and_online_cpu_nid(int cpu); +extern int timed_topology_update(int nsecs); #else static inline int start_topology_update(void) { @@ -108,16 +109,12 @@ static inline int find_and_online_cpu_nid(int cpu) { return 0; } +static inline int timed_topology_update(int nsecs) +{ + return 0; +} #endif /* CONFIG_NUMA && CONFIG_PPC_SPLPAR */ -#if defined(CONFIG_HOTPLUG_CPU) || defined(CONFIG_NEED_MULTIPLE_NODES) -#if defined(CONFIG_PPC_SPLPAR) -extern int timed_topology_update(int nsecs); -#else -#define timed_topology_update(nsecs) -#endif /* CONFIG_PPC_SPLPAR */ -#endif /* CONFIG_HOTPLUG_CPU || CONFIG_NEED_MULTIPLE_NODES */ - #include <asm-generic/topology.h> #ifdef CONFIG_SMP diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig index 97fe29316476..1851eaeee131 100644 --- a/arch/sh/Kconfig +++ b/arch/sh/Kconfig @@ -9,6 +9,7 @@ config SUPERH select HAVE_IDE if HAS_IOPORT_MAP select HAVE_MEMBLOCK select HAVE_MEMBLOCK_NODE_MAP + select NO_BOOTMEM select ARCH_DISCARD_MEMBLOCK select HAVE_OPROFILE select HAVE_GENERIC_DMA_COHERENT diff --git a/arch/sh/kernel/cpu/sh2/probe.c b/arch/sh/kernel/cpu/sh2/probe.c index 4205f6d42b69..a5bd03642678 100644 --- a/arch/sh/kernel/cpu/sh2/probe.c +++ b/arch/sh/kernel/cpu/sh2/probe.c @@ -43,7 +43,11 @@ void __ref cpu_probe(void) #endif #if defined(CONFIG_CPU_J2) +#if defined(CONFIG_SMP) unsigned cpu = hard_smp_processor_id(); +#else + unsigned cpu = 0; +#endif if (cpu == 0) of_scan_flat_dt(scan_cache, NULL); if (j2_ccr_base) __raw_writel(0x80000303, j2_ccr_base + 4*cpu); if (cpu != 0) return; diff --git a/arch/sh/kernel/setup.c b/arch/sh/kernel/setup.c index d34e998b809f..c286cf5da6e7 100644 --- a/arch/sh/kernel/setup.c +++ b/arch/sh/kernel/setup.c @@ -11,7 +11,6 @@ #include <linux/ioport.h> #include <linux/init.h> #include <linux/initrd.h> -#include <linux/bootmem.h> #include <linux/console.h> #include <linux/root_dev.h> #include <linux/utsname.h> diff --git a/arch/sh/mm/consistent.c b/arch/sh/mm/consistent.c index 8ce98691d822..f1b44697ad68 100644 --- a/arch/sh/mm/consistent.c +++ b/arch/sh/mm/consistent.c @@ -59,7 +59,9 @@ void *dma_generic_alloc_coherent(struct device *dev, size_t size, split_page(pfn_to_page(virt_to_phys(ret) >> PAGE_SHIFT), order); - *dma_handle = virt_to_phys(ret) - PFN_PHYS(dev->dma_pfn_offset); + *dma_handle = virt_to_phys(ret); + if (!WARN_ON(!dev)) + *dma_handle -= PFN_PHYS(dev->dma_pfn_offset); return ret_nocache; } @@ -69,9 +71,12 @@ void dma_generic_free_coherent(struct device *dev, size_t size, unsigned long attrs) { int order = get_order(size); - unsigned long pfn = (dma_handle >> PAGE_SHIFT) + dev->dma_pfn_offset; + unsigned long pfn = dma_handle >> PAGE_SHIFT; int k; + if (!WARN_ON(!dev)) + pfn += dev->dma_pfn_offset; + for (k = 0; k < (1 << order); k++) __free_pages(pfn_to_page(pfn + k), 0); @@ -143,7 +148,7 @@ int __init platform_resource_setup_memory(struct platform_device *pdev, if (!memsize) return 0; - buf = dma_alloc_coherent(NULL, memsize, &dma_handle, GFP_KERNEL); + buf = dma_alloc_coherent(&pdev->dev, memsize, &dma_handle, GFP_KERNEL); if (!buf) { pr_warning("%s: unable to allocate memory\n", name); return -ENOMEM; diff --git a/arch/sh/mm/init.c b/arch/sh/mm/init.c index ce0bbaa7e404..4034035fbede 100644 --- a/arch/sh/mm/init.c +++ b/arch/sh/mm/init.c @@ -211,59 +211,15 @@ void __init allocate_pgdat(unsigned int nid) NODE_DATA(nid) = __va(phys); memset(NODE_DATA(nid), 0, sizeof(struct pglist_data)); - - NODE_DATA(nid)->bdata = &bootmem_node_data[nid]; #endif NODE_DATA(nid)->node_start_pfn = start_pfn; NODE_DATA(nid)->node_spanned_pages = end_pfn - start_pfn; } -static void __init bootmem_init_one_node(unsigned int nid) -{ - unsigned long total_pages, paddr; - unsigned long end_pfn; - struct pglist_data *p; - - p = NODE_DATA(nid); - - /* Nothing to do.. */ - if (!p->node_spanned_pages) - return; - - end_pfn = pgdat_end_pfn(p); - - total_pages = bootmem_bootmap_pages(p->node_spanned_pages); - - paddr = memblock_alloc(total_pages << PAGE_SHIFT, PAGE_SIZE); - if (!paddr) - panic("Can't allocate bootmap for nid[%d]\n", nid); - - init_bootmem_node(p, paddr >> PAGE_SHIFT, p->node_start_pfn, end_pfn); - - free_bootmem_with_active_regions(nid, end_pfn); - - /* - * XXX Handle initial reservations for the system memory node - * only for the moment, we'll refactor this later for handling - * reservations in other nodes. - */ - if (nid == 0) { - struct memblock_region *reg; - - /* Reserve the sections we're already using. */ - for_each_memblock(reserved, reg) { - reserve_bootmem(reg->base, reg->size, BOOTMEM_DEFAULT); - } - } - - sparse_memory_present_with_active_regions(nid); -} - static void __init do_init_bootmem(void) { struct memblock_region *reg; - int i; /* Add active regions with valid PFNs. */ for_each_memblock(memory, reg) { @@ -279,9 +235,12 @@ static void __init do_init_bootmem(void) plat_mem_setup(); - for_each_online_node(i) - bootmem_init_one_node(i); + for_each_memblock(memory, reg) { + int nid = memblock_get_region_node(reg); + memory_present(nid, memblock_region_memory_base_pfn(reg), + memblock_region_memory_end_pfn(reg)); + } sparse_init(); } @@ -322,7 +281,6 @@ void __init paging_init(void) { unsigned long max_zone_pfns[MAX_NR_ZONES]; unsigned long vaddr, end; - int nid; sh_mv.mv_mem_init(); @@ -377,21 +335,7 @@ void __init paging_init(void) kmap_coherent_init(); memset(max_zone_pfns, 0, sizeof(max_zone_pfns)); - - for_each_online_node(nid) { - pg_data_t *pgdat = NODE_DATA(nid); - unsigned long low, start_pfn; - - start_pfn = pgdat->bdata->node_min_pfn; - low = pgdat->bdata->node_low_pfn; - - if (max_zone_pfns[ZONE_NORMAL] < low) - max_zone_pfns[ZONE_NORMAL] = low; - - printk("Node %u: start_pfn = 0x%lx, low = 0x%lx\n", - nid, start_pfn, low); - } - + max_zone_pfns[ZONE_NORMAL] = max_low_pfn; free_area_init_nodes(max_zone_pfns); } diff --git a/arch/sh/mm/numa.c b/arch/sh/mm/numa.c index 05713d190247..830e8b3684e4 100644 --- a/arch/sh/mm/numa.c +++ b/arch/sh/mm/numa.c @@ -8,7 +8,6 @@ * for more details. */ #include <linux/module.h> -#include <linux/bootmem.h> #include <linux/memblock.h> #include <linux/mm.h> #include <linux/numa.h> @@ -26,9 +25,7 @@ EXPORT_SYMBOL_GPL(node_data); */ void __init setup_bootmem_node(int nid, unsigned long start, unsigned long end) { - unsigned long bootmap_pages; unsigned long start_pfn, end_pfn; - unsigned long bootmem_paddr; /* Don't allow bogus node assignment */ BUG_ON(nid >= MAX_NUMNODES || nid <= 0); @@ -48,25 +45,9 @@ void __init setup_bootmem_node(int nid, unsigned long start, unsigned long end) SMP_CACHE_BYTES, end)); memset(NODE_DATA(nid), 0, sizeof(struct pglist_data)); - NODE_DATA(nid)->bdata = &bootmem_node_data[nid]; NODE_DATA(nid)->node_start_pfn = start_pfn; NODE_DATA(nid)->node_spanned_pages = end_pfn - start_pfn; - /* Node-local bootmap */ - bootmap_pages = bootmem_bootmap_pages(end_pfn - start_pfn); - bootmem_paddr = memblock_alloc_base(bootmap_pages << PAGE_SHIFT, - PAGE_SIZE, end); - init_bootmem_node(NODE_DATA(nid), bootmem_paddr >> PAGE_SHIFT, - start_pfn, end_pfn); - - free_bootmem_with_active_regions(nid, end_pfn); - - /* Reserve the pgdat and bootmap space with the bootmem allocator */ - reserve_bootmem_node(NODE_DATA(nid), start_pfn << PAGE_SHIFT, - sizeof(struct pglist_data), BOOTMEM_DEFAULT); - reserve_bootmem_node(NODE_DATA(nid), bootmem_paddr, - bootmap_pages << PAGE_SHIFT, BOOTMEM_DEFAULT); - /* It's up */ node_set_online(nid); diff --git a/arch/sparc/include/uapi/asm/oradax.h b/arch/sparc/include/uapi/asm/oradax.h index 722951908b0a..4f6676fe4bcc 100644 --- a/arch/sparc/include/uapi/asm/oradax.h +++ b/arch/sparc/include/uapi/asm/oradax.h @@ -3,7 +3,7 @@ * * This program is free software: you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by - * the Free Software Foundation, either version 3 of the License, or + * the Free Software Foundation, either version 2 of the License, or * (at your option) any later version. * * This program is distributed in the hope that it will be useful, diff --git a/arch/sparc/kernel/vio.c b/arch/sparc/kernel/vio.c index 1a0fa10cb6b7..32bae68e34c1 100644 --- a/arch/sparc/kernel/vio.c +++ b/arch/sparc/kernel/vio.c @@ -403,7 +403,7 @@ static struct vio_dev *vio_create_one(struct mdesc_handle *hp, u64 mp, if (err) { printk(KERN_ERR "VIO: Could not register device %s, err=%d\n", dev_name(&vdev->dev), err); - kfree(vdev); + put_device(&vdev->dev); return NULL; } if (vdev->dp) diff --git a/arch/x86/entry/vdso/vdso32/vdso-fakesections.c b/arch/x86/entry/vdso/vdso32/vdso-fakesections.c deleted file mode 100644 index 541468e25265..000000000000 --- a/arch/x86/entry/vdso/vdso32/vdso-fakesections.c +++ /dev/null @@ -1 +0,0 @@ -#include "../vdso-fakesections.c" diff --git a/arch/x86/events/core.c b/arch/x86/events/core.c index a6006e7bb729..45b2b1c93d04 100644 --- a/arch/x86/events/core.c +++ b/arch/x86/events/core.c @@ -27,6 +27,7 @@ #include <linux/cpu.h> #include <linux/bitops.h> #include <linux/device.h> +#include <linux/nospec.h> #include <asm/apic.h> #include <asm/stacktrace.h> @@ -304,17 +305,20 @@ set_ext_hw_attr(struct hw_perf_event *hwc, struct perf_event *event) config = attr->config; - cache_type = (config >> 0) & 0xff; + cache_type = (config >> 0) & 0xff; if (cache_type >= PERF_COUNT_HW_CACHE_MAX) return -EINVAL; + cache_type = array_index_nospec(cache_type, PERF_COUNT_HW_CACHE_MAX); cache_op = (config >> 8) & 0xff; if (cache_op >= PERF_COUNT_HW_CACHE_OP_MAX) return -EINVAL; + cache_op = array_index_nospec(cache_op, PERF_COUNT_HW_CACHE_OP_MAX); cache_result = (config >> 16) & 0xff; if (cache_result >= PERF_COUNT_HW_CACHE_RESULT_MAX) return -EINVAL; + cache_result = array_index_nospec(cache_result, PERF_COUNT_HW_CACHE_RESULT_MAX); val = hw_cache_event_ids[cache_type][cache_op][cache_result]; @@ -421,6 +425,8 @@ int x86_setup_perfctr(struct perf_event *event) if (attr->config >= x86_pmu.max_events) return -EINVAL; + attr->config = array_index_nospec((unsigned long)attr->config, x86_pmu.max_events); + /* * The generic map: */ diff --git a/arch/x86/events/intel/cstate.c b/arch/x86/events/intel/cstate.c index 9aca448bb8e6..9f8084f18d58 100644 --- a/arch/x86/events/intel/cstate.c +++ b/arch/x86/events/intel/cstate.c @@ -92,6 +92,7 @@ #include <linux/module.h> #include <linux/slab.h> #include <linux/perf_event.h> +#include <linux/nospec.h> #include <asm/cpu_device_id.h> #include <asm/intel-family.h> #include "../perf_event.h" @@ -302,6 +303,7 @@ static int cstate_pmu_event_init(struct perf_event *event) } else if (event->pmu == &cstate_pkg_pmu) { if (cfg >= PERF_CSTATE_PKG_EVENT_MAX) return -EINVAL; + cfg = array_index_nospec((unsigned long)cfg, PERF_CSTATE_PKG_EVENT_MAX); if (!pkg_msr[cfg].attr) return -EINVAL; event->hw.event_base = pkg_msr[cfg].msr; diff --git a/arch/x86/events/msr.c b/arch/x86/events/msr.c index e7edf19e64c2..b4771a6ddbc1 100644 --- a/arch/x86/events/msr.c +++ b/arch/x86/events/msr.c @@ -1,5 +1,6 @@ // SPDX-License-Identifier: GPL-2.0 #include <linux/perf_event.h> +#include <linux/nospec.h> #include <asm/intel-family.h> enum perf_msr_id { @@ -158,9 +159,6 @@ static int msr_event_init(struct perf_event *event) if (event->attr.type != event->pmu->type) return -ENOENT; - if (cfg >= PERF_MSR_EVENT_MAX) - return -EINVAL; - /* unsupported modes and filters */ if (event->attr.exclude_user || event->attr.exclude_kernel || @@ -171,6 +169,11 @@ static int msr_event_init(struct perf_event *event) event->attr.sample_period) /* no sampling */ return -EINVAL; + if (cfg >= PERF_MSR_EVENT_MAX) + return -EINVAL; + + cfg = array_index_nospec((unsigned long)cfg, PERF_MSR_EVENT_MAX); + if (!msr[cfg].attr) return -EINVAL; diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index 8a5b185735e1..ce243f7d2d4e 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -848,6 +848,11 @@ void get_cpu_cap(struct cpuinfo_x86 *c) c->x86_power = edx; } + if (c->extended_cpuid_level >= 0x80000008) { + cpuid(0x80000008, &eax, &ebx, &ecx, &edx); + c->x86_capability[CPUID_8000_0008_EBX] = ebx; + } + if (c->extended_cpuid_level >= 0x8000000a) c->x86_capability[CPUID_8000_000A_EDX] = cpuid_edx(0x8000000a); @@ -871,7 +876,6 @@ static void get_cpu_address_sizes(struct cpuinfo_x86 *c) c->x86_virt_bits = (eax >> 8) & 0xff; c->x86_phys_bits = eax & 0xff; - c->x86_capability[CPUID_8000_0008_EBX] = ebx; } #ifdef CONFIG_X86_32 else if (cpu_has(c, X86_FEATURE_PAE) || cpu_has(c, X86_FEATURE_PSE36)) diff --git a/arch/x86/kernel/tsc.c b/arch/x86/kernel/tsc.c index 91e6da48cbb6..74392d9d51e0 100644 --- a/arch/x86/kernel/tsc.c +++ b/arch/x86/kernel/tsc.c @@ -1067,6 +1067,7 @@ static struct clocksource clocksource_tsc_early = { .resume = tsc_resume, .mark_unstable = tsc_cs_mark_unstable, .tick_stable = tsc_cs_tick_stable, + .list = LIST_HEAD_INIT(clocksource_tsc_early.list), }; /* @@ -1086,6 +1087,7 @@ static struct clocksource clocksource_tsc = { .resume = tsc_resume, .mark_unstable = tsc_cs_mark_unstable, .tick_stable = tsc_cs_tick_stable, + .list = LIST_HEAD_INIT(clocksource_tsc.list), }; void mark_tsc_unstable(char *reason) @@ -1098,13 +1100,9 @@ void mark_tsc_unstable(char *reason) clear_sched_clock_stable(); disable_sched_clock_irqtime(); pr_info("Marking TSC unstable due to %s\n", reason); - /* Change only the rating, when not registered */ - if (clocksource_tsc.mult) { - clocksource_mark_unstable(&clocksource_tsc); - } else { - clocksource_tsc.flags |= CLOCK_SOURCE_UNSTABLE; - clocksource_tsc.rating = 0; - } + + clocksource_mark_unstable(&clocksource_tsc_early); + clocksource_mark_unstable(&clocksource_tsc); } EXPORT_SYMBOL_GPL(mark_tsc_unstable); @@ -1244,7 +1242,7 @@ static void tsc_refine_calibration_work(struct work_struct *work) /* Don't bother refining TSC on unstable systems */ if (tsc_unstable) - return; + goto unreg; /* * Since the work is started early in boot, we may be @@ -1297,11 +1295,12 @@ static void tsc_refine_calibration_work(struct work_struct *work) out: if (tsc_unstable) - return; + goto unreg; if (boot_cpu_has(X86_FEATURE_ART)) art_related_clocksource = &clocksource_tsc; clocksource_register_khz(&clocksource_tsc, tsc_khz); +unreg: clocksource_unregister(&clocksource_tsc_early); } @@ -1311,8 +1310,8 @@ static int __init init_tsc_clocksource(void) if (!boot_cpu_has(X86_FEATURE_TSC) || tsc_disabled > 0 || !tsc_khz) return 0; - if (check_tsc_unstable()) - return 0; + if (tsc_unstable) + goto unreg; if (tsc_clocksource_reliable) clocksource_tsc.flags &= ~CLOCK_SOURCE_MUST_VERIFY; @@ -1328,6 +1327,7 @@ static int __init init_tsc_clocksource(void) if (boot_cpu_has(X86_FEATURE_ART)) art_related_clocksource = &clocksource_tsc; clocksource_register_khz(&clocksource_tsc, tsc_khz); +unreg: clocksource_unregister(&clocksource_tsc_early); return 0; } diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c index 70dcb5548022..b74c9c1405b9 100644 --- a/arch/x86/kvm/lapic.c +++ b/arch/x86/kvm/lapic.c @@ -1463,23 +1463,6 @@ static void start_sw_tscdeadline(struct kvm_lapic *apic) local_irq_restore(flags); } -static void start_sw_period(struct kvm_lapic *apic) -{ - if (!apic->lapic_timer.period) - return; - - if (apic_lvtt_oneshot(apic) && - ktime_after(ktime_get(), - apic->lapic_timer.target_expiration)) { - apic_timer_expired(apic); - return; - } - - hrtimer_start(&apic->lapic_timer.timer, - apic->lapic_timer.target_expiration, - HRTIMER_MODE_ABS_PINNED); -} - static void update_target_expiration(struct kvm_lapic *apic, uint32_t old_divisor) { ktime_t now, remaining; @@ -1546,6 +1529,26 @@ static void advance_periodic_target_expiration(struct kvm_lapic *apic) apic->lapic_timer.period); } +static void start_sw_period(struct kvm_lapic *apic) +{ + if (!apic->lapic_timer.period) + return; + + if (ktime_after(ktime_get(), + apic->lapic_timer.target_expiration)) { + apic_timer_expired(apic); + + if (apic_lvtt_oneshot(apic)) + return; + + advance_periodic_target_expiration(apic); + } + + hrtimer_start(&apic->lapic_timer.timer, + apic->lapic_timer.target_expiration, + HRTIMER_MODE_ABS_PINNED); +} + bool kvm_lapic_hv_timer_in_use(struct kvm_vcpu *vcpu) { if (!lapic_in_kernel(vcpu)) diff --git a/arch/x86/net/bpf_jit_comp.c b/arch/x86/net/bpf_jit_comp.c index b725154182cc..263c8453815e 100644 --- a/arch/x86/net/bpf_jit_comp.c +++ b/arch/x86/net/bpf_jit_comp.c @@ -1027,7 +1027,17 @@ emit_cond_jmp: /* convert BPF opcode to x86 */ break; case BPF_JMP | BPF_JA: - jmp_offset = addrs[i + insn->off] - addrs[i]; + if (insn->off == -1) + /* -1 jmp instructions will always jump + * backwards two bytes. Explicitly handling + * this case avoids wasting too many passes + * when there are long sequences of replaced + * dead code. + */ + jmp_offset = -2; + else + jmp_offset = addrs[i + insn->off] - addrs[i]; + if (!jmp_offset) /* optimize out nop jumps */ break; @@ -1226,6 +1236,7 @@ skip_init_addrs: for (pass = 0; pass < 20 || image; pass++) { proglen = do_jit(prog, addrs, image, oldproglen, &ctx); if (proglen <= 0) { +out_image: image = NULL; if (header) bpf_jit_binary_free(header); @@ -1236,8 +1247,7 @@ skip_init_addrs: if (proglen != oldproglen) { pr_err("bpf_jit: proglen=%d != oldproglen=%d\n", proglen, oldproglen); - prog = orig_prog; - goto out_addrs; + goto out_image; } break; } @@ -1273,7 +1283,7 @@ skip_init_addrs: prog = orig_prog; } - if (!prog->is_func || extra_pass) { + if (!image || !prog->is_func || extra_pass) { out_addrs: kfree(addrs); kfree(jit_data); diff --git a/arch/x86/xen/enlighten_hvm.c b/arch/x86/xen/enlighten_hvm.c index 826898701045..19c1ff542387 100644 --- a/arch/x86/xen/enlighten_hvm.c +++ b/arch/x86/xen/enlighten_hvm.c @@ -65,6 +65,19 @@ static void __init xen_hvm_init_mem_mapping(void) { early_memunmap(HYPERVISOR_shared_info, PAGE_SIZE); HYPERVISOR_shared_info = __va(PFN_PHYS(shared_info_pfn)); + + /* + * The virtual address of the shared_info page has changed, so + * the vcpu_info pointer for VCPU 0 is now stale. + * + * The prepare_boot_cpu callback will re-initialize it via + * xen_vcpu_setup, but we can't rely on that to be called for + * old Xen versions (xen_have_vector_callback == 0). + * + * It is, in any case, bad to have a stale vcpu_info pointer + * so reset it now. + */ + xen_vcpu_info_reset(0); } static void __init init_hvm_pv_info(void) diff --git a/arch/x86/xen/enlighten_pv.c b/arch/x86/xen/enlighten_pv.c index c36d23aa6c35..357969a3697c 100644 --- a/arch/x86/xen/enlighten_pv.c +++ b/arch/x86/xen/enlighten_pv.c @@ -421,45 +421,33 @@ static void xen_load_gdt(const struct desc_ptr *dtr) { unsigned long va = dtr->address; unsigned int size = dtr->size + 1; - unsigned pages = DIV_ROUND_UP(size, PAGE_SIZE); - unsigned long frames[pages]; - int f; - - /* - * A GDT can be up to 64k in size, which corresponds to 8192 - * 8-byte entries, or 16 4k pages.. - */ + unsigned long pfn, mfn; + int level; + pte_t *ptep; + void *virt; - BUG_ON(size > 65536); + /* @size should be at most GDT_SIZE which is smaller than PAGE_SIZE. */ + BUG_ON(size > PAGE_SIZE); BUG_ON(va & ~PAGE_MASK); - for (f = 0; va < dtr->address + size; va += PAGE_SIZE, f++) { - int level; - pte_t *ptep; - unsigned long pfn, mfn; - void *virt; - - /* - * The GDT is per-cpu and is in the percpu data area. - * That can be virtually mapped, so we need to do a - * page-walk to get the underlying MFN for the - * hypercall. The page can also be in the kernel's - * linear range, so we need to RO that mapping too. - */ - ptep = lookup_address(va, &level); - BUG_ON(ptep == NULL); - - pfn = pte_pfn(*ptep); - mfn = pfn_to_mfn(pfn); - virt = __va(PFN_PHYS(pfn)); + /* + * The GDT is per-cpu and is in the percpu data area. + * That can be virtually mapped, so we need to do a + * page-walk to get the underlying MFN for the + * hypercall. The page can also be in the kernel's + * linear range, so we need to RO that mapping too. + */ + ptep = lookup_address(va, &level); + BUG_ON(ptep == NULL); - frames[f] = mfn; + pfn = pte_pfn(*ptep); + mfn = pfn_to_mfn(pfn); + virt = __va(PFN_PHYS(pfn)); - make_lowmem_page_readonly((void *)va); - make_lowmem_page_readonly(virt); - } + make_lowmem_page_readonly((void *)va); + make_lowmem_page_readonly(virt); - if (HYPERVISOR_set_gdt(frames, size / sizeof(struct desc_struct))) + if (HYPERVISOR_set_gdt(&mfn, size / sizeof(struct desc_struct))) BUG(); } @@ -470,34 +458,22 @@ static void __init xen_load_gdt_boot(const struct desc_ptr *dtr) { unsigned long va = dtr->address; unsigned int size = dtr->size + 1; - unsigned pages = DIV_ROUND_UP(size, PAGE_SIZE); - unsigned long frames[pages]; - int f; - - /* - * A GDT can be up to 64k in size, which corresponds to 8192 - * 8-byte entries, or 16 4k pages.. - */ + unsigned long pfn, mfn; + pte_t pte; - BUG_ON(size > 65536); + /* @size should be at most GDT_SIZE which is smaller than PAGE_SIZE. */ + BUG_ON(size > PAGE_SIZE); BUG_ON(va & ~PAGE_MASK); - for (f = 0; va < dtr->address + size; va += PAGE_SIZE, f++) { - pte_t pte; - unsigned long pfn, mfn; + pfn = virt_to_pfn(va); + mfn = pfn_to_mfn(pfn); - pfn = virt_to_pfn(va); - mfn = pfn_to_mfn(pfn); + pte = pfn_pte(pfn, PAGE_KERNEL_RO); - pte = pfn_pte(pfn, PAGE_KERNEL_RO); - - if (HYPERVISOR_update_va_mapping((unsigned long)va, pte, 0)) - BUG(); - - frames[f] = mfn; - } + if (HYPERVISOR_update_va_mapping((unsigned long)va, pte, 0)) + BUG(); - if (HYPERVISOR_set_gdt(frames, size / sizeof(struct desc_struct))) + if (HYPERVISOR_set_gdt(&mfn, size / sizeof(struct desc_struct))) BUG(); } diff --git a/block/blk-mq.c b/block/blk-mq.c index c3621453ad87..9ce9cac16c3f 100644 --- a/block/blk-mq.c +++ b/block/blk-mq.c @@ -95,18 +95,15 @@ static void blk_mq_check_inflight(struct blk_mq_hw_ctx *hctx, { struct mq_inflight *mi = priv; - if (blk_mq_rq_state(rq) == MQ_RQ_IN_FLIGHT) { - /* - * index[0] counts the specific partition that was asked - * for. index[1] counts the ones that are active on the - * whole device, so increment that if mi->part is indeed - * a partition, and not a whole device. - */ - if (rq->part == mi->part) - mi->inflight[0]++; - if (mi->part->partno) - mi->inflight[1]++; - } + /* + * index[0] counts the specific partition that was asked for. index[1] + * counts the ones that are active on the whole device, so increment + * that if mi->part is indeed a partition, and not a whole device. + */ + if (rq->part == mi->part) + mi->inflight[0]++; + if (mi->part->partno) + mi->inflight[1]++; } void blk_mq_in_flight(struct request_queue *q, struct hd_struct *part, @@ -118,6 +115,25 @@ void blk_mq_in_flight(struct request_queue *q, struct hd_struct *part, blk_mq_queue_tag_busy_iter(q, blk_mq_check_inflight, &mi); } +static void blk_mq_check_inflight_rw(struct blk_mq_hw_ctx *hctx, + struct request *rq, void *priv, + bool reserved) +{ + struct mq_inflight *mi = priv; + + if (rq->part == mi->part) + mi->inflight[rq_data_dir(rq)]++; +} + +void blk_mq_in_flight_rw(struct request_queue *q, struct hd_struct *part, + unsigned int inflight[2]) +{ + struct mq_inflight mi = { .part = part, .inflight = inflight, }; + + inflight[0] = inflight[1] = 0; + blk_mq_queue_tag_busy_iter(q, blk_mq_check_inflight_rw, &mi); +} + void blk_freeze_queue_start(struct request_queue *q) { int freeze_depth; diff --git a/block/blk-mq.h b/block/blk-mq.h index 89b5cd3a6c70..e1bb420dc5d6 100644 --- a/block/blk-mq.h +++ b/block/blk-mq.h @@ -188,7 +188,9 @@ static inline bool blk_mq_hw_queue_mapped(struct blk_mq_hw_ctx *hctx) } void blk_mq_in_flight(struct request_queue *q, struct hd_struct *part, - unsigned int inflight[2]); + unsigned int inflight[2]); +void blk_mq_in_flight_rw(struct request_queue *q, struct hd_struct *part, + unsigned int inflight[2]); static inline void blk_mq_put_dispatch_budget(struct blk_mq_hw_ctx *hctx) { diff --git a/block/genhd.c b/block/genhd.c index dc7e089373b9..c4513fe1adda 100644 --- a/block/genhd.c +++ b/block/genhd.c @@ -82,6 +82,18 @@ void part_in_flight(struct request_queue *q, struct hd_struct *part, } } +void part_in_flight_rw(struct request_queue *q, struct hd_struct *part, + unsigned int inflight[2]) +{ + if (q->mq_ops) { + blk_mq_in_flight_rw(q, part, inflight); + return; + } + + inflight[0] = atomic_read(&part->in_flight[0]); + inflight[1] = atomic_read(&part->in_flight[1]); +} + struct hd_struct *__disk_get_part(struct gendisk *disk, int partno) { struct disk_part_tbl *ptbl = rcu_dereference(disk->part_tbl); diff --git a/block/partition-generic.c b/block/partition-generic.c index 08dabcd8b6ae..db57cced9b98 100644 --- a/block/partition-generic.c +++ b/block/partition-generic.c @@ -145,13 +145,15 @@ ssize_t part_stat_show(struct device *dev, jiffies_to_msecs(part_stat_read(p, time_in_queue))); } -ssize_t part_inflight_show(struct device *dev, - struct device_attribute *attr, char *buf) +ssize_t part_inflight_show(struct device *dev, struct device_attribute *attr, + char *buf) { struct hd_struct *p = dev_to_part(dev); + struct request_queue *q = part_to_disk(p)->queue; + unsigned int inflight[2]; - return sprintf(buf, "%8u %8u\n", atomic_read(&p->in_flight[0]), - atomic_read(&p->in_flight[1])); + part_in_flight_rw(q, p, inflight); + return sprintf(buf, "%8u %8u\n", inflight[0], inflight[1]); } #ifdef CONFIG_FAIL_MAKE_REQUEST diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c index 1ff17799769d..6389c88b3500 100644 --- a/drivers/ata/ahci.c +++ b/drivers/ata/ahci.c @@ -698,7 +698,7 @@ static int ahci_vt8251_hardreset(struct ata_link *link, unsigned int *class, DPRINTK("ENTER\n"); - ahci_stop_engine(ap); + hpriv->stop_engine(ap); rc = sata_link_hardreset(link, sata_ehc_deb_timing(&link->eh_context), deadline, &online, NULL); @@ -724,7 +724,7 @@ static int ahci_p5wdh_hardreset(struct ata_link *link, unsigned int *class, bool online; int rc; - ahci_stop_engine(ap); + hpriv->stop_engine(ap); /* clear D2H reception area to properly wait for D2H FIS */ ata_tf_init(link->device, &tf); @@ -788,7 +788,7 @@ static int ahci_avn_hardreset(struct ata_link *link, unsigned int *class, DPRINTK("ENTER\n"); - ahci_stop_engine(ap); + hpriv->stop_engine(ap); for (i = 0; i < 2; i++) { u16 val; diff --git a/drivers/ata/ahci.h b/drivers/ata/ahci.h index 4356ef1d28a8..824bd399f02e 100644 --- a/drivers/ata/ahci.h +++ b/drivers/ata/ahci.h @@ -350,7 +350,6 @@ struct ahci_host_priv { u32 em_msg_type; /* EM message type */ bool got_runtime_pm; /* Did we do pm_runtime_get? */ struct clk *clks[AHCI_MAX_CLKS]; /* Optional */ - struct reset_control *rsts; /* Optional */ struct regulator **target_pwrs; /* Optional */ /* * If platform uses PHYs. There is a 1:1 relation between the port number and @@ -366,6 +365,13 @@ struct ahci_host_priv { * be overridden anytime before the host is activated. */ void (*start_engine)(struct ata_port *ap); + /* + * Optional ahci_stop_engine override, if not set this gets set to the + * default ahci_stop_engine during ahci_save_initial_config, this can + * be overridden anytime before the host is activated. + */ + int (*stop_engine)(struct ata_port *ap); + irqreturn_t (*irq_handler)(int irq, void *dev_instance); /* only required for per-port MSI(-X) support */ diff --git a/drivers/ata/ahci_mvebu.c b/drivers/ata/ahci_mvebu.c index de7128d81e9c..0045dacd814b 100644 --- a/drivers/ata/ahci_mvebu.c +++ b/drivers/ata/ahci_mvebu.c @@ -62,6 +62,60 @@ static void ahci_mvebu_regret_option(struct ahci_host_priv *hpriv) writel(0x80, hpriv->mmio + AHCI_VENDOR_SPECIFIC_0_DATA); } +/** + * ahci_mvebu_stop_engine + * + * @ap: Target ata port + * + * Errata Ref#226 - SATA Disk HOT swap issue when connected through + * Port Multiplier in FIS-based Switching mode. + * + * To avoid the issue, according to design, the bits[11:8, 0] of + * register PxFBS are cleared when Port Command and Status (0x18) bit[0] + * changes its value from 1 to 0, i.e. falling edge of Port + * Command and Status bit[0] sends PULSE that resets PxFBS + * bits[11:8; 0]. + * + * This function is used to override function of "ahci_stop_engine" + * from libahci.c by adding the mvebu work around(WA) to save PxFBS + * value before the PxCMD ST write of 0, then restore PxFBS value. + * + * Return: 0 on success; Error code otherwise. + */ +int ahci_mvebu_stop_engine(struct ata_port *ap) +{ + void __iomem *port_mmio = ahci_port_base(ap); + u32 tmp, port_fbs; + + tmp = readl(port_mmio + PORT_CMD); + + /* check if the HBA is idle */ + if ((tmp & (PORT_CMD_START | PORT_CMD_LIST_ON)) == 0) + return 0; + + /* save the port PxFBS register for later restore */ + port_fbs = readl(port_mmio + PORT_FBS); + + /* setting HBA to idle */ + tmp &= ~PORT_CMD_START; + writel(tmp, port_mmio + PORT_CMD); + + /* + * bit #15 PxCMD signal doesn't clear PxFBS, + * restore the PxFBS register right after clearing the PxCMD ST, + * no need to wait for the PxCMD bit #15. + */ + writel(port_fbs, port_mmio + PORT_FBS); + + /* wait for engine to stop. This could be as long as 500 msec */ + tmp = ata_wait_register(ap, port_mmio + PORT_CMD, + PORT_CMD_LIST_ON, PORT_CMD_LIST_ON, 1, 500); + if (tmp & PORT_CMD_LIST_ON) + return -EIO; + + return 0; +} + #ifdef CONFIG_PM_SLEEP static int ahci_mvebu_suspend(struct platform_device *pdev, pm_message_t state) { @@ -112,6 +166,8 @@ static int ahci_mvebu_probe(struct platform_device *pdev) if (rc) return rc; + hpriv->stop_engine = ahci_mvebu_stop_engine; + if (of_device_is_compatible(pdev->dev.of_node, "marvell,armada-380-ahci")) { dram = mv_mbus_dram_info(); diff --git a/drivers/ata/ahci_qoriq.c b/drivers/ata/ahci_qoriq.c index 2685f28160f7..cfdef4d44ae9 100644 --- a/drivers/ata/ahci_qoriq.c +++ b/drivers/ata/ahci_qoriq.c @@ -96,7 +96,7 @@ static int ahci_qoriq_hardreset(struct ata_link *link, unsigned int *class, DPRINTK("ENTER\n"); - ahci_stop_engine(ap); + hpriv->stop_engine(ap); /* * There is a errata on ls1021a Rev1.0 and Rev2.0 which is: diff --git a/drivers/ata/ahci_xgene.c b/drivers/ata/ahci_xgene.c index c2b5941d9184..ad58da7c9aff 100644 --- a/drivers/ata/ahci_xgene.c +++ b/drivers/ata/ahci_xgene.c @@ -165,7 +165,7 @@ static int xgene_ahci_restart_engine(struct ata_port *ap) PORT_CMD_ISSUE, 0x0, 1, 100)) return -EBUSY; - ahci_stop_engine(ap); + hpriv->stop_engine(ap); ahci_start_fis_rx(ap); /* @@ -421,7 +421,7 @@ static int xgene_ahci_hardreset(struct ata_link *link, unsigned int *class, portrxfis_saved = readl(port_mmio + PORT_FIS_ADDR); portrxfishi_saved = readl(port_mmio + PORT_FIS_ADDR_HI); - ahci_stop_engine(ap); + hpriv->stop_engine(ap); rc = xgene_ahci_do_hardreset(link, deadline, &online); diff --git a/drivers/ata/libahci.c b/drivers/ata/libahci.c index 7adcf3caabd0..e5d90977caec 100644 --- a/drivers/ata/libahci.c +++ b/drivers/ata/libahci.c @@ -560,6 +560,9 @@ void ahci_save_initial_config(struct device *dev, struct ahci_host_priv *hpriv) if (!hpriv->start_engine) hpriv->start_engine = ahci_start_engine; + if (!hpriv->stop_engine) + hpriv->stop_engine = ahci_stop_engine; + if (!hpriv->irq_handler) hpriv->irq_handler = ahci_single_level_irq_intr; } @@ -897,9 +900,10 @@ static void ahci_start_port(struct ata_port *ap) static int ahci_deinit_port(struct ata_port *ap, const char **emsg) { int rc; + struct ahci_host_priv *hpriv = ap->host->private_data; /* disable DMA */ - rc = ahci_stop_engine(ap); + rc = hpriv->stop_engine(ap); if (rc) { *emsg = "failed to stop engine"; return rc; @@ -1310,7 +1314,7 @@ int ahci_kick_engine(struct ata_port *ap) int busy, rc; /* stop engine */ - rc = ahci_stop_engine(ap); + rc = hpriv->stop_engine(ap); if (rc) goto out_restart; @@ -1549,7 +1553,7 @@ int ahci_do_hardreset(struct ata_link *link, unsigned int *class, DPRINTK("ENTER\n"); - ahci_stop_engine(ap); + hpriv->stop_engine(ap); /* clear D2H reception area to properly wait for D2H FIS */ ata_tf_init(link->device, &tf); @@ -2075,14 +2079,14 @@ void ahci_error_handler(struct ata_port *ap) if (!(ap->pflags & ATA_PFLAG_FROZEN)) { /* restart engine */ - ahci_stop_engine(ap); + hpriv->stop_engine(ap); hpriv->start_engine(ap); } sata_pmp_error_handler(ap); if (!ata_dev_enabled(ap->link.device)) - ahci_stop_engine(ap); + hpriv->stop_engine(ap); } EXPORT_SYMBOL_GPL(ahci_error_handler); @@ -2129,7 +2133,7 @@ static void ahci_set_aggressive_devslp(struct ata_port *ap, bool sleep) return; /* set DITO, MDAT, DETO and enable DevSlp, need to stop engine first */ - rc = ahci_stop_engine(ap); + rc = hpriv->stop_engine(ap); if (rc) return; @@ -2189,7 +2193,7 @@ static void ahci_enable_fbs(struct ata_port *ap) return; } - rc = ahci_stop_engine(ap); + rc = hpriv->stop_engine(ap); if (rc) return; @@ -2222,7 +2226,7 @@ static void ahci_disable_fbs(struct ata_port *ap) return; } - rc = ahci_stop_engine(ap); + rc = hpriv->stop_engine(ap); if (rc) return; diff --git a/drivers/ata/libahci_platform.c b/drivers/ata/libahci_platform.c index 46a762442dc5..30cc8f1a31e1 100644 --- a/drivers/ata/libahci_platform.c +++ b/drivers/ata/libahci_platform.c @@ -25,7 +25,6 @@ #include <linux/phy/phy.h> #include <linux/pm_runtime.h> #include <linux/of_platform.h> -#include <linux/reset.h> #include "ahci.h" static void ahci_host_stop(struct ata_host *host); @@ -196,8 +195,7 @@ EXPORT_SYMBOL_GPL(ahci_platform_disable_regulators); * following order: * 1) Regulator * 2) Clocks (through ahci_platform_enable_clks) - * 3) Resets - * 4) Phys + * 3) Phys * * If resource enabling fails at any point the previous enabled resources * are disabled in reverse order. @@ -217,19 +215,12 @@ int ahci_platform_enable_resources(struct ahci_host_priv *hpriv) if (rc) goto disable_regulator; - rc = reset_control_deassert(hpriv->rsts); - if (rc) - goto disable_clks; - rc = ahci_platform_enable_phys(hpriv); if (rc) - goto disable_resets; + goto disable_clks; return 0; -disable_resets: - reset_control_assert(hpriv->rsts); - disable_clks: ahci_platform_disable_clks(hpriv); @@ -248,15 +239,12 @@ EXPORT_SYMBOL_GPL(ahci_platform_enable_resources); * following order: * 1) Phys * 2) Clocks (through ahci_platform_disable_clks) - * 3) Resets - * 4) Regulator + * 3) Regulator */ void ahci_platform_disable_resources(struct ahci_host_priv *hpriv) { ahci_platform_disable_phys(hpriv); - reset_control_assert(hpriv->rsts); - ahci_platform_disable_clks(hpriv); ahci_platform_disable_regulators(hpriv); @@ -405,12 +393,6 @@ struct ahci_host_priv *ahci_platform_get_resources(struct platform_device *pdev) hpriv->clks[i] = clk; } - hpriv->rsts = devm_reset_control_array_get_optional_shared(dev); - if (IS_ERR(hpriv->rsts)) { - rc = PTR_ERR(hpriv->rsts); - goto err_out; - } - hpriv->nports = child_nodes = of_get_child_count(dev->of_node); /* diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c index 8bc71ca61e7f..68596bd4cf06 100644 --- a/drivers/ata/libata-core.c +++ b/drivers/ata/libata-core.c @@ -4549,6 +4549,12 @@ static const struct ata_blacklist_entry ata_device_blacklist [] = { ATA_HORKAGE_ZERO_AFTER_TRIM | ATA_HORKAGE_NOLPM, }, + /* This specific Samsung model/firmware-rev does not handle LPM well */ + { "SAMSUNG MZMPC128HBFU-000MV", "CXM14M1Q", ATA_HORKAGE_NOLPM, }, + + /* Sandisk devices which are known to not handle LPM well */ + { "SanDisk SD7UB3Q*G1001", NULL, ATA_HORKAGE_NOLPM, }, + /* devices that don't properly handle queued TRIM commands */ { "Micron_M500_*", NULL, ATA_HORKAGE_NO_NCQ_TRIM | ATA_HORKAGE_ZERO_AFTER_TRIM, }, diff --git a/drivers/ata/libata-eh.c b/drivers/ata/libata-eh.c index c016829a38fd..513b260bcff1 100644 --- a/drivers/ata/libata-eh.c +++ b/drivers/ata/libata-eh.c @@ -175,8 +175,8 @@ static void ata_eh_handle_port_resume(struct ata_port *ap) { } #endif /* CONFIG_PM */ -static void __ata_ehi_pushv_desc(struct ata_eh_info *ehi, const char *fmt, - va_list args) +static __printf(2, 0) void __ata_ehi_pushv_desc(struct ata_eh_info *ehi, + const char *fmt, va_list args) { ehi->desc_len += vscnprintf(ehi->desc + ehi->desc_len, ATA_EH_DESC_LEN - ehi->desc_len, diff --git a/drivers/ata/sata_highbank.c b/drivers/ata/sata_highbank.c index aafb8cc03523..e67815b896fc 100644 --- a/drivers/ata/sata_highbank.c +++ b/drivers/ata/sata_highbank.c @@ -410,7 +410,7 @@ static int ahci_highbank_hardreset(struct ata_link *link, unsigned int *class, int rc; int retry = 100; - ahci_stop_engine(ap); + hpriv->stop_engine(ap); /* clear D2H reception area to properly wait for D2H FIS */ ata_tf_init(link->device, &tf); diff --git a/drivers/ata/sata_sil24.c b/drivers/ata/sata_sil24.c index 4b1995e2d044..010ca101d412 100644 --- a/drivers/ata/sata_sil24.c +++ b/drivers/ata/sata_sil24.c @@ -285,13 +285,13 @@ static const struct sil24_cerr_info { [PORT_CERR_INCONSISTENT] = { AC_ERR_HSM, ATA_EH_RESET, "protocol mismatch" }, [PORT_CERR_DIRECTION] = { AC_ERR_HSM, ATA_EH_RESET, - "data directon mismatch" }, + "data direction mismatch" }, [PORT_CERR_UNDERRUN] = { AC_ERR_HSM, ATA_EH_RESET, "ran out of SGEs while writing" }, [PORT_CERR_OVERRUN] = { AC_ERR_HSM, ATA_EH_RESET, "ran out of SGEs while reading" }, [PORT_CERR_PKT_PROT] = { AC_ERR_HSM, ATA_EH_RESET, - "invalid data directon for ATAPI CDB" }, + "invalid data direction for ATAPI CDB" }, [PORT_CERR_SGT_BOUNDARY] = { AC_ERR_SYSTEM, ATA_EH_RESET, "SGT not on qword boundary" }, [PORT_CERR_SGT_TGTABRT] = { AC_ERR_HOST_BUS, ATA_EH_RESET, diff --git a/drivers/atm/firestream.c b/drivers/atm/firestream.c index d97c05690faa..4e46dc9e41ad 100644 --- a/drivers/atm/firestream.c +++ b/drivers/atm/firestream.c @@ -191,7 +191,7 @@ static char *res_strings[] = { "reserved 37", "reserved 38", "reserved 39", - "reseverd 40", + "reserved 40", "reserved 41", "reserved 42", "reserved 43", diff --git a/drivers/atm/zatm.c b/drivers/atm/zatm.c index 1ef67db03c8e..9c9a22958717 100644 --- a/drivers/atm/zatm.c +++ b/drivers/atm/zatm.c @@ -28,6 +28,7 @@ #include <asm/io.h> #include <linux/atomic.h> #include <linux/uaccess.h> +#include <linux/nospec.h> #include "uPD98401.h" #include "uPD98402.h" @@ -1458,6 +1459,8 @@ static int zatm_ioctl(struct atm_dev *dev,unsigned int cmd,void __user *arg) return -EFAULT; if (pool < 0 || pool > ZATM_LAST_POOL) return -EINVAL; + pool = array_index_nospec(pool, + ZATM_LAST_POOL + 1); spin_lock_irqsave(&zatm_dev->lock, flags); info = zatm_dev->pool_info[pool]; if (cmd == ZATM_GETPOOLZ) { diff --git a/drivers/block/rbd.c b/drivers/block/rbd.c index 8e8b04cc569a..33b36fea1d73 100644 --- a/drivers/block/rbd.c +++ b/drivers/block/rbd.c @@ -2366,7 +2366,9 @@ static int rbd_obj_issue_copyup(struct rbd_obj_request *obj_req, u32 bytes) osd_req_op_cls_init(obj_req->osd_req, 0, CEPH_OSD_OP_CALL, "rbd", "copyup"); osd_req_op_cls_request_data_bvecs(obj_req->osd_req, 0, - obj_req->copyup_bvecs, bytes); + obj_req->copyup_bvecs, + obj_req->copyup_bvec_count, + bytes); switch (obj_req->img_request->op_type) { case OBJ_OP_WRITE: diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c index c8c8b0b8d333..b937cc1e2c07 100644 --- a/drivers/bluetooth/btusb.c +++ b/drivers/bluetooth/btusb.c @@ -231,6 +231,7 @@ static const struct usb_device_id blacklist_table[] = { { USB_DEVICE(0x0930, 0x0227), .driver_info = BTUSB_ATH3012 }, { USB_DEVICE(0x0b05, 0x17d0), .driver_info = BTUSB_ATH3012 }, { USB_DEVICE(0x0cf3, 0x0036), .driver_info = BTUSB_ATH3012 }, + { USB_DEVICE(0x0cf3, 0x3004), .driver_info = BTUSB_ATH3012 }, { USB_DEVICE(0x0cf3, 0x3008), .driver_info = BTUSB_ATH3012 }, { USB_DEVICE(0x0cf3, 0x311d), .driver_info = BTUSB_ATH3012 }, { USB_DEVICE(0x0cf3, 0x311e), .driver_info = BTUSB_ATH3012 }, @@ -263,7 +264,6 @@ static const struct usb_device_id blacklist_table[] = { { USB_DEVICE(0x0489, 0xe03c), .driver_info = BTUSB_ATH3012 }, /* QCA ROME chipset */ - { USB_DEVICE(0x0cf3, 0x3004), .driver_info = BTUSB_QCA_ROME }, { USB_DEVICE(0x0cf3, 0xe007), .driver_info = BTUSB_QCA_ROME }, { USB_DEVICE(0x0cf3, 0xe009), .driver_info = BTUSB_QCA_ROME }, { USB_DEVICE(0x0cf3, 0xe010), .driver_info = BTUSB_QCA_ROME }, @@ -399,6 +399,13 @@ static const struct dmi_system_id btusb_needs_reset_resume_table[] = { DMI_MATCH(DMI_PRODUCT_NAME, "OptiPlex 3060"), }, }, + { + /* Dell XPS 9360 (QCA ROME device 0cf3:e300) */ + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."), + DMI_MATCH(DMI_PRODUCT_NAME, "XPS 13 9360"), + }, + }, {} }; @@ -2852,6 +2859,12 @@ static int btusb_config_oob_wake(struct hci_dev *hdev) } #endif +static void btusb_check_needs_reset_resume(struct usb_interface *intf) +{ + if (dmi_check_system(btusb_needs_reset_resume_table)) + interface_to_usbdev(intf)->quirks |= USB_QUIRK_RESET_RESUME; +} + static int btusb_probe(struct usb_interface *intf, const struct usb_device_id *id) { @@ -2974,9 +2987,6 @@ static int btusb_probe(struct usb_interface *intf, hdev->send = btusb_send_frame; hdev->notify = btusb_notify; - if (dmi_check_system(btusb_needs_reset_resume_table)) - interface_to_usbdev(intf)->quirks |= USB_QUIRK_RESET_RESUME; - #ifdef CONFIG_PM err = btusb_config_oob_wake(hdev); if (err) @@ -3064,6 +3074,7 @@ static int btusb_probe(struct usb_interface *intf, data->setup_on_usb = btusb_setup_qca; hdev->set_bdaddr = btusb_set_bdaddr_ath3012; set_bit(HCI_QUIRK_SIMULTANEOUS_DISCOVERY, &hdev->quirks); + btusb_check_needs_reset_resume(intf); } #ifdef CONFIG_BT_HCIBTUSB_RTL diff --git a/drivers/char/agp/uninorth-agp.c b/drivers/char/agp/uninorth-agp.c index c381c8e396fc..79d8c84693a1 100644 --- a/drivers/char/agp/uninorth-agp.c +++ b/drivers/char/agp/uninorth-agp.c @@ -195,7 +195,7 @@ static int uninorth_insert_memory(struct agp_memory *mem, off_t pg_start, int ty return 0; } -int uninorth_remove_memory(struct agp_memory *mem, off_t pg_start, int type) +static int uninorth_remove_memory(struct agp_memory *mem, off_t pg_start, int type) { size_t i; u32 *gp; @@ -470,7 +470,7 @@ static int uninorth_free_gatt_table(struct agp_bridge_data *bridge) return 0; } -void null_cache_flush(void) +static void null_cache_flush(void) { mb(); } diff --git a/drivers/clk/clk-cs2000-cp.c b/drivers/clk/clk-cs2000-cp.c index c58019750b7e..a2f8c42e527a 100644 --- a/drivers/clk/clk-cs2000-cp.c +++ b/drivers/clk/clk-cs2000-cp.c @@ -541,7 +541,7 @@ probe_err: return ret; } -static int cs2000_resume(struct device *dev) +static int __maybe_unused cs2000_resume(struct device *dev) { struct cs2000_priv *priv = dev_get_drvdata(dev); diff --git a/drivers/clk/clk-mux.c b/drivers/clk/clk-mux.c index ac4a042f8658..1628b93655ed 100644 --- a/drivers/clk/clk-mux.c +++ b/drivers/clk/clk-mux.c @@ -112,10 +112,18 @@ static int clk_mux_set_parent(struct clk_hw *hw, u8 index) return 0; } +static int clk_mux_determine_rate(struct clk_hw *hw, + struct clk_rate_request *req) +{ + struct clk_mux *mux = to_clk_mux(hw); + + return clk_mux_determine_rate_flags(hw, req, mux->flags); +} + const struct clk_ops clk_mux_ops = { .get_parent = clk_mux_get_parent, .set_parent = clk_mux_set_parent, - .determine_rate = __clk_mux_determine_rate, + .determine_rate = clk_mux_determine_rate, }; EXPORT_SYMBOL_GPL(clk_mux_ops); diff --git a/drivers/clk/clk-stm32mp1.c b/drivers/clk/clk-stm32mp1.c index f1d5967b4b39..edd3cf451401 100644 --- a/drivers/clk/clk-stm32mp1.c +++ b/drivers/clk/clk-stm32mp1.c @@ -216,7 +216,7 @@ static const char * const usart1_src[] = { "pclk5", "pll3_q", "ck_hsi", "ck_csi", "pll4_q", "ck_hse" }; -const char * const usart234578_src[] = { +static const char * const usart234578_src[] = { "pclk1", "pll4_q", "ck_hsi", "ck_csi", "ck_hse" }; @@ -224,10 +224,6 @@ static const char * const usart6_src[] = { "pclk2", "pll4_q", "ck_hsi", "ck_csi", "ck_hse" }; -static const char * const dfsdm_src[] = { - "pclk2", "ck_mcu" -}; - static const char * const fdcan_src[] = { "ck_hse", "pll3_q", "pll4_q" }; @@ -316,10 +312,8 @@ struct stm32_clk_mgate { struct clock_config { u32 id; const char *name; - union { - const char *parent_name; - const char * const *parent_names; - }; + const char *parent_name; + const char * const *parent_names; int num_parents; unsigned long flags; void *cfg; @@ -469,7 +463,7 @@ static void mp1_gate_clk_disable(struct clk_hw *hw) } } -const struct clk_ops mp1_gate_clk_ops = { +static const struct clk_ops mp1_gate_clk_ops = { .enable = mp1_gate_clk_enable, .disable = mp1_gate_clk_disable, .is_enabled = clk_gate_is_enabled, @@ -698,7 +692,7 @@ static void mp1_mgate_clk_disable(struct clk_hw *hw) mp1_gate_clk_disable(hw); } -const struct clk_ops mp1_mgate_clk_ops = { +static const struct clk_ops mp1_mgate_clk_ops = { .enable = mp1_mgate_clk_enable, .disable = mp1_mgate_clk_disable, .is_enabled = clk_gate_is_enabled, @@ -732,7 +726,7 @@ static int clk_mmux_set_parent(struct clk_hw *hw, u8 index) return 0; } -const struct clk_ops clk_mmux_ops = { +static const struct clk_ops clk_mmux_ops = { .get_parent = clk_mmux_get_parent, .set_parent = clk_mmux_set_parent, .determine_rate = __clk_mux_determine_rate, @@ -1048,10 +1042,10 @@ struct stm32_pll_cfg { u32 offset; }; -struct clk_hw *_clk_register_pll(struct device *dev, - struct clk_hw_onecell_data *clk_data, - void __iomem *base, spinlock_t *lock, - const struct clock_config *cfg) +static struct clk_hw *_clk_register_pll(struct device *dev, + struct clk_hw_onecell_data *clk_data, + void __iomem *base, spinlock_t *lock, + const struct clock_config *cfg) { struct stm32_pll_cfg *stm_pll_cfg = cfg->cfg; @@ -1405,7 +1399,8 @@ enum { G_USBH, G_ETHSTP, G_RTCAPB, - G_TZC, + G_TZC1, + G_TZC2, G_TZPC, G_IWDG1, G_BSEC, @@ -1417,7 +1412,7 @@ enum { G_LAST }; -struct stm32_mgate mp1_mgate[G_LAST]; +static struct stm32_mgate mp1_mgate[G_LAST]; #define _K_GATE(_id, _gate_offset, _gate_bit_idx, _gate_flags,\ _mgate, _ops)\ @@ -1440,7 +1435,7 @@ struct stm32_mgate mp1_mgate[G_LAST]; &mp1_mgate[_id], &mp1_mgate_clk_ops) /* Peripheral gates */ -struct stm32_gate_cfg per_gate_cfg[G_LAST] = { +static struct stm32_gate_cfg per_gate_cfg[G_LAST] = { /* Multi gates */ K_GATE(G_MDIO, RCC_APB1ENSETR, 31, 0), K_MGATE(G_DAC12, RCC_APB1ENSETR, 29, 0), @@ -1506,7 +1501,8 @@ struct stm32_gate_cfg per_gate_cfg[G_LAST] = { K_GATE(G_BSEC, RCC_APB5ENSETR, 16, 0), K_GATE(G_IWDG1, RCC_APB5ENSETR, 15, 0), K_GATE(G_TZPC, RCC_APB5ENSETR, 13, 0), - K_GATE(G_TZC, RCC_APB5ENSETR, 12, 0), + K_GATE(G_TZC2, RCC_APB5ENSETR, 12, 0), + K_GATE(G_TZC1, RCC_APB5ENSETR, 11, 0), K_GATE(G_RTCAPB, RCC_APB5ENSETR, 8, 0), K_MGATE(G_USART1, RCC_APB5ENSETR, 4, 0), K_MGATE(G_I2C6, RCC_APB5ENSETR, 3, 0), @@ -1600,7 +1596,7 @@ enum { M_LAST }; -struct stm32_mmux ker_mux[M_LAST]; +static struct stm32_mmux ker_mux[M_LAST]; #define _K_MUX(_id, _offset, _shift, _width, _mux_flags, _mmux, _ops)\ [_id] = {\ @@ -1623,7 +1619,7 @@ struct stm32_mmux ker_mux[M_LAST]; _K_MUX(_id, _offset, _shift, _width, _mux_flags,\ &ker_mux[_id], &clk_mmux_ops) -const struct stm32_mux_cfg ker_mux_cfg[M_LAST] = { +static const struct stm32_mux_cfg ker_mux_cfg[M_LAST] = { /* Kernel multi mux */ K_MMUX(M_SDMMC12, RCC_SDMMC12CKSELR, 0, 3, 0), K_MMUX(M_SPI23, RCC_SPI2S23CKSELR, 0, 3, 0), @@ -1860,7 +1856,8 @@ static const struct clock_config stm32mp1_clock_cfg[] = { PCLK(USART1, "usart1", "pclk5", 0, G_USART1), PCLK(RTCAPB, "rtcapb", "pclk5", CLK_IGNORE_UNUSED | CLK_IS_CRITICAL, G_RTCAPB), - PCLK(TZC, "tzc", "pclk5", CLK_IGNORE_UNUSED, G_TZC), + PCLK(TZC1, "tzc1", "ck_axi", CLK_IGNORE_UNUSED, G_TZC1), + PCLK(TZC2, "tzc2", "ck_axi", CLK_IGNORE_UNUSED, G_TZC2), PCLK(TZPC, "tzpc", "pclk5", CLK_IGNORE_UNUSED, G_TZPC), PCLK(IWDG1, "iwdg1", "pclk5", 0, G_IWDG1), PCLK(BSEC, "bsec", "pclk5", CLK_IGNORE_UNUSED, G_BSEC), @@ -1916,8 +1913,7 @@ static const struct clock_config stm32mp1_clock_cfg[] = { KCLK(RNG1_K, "rng1_k", rng_src, 0, G_RNG1, M_RNG1), KCLK(RNG2_K, "rng2_k", rng_src, 0, G_RNG2, M_RNG2), KCLK(USBPHY_K, "usbphy_k", usbphy_src, 0, G_USBPHY, M_USBPHY), - KCLK(STGEN_K, "stgen_k", stgen_src, CLK_IGNORE_UNUSED, - G_STGEN, M_STGEN), + KCLK(STGEN_K, "stgen_k", stgen_src, CLK_IS_CRITICAL, G_STGEN, M_STGEN), KCLK(SPDIF_K, "spdif_k", spdif_src, 0, G_SPDIF, M_SPDIF), KCLK(SPI1_K, "spi1_k", spi123_src, 0, G_SPI1, M_SPI1), KCLK(SPI2_K, "spi2_k", spi123_src, 0, G_SPI2, M_SPI23), @@ -1948,8 +1944,8 @@ static const struct clock_config stm32mp1_clock_cfg[] = { KCLK(FDCAN_K, "fdcan_k", fdcan_src, 0, G_FDCAN, M_FDCAN), KCLK(SAI1_K, "sai1_k", sai_src, 0, G_SAI1, M_SAI1), KCLK(SAI2_K, "sai2_k", sai2_src, 0, G_SAI2, M_SAI2), - KCLK(SAI3_K, "sai3_k", sai_src, 0, G_SAI2, M_SAI3), - KCLK(SAI4_K, "sai4_k", sai_src, 0, G_SAI2, M_SAI4), + KCLK(SAI3_K, "sai3_k", sai_src, 0, G_SAI3, M_SAI3), + KCLK(SAI4_K, "sai4_k", sai_src, 0, G_SAI4, M_SAI4), KCLK(ADC12_K, "adc12_k", adc12_src, 0, G_ADC12, M_ADC12), KCLK(DSI_K, "dsi_k", dsi_src, 0, G_DSI, M_DSI), KCLK(ADFSDM_K, "adfsdm_k", sai_src, 0, G_ADFSDM, M_SAI1), @@ -1992,10 +1988,6 @@ static const struct clock_config stm32mp1_clock_cfg[] = { _DIV(RCC_MCO2CFGR, 4, 4, 0, NULL)), /* Debug clocks */ - FIXED_FACTOR(NO_ID, "ck_axi_div2", "ck_axi", 0, 1, 2), - - GATE(DBG, "ck_apb_dbg", "ck_axi_div2", 0, RCC_DBGCFGR, 8, 0), - GATE(CK_DBG, "ck_sys_dbg", "ck_axi", 0, RCC_DBGCFGR, 8, 0), COMPOSITE(CK_TRACE, "ck_trace", ck_trace_src, CLK_OPS_PARENT_ENABLE, diff --git a/drivers/clk/clk.c b/drivers/clk/clk.c index ea67ac81c6f9..7af555f0e60c 100644 --- a/drivers/clk/clk.c +++ b/drivers/clk/clk.c @@ -426,9 +426,9 @@ static bool mux_is_better_rate(unsigned long rate, unsigned long now, return now <= rate && now > best; } -static int -clk_mux_determine_rate_flags(struct clk_hw *hw, struct clk_rate_request *req, - unsigned long flags) +int clk_mux_determine_rate_flags(struct clk_hw *hw, + struct clk_rate_request *req, + unsigned long flags) { struct clk_core *core = hw->core, *parent, *best_parent = NULL; int i, num_parents, ret; @@ -488,6 +488,7 @@ out: return 0; } +EXPORT_SYMBOL_GPL(clk_mux_determine_rate_flags); struct clk *__clk_lookup(const char *name) { diff --git a/drivers/clk/meson/clk-regmap.c b/drivers/clk/meson/clk-regmap.c index 3645fdb62343..ab7a3556f5b2 100644 --- a/drivers/clk/meson/clk-regmap.c +++ b/drivers/clk/meson/clk-regmap.c @@ -153,10 +153,19 @@ static int clk_regmap_mux_set_parent(struct clk_hw *hw, u8 index) val << mux->shift); } +static int clk_regmap_mux_determine_rate(struct clk_hw *hw, + struct clk_rate_request *req) +{ + struct clk_regmap *clk = to_clk_regmap(hw); + struct clk_regmap_mux_data *mux = clk_get_regmap_mux_data(clk); + + return clk_mux_determine_rate_flags(hw, req, mux->flags); +} + const struct clk_ops clk_regmap_mux_ops = { .get_parent = clk_regmap_mux_get_parent, .set_parent = clk_regmap_mux_set_parent, - .determine_rate = __clk_mux_determine_rate, + .determine_rate = clk_regmap_mux_determine_rate, }; EXPORT_SYMBOL_GPL(clk_regmap_mux_ops); diff --git a/drivers/clk/meson/gxbb-aoclk.h b/drivers/clk/meson/gxbb-aoclk.h index 0be78383f257..badc4c22b4ee 100644 --- a/drivers/clk/meson/gxbb-aoclk.h +++ b/drivers/clk/meson/gxbb-aoclk.h @@ -17,8 +17,6 @@ #define AO_RTC_ALT_CLK_CNTL0 0x94 #define AO_RTC_ALT_CLK_CNTL1 0x98 -extern const struct clk_ops meson_aoclk_gate_regmap_ops; - struct aoclk_cec_32k { struct clk_hw hw; struct regmap *regmap; diff --git a/drivers/clk/meson/meson8b.c b/drivers/clk/meson/meson8b.c index cc2992493e0b..d0524ec71aad 100644 --- a/drivers/clk/meson/meson8b.c +++ b/drivers/clk/meson/meson8b.c @@ -253,7 +253,7 @@ static struct clk_fixed_factor meson8b_fclk_div3_div = { .mult = 1, .div = 3, .hw.init = &(struct clk_init_data){ - .name = "fclk_div_div3", + .name = "fclk_div3_div", .ops = &clk_fixed_factor_ops, .parent_names = (const char *[]){ "fixed_pll" }, .num_parents = 1, @@ -632,7 +632,8 @@ static struct clk_regmap meson8b_cpu_clk = { .hw.init = &(struct clk_init_data){ .name = "cpu_clk", .ops = &clk_regmap_mux_ro_ops, - .parent_names = (const char *[]){ "xtal", "cpu_out_sel" }, + .parent_names = (const char *[]){ "xtal", + "cpu_scale_out_sel" }, .num_parents = 2, .flags = (CLK_SET_RATE_PARENT | CLK_SET_RATE_NO_REPARENT), diff --git a/drivers/cpufreq/cppc_cpufreq.c b/drivers/cpufreq/cppc_cpufreq.c index bc5fc1630876..b15115a48775 100644 --- a/drivers/cpufreq/cppc_cpufreq.c +++ b/drivers/cpufreq/cppc_cpufreq.c @@ -126,6 +126,49 @@ static void cppc_cpufreq_stop_cpu(struct cpufreq_policy *policy) cpu->perf_caps.lowest_perf, cpu_num, ret); } +/* + * The PCC subspace describes the rate at which platform can accept commands + * on the shared PCC channel (including READs which do not count towards freq + * trasition requests), so ideally we need to use the PCC values as a fallback + * if we don't have a platform specific transition_delay_us + */ +#ifdef CONFIG_ARM64 +#include <asm/cputype.h> + +static unsigned int cppc_cpufreq_get_transition_delay_us(int cpu) +{ + unsigned long implementor = read_cpuid_implementor(); + unsigned long part_num = read_cpuid_part_number(); + unsigned int delay_us = 0; + + switch (implementor) { + case ARM_CPU_IMP_QCOM: + switch (part_num) { + case QCOM_CPU_PART_FALKOR_V1: + case QCOM_CPU_PART_FALKOR: + delay_us = 10000; + break; + default: + delay_us = cppc_get_transition_latency(cpu) / NSEC_PER_USEC; + break; + } + break; + default: + delay_us = cppc_get_transition_latency(cpu) / NSEC_PER_USEC; + break; + } + + return delay_us; +} + +#else + +static unsigned int cppc_cpufreq_get_transition_delay_us(int cpu) +{ + return cppc_get_transition_latency(cpu) / NSEC_PER_USEC; +} +#endif + static int cppc_cpufreq_cpu_init(struct cpufreq_policy *policy) { struct cppc_cpudata *cpu; @@ -162,8 +205,7 @@ static int cppc_cpufreq_cpu_init(struct cpufreq_policy *policy) cpu->perf_caps.highest_perf; policy->cpuinfo.max_freq = cppc_dmi_max_khz; - policy->transition_delay_us = cppc_get_transition_latency(cpu_num) / - NSEC_PER_USEC; + policy->transition_delay_us = cppc_cpufreq_get_transition_delay_us(cpu_num); policy->shared_type = cpu->shared_type; if (policy->shared_type == CPUFREQ_SHARED_TYPE_ANY) { diff --git a/drivers/gpio/gpio-aspeed.c b/drivers/gpio/gpio-aspeed.c index 77e485557498..6f693b7d5220 100644 --- a/drivers/gpio/gpio-aspeed.c +++ b/drivers/gpio/gpio-aspeed.c @@ -384,7 +384,7 @@ static void aspeed_gpio_irq_set_mask(struct irq_data *d, bool set) if (set) reg |= bit; else - reg &= bit; + reg &= ~bit; iowrite32(reg, addr); spin_unlock_irqrestore(&gpio->lock, flags); diff --git a/drivers/gpio/gpio-pci-idio-16.c b/drivers/gpio/gpio-pci-idio-16.c index 1948724d8c36..25d16b2af1c3 100644 --- a/drivers/gpio/gpio-pci-idio-16.c +++ b/drivers/gpio/gpio-pci-idio-16.c @@ -116,9 +116,9 @@ static int idio_16_gpio_get_multiple(struct gpio_chip *chip, unsigned long word_mask; const unsigned long port_mask = GENMASK(gpio_reg_size - 1, 0); unsigned long port_state; - u8 __iomem ports[] = { - idio16gpio->reg->out0_7, idio16gpio->reg->out8_15, - idio16gpio->reg->in0_7, idio16gpio->reg->in8_15, + void __iomem *ports[] = { + &idio16gpio->reg->out0_7, &idio16gpio->reg->out8_15, + &idio16gpio->reg->in0_7, &idio16gpio->reg->in8_15, }; /* clear bits array to a clean slate */ @@ -143,7 +143,7 @@ static int idio_16_gpio_get_multiple(struct gpio_chip *chip, } /* read bits from current gpio port */ - port_state = ioread8(ports + i); + port_state = ioread8(ports[i]); /* store acquired bits at respective bits array offset */ bits[word_index] |= port_state << word_offset; diff --git a/drivers/gpio/gpio-pcie-idio-24.c b/drivers/gpio/gpio-pcie-idio-24.c index 835607ecf658..f953541e7890 100644 --- a/drivers/gpio/gpio-pcie-idio-24.c +++ b/drivers/gpio/gpio-pcie-idio-24.c @@ -206,10 +206,10 @@ static int idio_24_gpio_get_multiple(struct gpio_chip *chip, unsigned long word_mask; const unsigned long port_mask = GENMASK(gpio_reg_size - 1, 0); unsigned long port_state; - u8 __iomem ports[] = { - idio24gpio->reg->out0_7, idio24gpio->reg->out8_15, - idio24gpio->reg->out16_23, idio24gpio->reg->in0_7, - idio24gpio->reg->in8_15, idio24gpio->reg->in16_23, + void __iomem *ports[] = { + &idio24gpio->reg->out0_7, &idio24gpio->reg->out8_15, + &idio24gpio->reg->out16_23, &idio24gpio->reg->in0_7, + &idio24gpio->reg->in8_15, &idio24gpio->reg->in16_23, }; const unsigned long out_mode_mask = BIT(1); @@ -217,7 +217,7 @@ static int idio_24_gpio_get_multiple(struct gpio_chip *chip, bitmap_zero(bits, chip->ngpio); /* get bits are evaluated a gpio port register at a time */ - for (i = 0; i < ARRAY_SIZE(ports); i++) { + for (i = 0; i < ARRAY_SIZE(ports) + 1; i++) { /* gpio offset in bits array */ bits_offset = i * gpio_reg_size; @@ -236,7 +236,7 @@ static int idio_24_gpio_get_multiple(struct gpio_chip *chip, /* read bits from current gpio port (port 6 is TTL GPIO) */ if (i < 6) - port_state = ioread8(ports + i); + port_state = ioread8(ports[i]); else if (ioread8(&idio24gpio->reg->ctl) & out_mode_mask) port_state = ioread8(&idio24gpio->reg->ttl_out0_7); else @@ -301,9 +301,9 @@ static void idio_24_gpio_set_multiple(struct gpio_chip *chip, const unsigned long port_mask = GENMASK(gpio_reg_size, 0); unsigned long flags; unsigned int out_state; - u8 __iomem ports[] = { - idio24gpio->reg->out0_7, idio24gpio->reg->out8_15, - idio24gpio->reg->out16_23 + void __iomem *ports[] = { + &idio24gpio->reg->out0_7, &idio24gpio->reg->out8_15, + &idio24gpio->reg->out16_23 }; const unsigned long out_mode_mask = BIT(1); const unsigned int ttl_offset = 48; @@ -327,9 +327,9 @@ static void idio_24_gpio_set_multiple(struct gpio_chip *chip, raw_spin_lock_irqsave(&idio24gpio->lock, flags); /* process output lines */ - out_state = ioread8(ports + i) & ~gpio_mask; + out_state = ioread8(ports[i]) & ~gpio_mask; out_state |= (*bits >> bits_offset) & gpio_mask; - iowrite8(out_state, ports + i); + iowrite8(out_state, ports[i]); raw_spin_unlock_irqrestore(&idio24gpio->lock, flags); } diff --git a/drivers/gpio/gpiolib.c b/drivers/gpio/gpiolib.c index 43aeb07343ec..d8ccb500872f 100644 --- a/drivers/gpio/gpiolib.c +++ b/drivers/gpio/gpiolib.c @@ -497,7 +497,7 @@ static int linehandle_create(struct gpio_device *gdev, void __user *ip) struct gpiohandle_request handlereq; struct linehandle_state *lh; struct file *file; - int fd, i, ret; + int fd, i, count = 0, ret; u32 lflags; if (copy_from_user(&handlereq, ip, sizeof(handlereq))) @@ -558,6 +558,7 @@ static int linehandle_create(struct gpio_device *gdev, void __user *ip) if (ret) goto out_free_descs; lh->descs[i] = desc; + count = i; if (lflags & GPIOHANDLE_REQUEST_ACTIVE_LOW) set_bit(FLAG_ACTIVE_LOW, &desc->flags); @@ -628,7 +629,7 @@ static int linehandle_create(struct gpio_device *gdev, void __user *ip) out_put_unused_fd: put_unused_fd(fd); out_free_descs: - for (; i >= 0; i--) + for (i = 0; i < count; i++) gpiod_free(lh->descs[i]); kfree(lh->label); out_free_lh: @@ -902,7 +903,7 @@ static int lineevent_create(struct gpio_device *gdev, void __user *ip) desc = &gdev->descs[offset]; ret = gpiod_request(desc, le->label); if (ret) - goto out_free_desc; + goto out_free_label; le->desc = desc; le->eflags = eflags; diff --git a/drivers/gpu/drm/amd/amdgpu/Makefile b/drivers/gpu/drm/amd/amdgpu/Makefile index 68e9f584c570..bfd332c95b61 100644 --- a/drivers/gpu/drm/amd/amdgpu/Makefile +++ b/drivers/gpu/drm/amd/amdgpu/Makefile @@ -56,17 +56,18 @@ amdgpu-y += amdgpu_device.o amdgpu_kms.o \ # add asic specific block amdgpu-$(CONFIG_DRM_AMDGPU_CIK)+= cik.o cik_ih.o kv_smc.o kv_dpm.o \ - ci_smc.o ci_dpm.o dce_v8_0.o gfx_v7_0.o cik_sdma.o uvd_v4_2.o vce_v2_0.o \ - amdgpu_amdkfd_gfx_v7.o + ci_smc.o ci_dpm.o dce_v8_0.o gfx_v7_0.o cik_sdma.o uvd_v4_2.o vce_v2_0.o amdgpu-$(CONFIG_DRM_AMDGPU_SI)+= si.o gmc_v6_0.o gfx_v6_0.o si_ih.o si_dma.o dce_v6_0.o si_dpm.o si_smc.o amdgpu-y += \ - vi.o mxgpu_vi.o nbio_v6_1.o soc15.o emu_soc.o mxgpu_ai.o nbio_v7_0.o vega10_reg_init.o + vi.o mxgpu_vi.o nbio_v6_1.o soc15.o emu_soc.o mxgpu_ai.o nbio_v7_0.o vega10_reg_init.o \ + vega20_reg_init.o # add DF block amdgpu-y += \ - df_v1_7.o + df_v1_7.o \ + df_v3_6.o # add GMC block amdgpu-y += \ @@ -130,13 +131,21 @@ amdgpu-y += \ vcn_v1_0.o # add amdkfd interfaces +amdgpu-y += amdgpu_amdkfd.o + +ifneq ($(CONFIG_HSA_AMD),) amdgpu-y += \ - amdgpu_amdkfd.o \ amdgpu_amdkfd_fence.o \ amdgpu_amdkfd_gpuvm.o \ amdgpu_amdkfd_gfx_v8.o \ amdgpu_amdkfd_gfx_v9.o +ifneq ($(CONFIG_DRM_AMDGPU_CIK),) +amdgpu-y += amdgpu_amdkfd_gfx_v7.o +endif + +endif + # add cgs amdgpu-y += amdgpu_cgs.o diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h index 03a2c0be0bf2..a59c07590cee 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h @@ -1401,6 +1401,8 @@ struct amdgpu_df_funcs { bool enable); void (*get_clockgating_state)(struct amdgpu_device *adev, u32 *flags); + void (*enable_ecc_force_par_wr_rmw)(struct amdgpu_device *adev, + bool enable); }; /* Define the HW IP blocks will be used in driver , add more if necessary */ enum amd_hw_ip_block_type { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c index bd36ee9f7e6d..8f6f45567bfa 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c @@ -50,15 +50,21 @@ int amdgpu_amdkfd_init(void) kgd2kfd = NULL; } + #elif defined(CONFIG_HSA_AMD) + ret = kgd2kfd_init(KFD_INTERFACE_VERSION, &kgd2kfd); if (ret) kgd2kfd = NULL; #else + kgd2kfd = NULL; ret = -ENOENT; #endif + +#if defined(CONFIG_HSA_AMD_MODULE) || defined(CONFIG_HSA_AMD) amdgpu_amdkfd_gpuvm_init_mem_limits(); +#endif return ret; } @@ -97,7 +103,7 @@ void amdgpu_amdkfd_device_probe(struct amdgpu_device *adev) kfd2kgd = amdgpu_amdkfd_gfx_9_0_get_functions(); break; default: - dev_dbg(adev->dev, "kfd not supported on this ASIC\n"); + dev_info(adev->dev, "kfd not supported on this ASIC\n"); return; } @@ -464,3 +470,44 @@ bool amdgpu_amdkfd_is_kfd_vmid(struct amdgpu_device *adev, u32 vmid) return false; } + +#if !defined(CONFIG_HSA_AMD_MODULE) && !defined(CONFIG_HSA_AMD) +bool amdkfd_fence_check_mm(struct dma_fence *f, struct mm_struct *mm) +{ + return false; +} + +void amdgpu_amdkfd_unreserve_system_memory_limit(struct amdgpu_bo *bo) +{ +} + +void amdgpu_amdkfd_gpuvm_destroy_cb(struct amdgpu_device *adev, + struct amdgpu_vm *vm) +{ +} + +struct amdgpu_amdkfd_fence *to_amdgpu_amdkfd_fence(struct dma_fence *f) +{ + return NULL; +} + +int amdgpu_amdkfd_evict_userptr(struct kgd_mem *mem, struct mm_struct *mm) +{ + return 0; +} + +struct kfd2kgd_calls *amdgpu_amdkfd_gfx_7_get_functions(void) +{ + return NULL; +} + +struct kfd2kgd_calls *amdgpu_amdkfd_gfx_8_0_get_functions(void) +{ + return NULL; +} + +struct kfd2kgd_calls *amdgpu_amdkfd_gfx_9_0_get_functions(void) +{ + return NULL; +} +#endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h index 12367a9951e8..a8418a3f4e9d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h @@ -156,14 +156,14 @@ uint64_t amdgpu_amdkfd_get_vram_usage(struct kgd_dev *kgd); /* GPUVM API */ int amdgpu_amdkfd_gpuvm_create_process_vm(struct kgd_dev *kgd, void **vm, - void **process_info, - struct dma_fence **ef); + void **process_info, + struct dma_fence **ef); int amdgpu_amdkfd_gpuvm_acquire_process_vm(struct kgd_dev *kgd, - struct file *filp, - void **vm, void **process_info, - struct dma_fence **ef); + struct file *filp, + void **vm, void **process_info, + struct dma_fence **ef); void amdgpu_amdkfd_gpuvm_destroy_cb(struct amdgpu_device *adev, - struct amdgpu_vm *vm); + struct amdgpu_vm *vm); void amdgpu_amdkfd_gpuvm_destroy_process_vm(struct kgd_dev *kgd, void *vm); uint32_t amdgpu_amdkfd_gpuvm_get_process_page_dir(void *vm); int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu( diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c index 8f37991df61b..f0c0d3953f69 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c @@ -470,9 +470,9 @@ static int kgd_hqd_load(struct kgd_dev *kgd, void *mqd, uint32_t pipe_id, WREG32(SOC15_REG_OFFSET(GC, 0, mmCP_HQD_PQ_WPTR_HI), upper_32_bits(guessed_wptr)); WREG32(SOC15_REG_OFFSET(GC, 0, mmCP_HQD_PQ_WPTR_POLL_ADDR), - lower_32_bits((uint64_t)wptr)); + lower_32_bits((uintptr_t)wptr)); WREG32(SOC15_REG_OFFSET(GC, 0, mmCP_HQD_PQ_WPTR_POLL_ADDR_HI), - upper_32_bits((uint64_t)wptr)); + upper_32_bits((uintptr_t)wptr)); WREG32(SOC15_REG_OFFSET(GC, 0, mmCP_PQ_WPTR_POLL_CNTL1), get_queue_mask(adev, pipe_id, queue_id)); } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c index a0f48cb9b8f0..236915849cfe 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c @@ -322,3 +322,47 @@ int amdgpu_atomfirmware_get_clock_info(struct amdgpu_device *adev) return ret; } + +union gfx_info { + struct atom_gfx_info_v2_4 v24; +}; + +int amdgpu_atomfirmware_get_gfx_info(struct amdgpu_device *adev) +{ + struct amdgpu_mode_info *mode_info = &adev->mode_info; + int index; + uint8_t frev, crev; + uint16_t data_offset; + + index = get_index_into_master_table(atom_master_list_of_data_tables_v2_1, + gfx_info); + if (amdgpu_atom_parse_data_header(mode_info->atom_context, index, NULL, + &frev, &crev, &data_offset)) { + union gfx_info *gfx_info = (union gfx_info *) + (mode_info->atom_context->bios + data_offset); + switch (crev) { + case 4: + adev->gfx.config.max_shader_engines = gfx_info->v24.gc_num_se; + adev->gfx.config.max_cu_per_sh = gfx_info->v24.gc_num_cu_per_sh; + adev->gfx.config.max_sh_per_se = gfx_info->v24.gc_num_sh_per_se; + adev->gfx.config.max_backends_per_se = gfx_info->v24.gc_num_rb_per_se; + adev->gfx.config.max_texture_channel_caches = gfx_info->v24.gc_num_tccs; + adev->gfx.config.max_gprs = le16_to_cpu(gfx_info->v24.gc_num_gprs); + adev->gfx.config.max_gs_threads = gfx_info->v24.gc_num_max_gs_thds; + adev->gfx.config.gs_vgt_table_depth = gfx_info->v24.gc_gs_table_depth; + adev->gfx.config.gs_prim_buffer_depth = + le16_to_cpu(gfx_info->v24.gc_gsprim_buff_depth); + adev->gfx.config.double_offchip_lds_buf = + gfx_info->v24.gc_double_offchip_lds_buffer; + adev->gfx.cu_info.wave_front_size = le16_to_cpu(gfx_info->v24.gc_wave_size); + adev->gfx.cu_info.max_waves_per_simd = le16_to_cpu(gfx_info->v24.gc_max_waves_per_simd); + adev->gfx.cu_info.max_scratch_slots_per_cu = gfx_info->v24.gc_max_scratch_slots_per_cu; + adev->gfx.cu_info.lds_size = le16_to_cpu(gfx_info->v24.gc_lds_size); + return 0; + default: + return -EINVAL; + } + + } + return -EINVAL; +} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.h index 7689c961c4ef..20f158fd3b76 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.h @@ -30,5 +30,6 @@ int amdgpu_atomfirmware_allocate_fb_scratch(struct amdgpu_device *adev); int amdgpu_atomfirmware_get_vram_width(struct amdgpu_device *adev); int amdgpu_atomfirmware_get_vram_type(struct amdgpu_device *adev); int amdgpu_atomfirmware_get_clock_info(struct amdgpu_device *adev); +int amdgpu_atomfirmware_get_gfx_info(struct amdgpu_device *adev); #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c index 5b3d3bf5b599..e950730f1933 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c @@ -400,6 +400,9 @@ static int amdgpu_cgs_get_firmware_info(struct cgs_device *cgs_device, case CHIP_VEGA12: strcpy(fw_name, "amdgpu/vega12_smc.bin"); break; + case CHIP_VEGA20: + strcpy(fw_name, "amdgpu/vega20_smc.bin"); + break; default: DRM_ERROR("SMC firmware not supported\n"); return -EINVAL; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c index a8e531d604fa..c5bb36275e93 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c @@ -173,9 +173,14 @@ static void amdgpu_ctx_do_release(struct kref *ref) ctx = container_of(ref, struct amdgpu_ctx, refcount); - for (i = 0; i < ctx->adev->num_rings; i++) + for (i = 0; i < ctx->adev->num_rings; i++) { + + if (ctx->adev->rings[i] == &ctx->adev->gfx.kiq.ring) + continue; + drm_sched_entity_fini(&ctx->adev->rings[i]->sched, &ctx->rings[i].entity); + } amdgpu_ctx_fini(ref); } @@ -452,12 +457,17 @@ void amdgpu_ctx_mgr_entity_fini(struct amdgpu_ctx_mgr *mgr) if (!ctx->adev) return; - for (i = 0; i < ctx->adev->num_rings; i++) + for (i = 0; i < ctx->adev->num_rings; i++) { + + if (ctx->adev->rings[i] == &ctx->adev->gfx.kiq.ring) + continue; + if (kref_read(&ctx->refcount) == 1) drm_sched_entity_do_release(&ctx->adev->rings[i]->sched, &ctx->rings[i].entity); else DRM_ERROR("ctx %p is still alive\n", ctx); + } } } @@ -474,12 +484,17 @@ void amdgpu_ctx_mgr_entity_cleanup(struct amdgpu_ctx_mgr *mgr) if (!ctx->adev) return; - for (i = 0; i < ctx->adev->num_rings; i++) + for (i = 0; i < ctx->adev->num_rings; i++) { + + if (ctx->adev->rings[i] == &ctx->adev->gfx.kiq.ring) + continue; + if (kref_read(&ctx->refcount) == 1) drm_sched_entity_cleanup(&ctx->adev->rings[i]->sched, &ctx->rings[i].entity); else DRM_ERROR("ctx %p is still alive\n", ctx); + } } } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index 9fb20a53d5b2..290e279abf0d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -86,6 +86,7 @@ static const char *amdgpu_asic_name[] = { "VEGAM", "VEGA10", "VEGA12", + "VEGA20", "RAVEN", "LAST", }; @@ -1387,6 +1388,7 @@ static int amdgpu_device_parse_gpu_info_fw(struct amdgpu_device *adev) case CHIP_KABINI: case CHIP_MULLINS: #endif + case CHIP_VEGA20: default: return 0; case CHIP_VEGA10: @@ -1521,6 +1523,7 @@ static int amdgpu_device_ip_early_init(struct amdgpu_device *adev) #endif case CHIP_VEGA10: case CHIP_VEGA12: + case CHIP_VEGA20: case CHIP_RAVEN: if (adev->asic_type == CHIP_RAVEN) adev->family = AMDGPU_FAMILY_RV; @@ -1715,6 +1718,7 @@ static int amdgpu_device_ip_late_set_cg_state(struct amdgpu_device *adev) /* skip CG for VCE/UVD, it's handled specially */ if (adev->ip_blocks[i].version->type != AMD_IP_BLOCK_TYPE_UVD && adev->ip_blocks[i].version->type != AMD_IP_BLOCK_TYPE_VCE && + adev->ip_blocks[i].version->type != AMD_IP_BLOCK_TYPE_VCN && adev->ip_blocks[i].version->funcs->set_clockgating_state) { /* enable clockgating to save power */ r = adev->ip_blocks[i].version->funcs->set_clockgating_state((void *)adev, @@ -1814,6 +1818,7 @@ static int amdgpu_device_ip_fini(struct amdgpu_device *adev) if (adev->ip_blocks[i].version->type != AMD_IP_BLOCK_TYPE_UVD && adev->ip_blocks[i].version->type != AMD_IP_BLOCK_TYPE_VCE && + adev->ip_blocks[i].version->type != AMD_IP_BLOCK_TYPE_VCN && adev->ip_blocks[i].version->funcs->set_clockgating_state) { /* ungate blocks before hw fini so that we can shutdown the blocks safely */ r = adev->ip_blocks[i].version->funcs->set_clockgating_state((void *)adev, @@ -2155,6 +2160,7 @@ bool amdgpu_device_asic_has_dc_support(enum amd_asic_type asic_type) case CHIP_FIJI: case CHIP_VEGA10: case CHIP_VEGA12: + case CHIP_VEGA20: #if defined(CONFIG_DRM_AMD_DC_DCN1_0) case CHIP_RAVEN: #endif @@ -3172,7 +3178,6 @@ error: int amdgpu_device_gpu_recover(struct amdgpu_device *adev, struct amdgpu_job *job, bool force) { - struct drm_atomic_state *state = NULL; int i, r, resched; if (!force && !amdgpu_device_ip_check_soft_reset(adev)) { @@ -3195,10 +3200,6 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev, /* block TTM */ resched = ttm_bo_lock_delayed_workqueue(&adev->mman.bdev); - /* store modesetting */ - if (amdgpu_device_has_dc_support(adev)) - state = drm_atomic_helper_suspend(adev->ddev); - /* block all schedulers and reset given job's ring */ for (i = 0; i < AMDGPU_MAX_RINGS; ++i) { struct amdgpu_ring *ring = adev->rings[i]; @@ -3238,10 +3239,7 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev, kthread_unpark(ring->sched.thread); } - if (amdgpu_device_has_dc_support(adev)) { - if (drm_atomic_helper_resume(adev->ddev, state)) - dev_info(adev->dev, "drm resume failed:%d\n", r); - } else { + if (!amdgpu_device_has_dc_support(adev)) { drm_helper_resume_force_mode(adev->ddev); } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c index 739e7e09c8b0..b0bf2f24da48 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c @@ -560,6 +560,13 @@ static const struct pci_device_id pciidlist[] = { {0x1002, 0x69A2, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA12}, {0x1002, 0x69A3, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA12}, {0x1002, 0x69AF, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA12}, + /* Vega 20 */ + {0x1002, 0x66A0, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA20|AMD_EXP_HW_SUPPORT}, + {0x1002, 0x66A1, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA20|AMD_EXP_HW_SUPPORT}, + {0x1002, 0x66A2, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA20|AMD_EXP_HW_SUPPORT}, + {0x1002, 0x66A3, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA20|AMD_EXP_HW_SUPPORT}, + {0x1002, 0x66A7, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA20|AMD_EXP_HW_SUPPORT}, + {0x1002, 0x66AF, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA20|AMD_EXP_HW_SUPPORT}, /* Raven */ {0x1002, 0x15dd, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_RAVEN|AMD_IS_APU}, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c index d09fcab2398f..39ec6b8890a1 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c @@ -376,14 +376,14 @@ int amdgpu_fence_driver_start_ring(struct amdgpu_ring *ring, struct amdgpu_device *adev = ring->adev; uint64_t index; - if (ring != &adev->uvd.ring) { + if (ring != &adev->uvd.inst[ring->me].ring) { ring->fence_drv.cpu_addr = &adev->wb.wb[ring->fence_offs]; ring->fence_drv.gpu_addr = adev->wb.gpu_addr + (ring->fence_offs * 4); } else { /* put fence directly behind firmware */ index = ALIGN(adev->uvd.fw->size, 8); - ring->fence_drv.cpu_addr = adev->uvd.cpu_addr + index; - ring->fence_drv.gpu_addr = adev->uvd.gpu_addr + index; + ring->fence_drv.cpu_addr = adev->uvd.inst[ring->me].cpu_addr + index; + ring->fence_drv.gpu_addr = adev->uvd.inst[ring->me].gpu_addr + index; } amdgpu_fence_write(ring, atomic_read(&ring->fence_drv.last_seq)); amdgpu_irq_get(adev, irq_src, irq_type); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c index eb4785e51573..91517b166a3b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c @@ -286,7 +286,7 @@ static int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file struct drm_crtc *crtc; uint32_t ui32 = 0; uint64_t ui64 = 0; - int i, found; + int i, j, found; int ui32_size = sizeof(ui32); if (!info->return_size || !info->return_pointer) @@ -348,7 +348,8 @@ static int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file break; case AMDGPU_HW_IP_UVD: type = AMD_IP_BLOCK_TYPE_UVD; - ring_mask = adev->uvd.ring.ready ? 1 : 0; + for (i = 0; i < adev->uvd.num_uvd_inst; i++) + ring_mask |= ((adev->uvd.inst[i].ring.ready ? 1 : 0) << i); ib_start_alignment = AMDGPU_GPU_PAGE_SIZE; ib_size_alignment = 16; break; @@ -361,8 +362,11 @@ static int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file break; case AMDGPU_HW_IP_UVD_ENC: type = AMD_IP_BLOCK_TYPE_UVD; - for (i = 0; i < adev->uvd.num_enc_rings; i++) - ring_mask |= ((adev->uvd.ring_enc[i].ready ? 1 : 0) << i); + for (i = 0; i < adev->uvd.num_uvd_inst; i++) + for (j = 0; j < adev->uvd.num_enc_rings; j++) + ring_mask |= + ((adev->uvd.inst[i].ring_enc[j].ready ? 1 : 0) << + (j + i * adev->uvd.num_enc_rings)); ib_start_alignment = AMDGPU_GPU_PAGE_SIZE; ib_size_alignment = 1; break; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c index c7d43e064fc7..9f1a5bd39ae8 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c @@ -52,6 +52,7 @@ static int psp_sw_init(void *handle) switch (adev->asic_type) { case CHIP_VEGA10: case CHIP_VEGA12: + case CHIP_VEGA20: psp_v3_1_set_psp_funcs(psp); break; case CHIP_RAVEN: diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_queue_mgr.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_queue_mgr.c index 262c1267249e..8af16e81c7d4 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_queue_mgr.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_queue_mgr.c @@ -66,6 +66,8 @@ static int amdgpu_identity_map(struct amdgpu_device *adev, u32 ring, struct amdgpu_ring **out_ring) { + u32 instance; + switch (mapper->hw_ip) { case AMDGPU_HW_IP_GFX: *out_ring = &adev->gfx.gfx_ring[ring]; @@ -77,13 +79,16 @@ static int amdgpu_identity_map(struct amdgpu_device *adev, *out_ring = &adev->sdma.instance[ring].ring; break; case AMDGPU_HW_IP_UVD: - *out_ring = &adev->uvd.ring; + instance = ring; + *out_ring = &adev->uvd.inst[instance].ring; break; case AMDGPU_HW_IP_VCE: *out_ring = &adev->vce.ring[ring]; break; case AMDGPU_HW_IP_UVD_ENC: - *out_ring = &adev->uvd.ring_enc[ring]; + instance = ring / adev->uvd.num_enc_rings; + *out_ring = + &adev->uvd.inst[instance].ring_enc[ring%adev->uvd.num_enc_rings]; break; case AMDGPU_HW_IP_VCN_DEC: *out_ring = &adev->vcn.ring_dec; @@ -240,13 +245,14 @@ int amdgpu_queue_mgr_map(struct amdgpu_device *adev, ip_num_rings = adev->sdma.num_instances; break; case AMDGPU_HW_IP_UVD: - ip_num_rings = 1; + ip_num_rings = adev->uvd.num_uvd_inst; break; case AMDGPU_HW_IP_VCE: ip_num_rings = adev->vce.num_rings; break; case AMDGPU_HW_IP_UVD_ENC: - ip_num_rings = adev->uvd.num_enc_rings; + ip_num_rings = + adev->uvd.num_enc_rings * adev->uvd.num_uvd_inst; break; case AMDGPU_HW_IP_VCN_DEC: ip_num_rings = 1; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c index 49cad08b5c16..c6850b629d0e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c @@ -362,6 +362,7 @@ void amdgpu_ring_fini(struct amdgpu_ring *ring) dma_fence_put(ring->vmid_wait); ring->vmid_wait = NULL; + ring->me = 0; ring->adev->rings[ring->idx] = NULL; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h index 4f8dac2d36a5..1513124c5659 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h @@ -29,7 +29,7 @@ #include <drm/drm_print.h> /* max number of rings */ -#define AMDGPU_MAX_RINGS 18 +#define AMDGPU_MAX_RINGS 21 #define AMDGPU_MAX_GFX_RINGS 1 #define AMDGPU_MAX_COMPUTE_RINGS 8 #define AMDGPU_MAX_VCE_RINGS 3 diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c index 69a2b25b3696..e93a0a237dc3 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c @@ -63,16 +63,44 @@ static void amdgpu_ttm_debugfs_fini(struct amdgpu_device *adev); /* * Global memory. */ + +/** + * amdgpu_ttm_mem_global_init - Initialize and acquire reference to + * memory object + * + * @ref: Object for initialization. + * + * This is called by drm_global_item_ref() when an object is being + * initialized. + */ static int amdgpu_ttm_mem_global_init(struct drm_global_reference *ref) { return ttm_mem_global_init(ref->object); } +/** + * amdgpu_ttm_mem_global_release - Drop reference to a memory object + * + * @ref: Object being removed + * + * This is called by drm_global_item_unref() when an object is being + * released. + */ static void amdgpu_ttm_mem_global_release(struct drm_global_reference *ref) { ttm_mem_global_release(ref->object); } +/** + * amdgpu_ttm_global_init - Initialize global TTM memory reference + * structures. + * + * @adev: AMDGPU device for which the global structures need to be + * registered. + * + * This is called as part of the AMDGPU ttm init from amdgpu_ttm_init() + * during bring up. + */ static int amdgpu_ttm_global_init(struct amdgpu_device *adev) { struct drm_global_reference *global_ref; @@ -80,7 +108,9 @@ static int amdgpu_ttm_global_init(struct amdgpu_device *adev) struct drm_sched_rq *rq; int r; + /* ensure reference is false in case init fails */ adev->mman.mem_global_referenced = false; + global_ref = &adev->mman.mem_global_ref; global_ref->global_type = DRM_GLOBAL_TTM_MEM; global_ref->size = sizeof(struct ttm_mem_global); @@ -146,6 +176,18 @@ static int amdgpu_invalidate_caches(struct ttm_bo_device *bdev, uint32_t flags) return 0; } +/** + * amdgpu_init_mem_type - Initialize a memory manager for a specific + * type of memory request. + * + * @bdev: The TTM BO device object (contains a reference to + * amdgpu_device) + * @type: The type of memory requested + * @man: + * + * This is called by ttm_bo_init_mm() when a buffer object is being + * initialized. + */ static int amdgpu_init_mem_type(struct ttm_bo_device *bdev, uint32_t type, struct ttm_mem_type_manager *man) { @@ -161,6 +203,7 @@ static int amdgpu_init_mem_type(struct ttm_bo_device *bdev, uint32_t type, man->default_caching = TTM_PL_FLAG_CACHED; break; case TTM_PL_TT: + /* GTT memory */ man->func = &amdgpu_gtt_mgr_func; man->gpu_offset = adev->gmc.gart_start; man->available_caching = TTM_PL_MASK_CACHING; @@ -193,6 +236,14 @@ static int amdgpu_init_mem_type(struct ttm_bo_device *bdev, uint32_t type, return 0; } +/** + * amdgpu_evict_flags - Compute placement flags + * + * @bo: The buffer object to evict + * @placement: Possible destination(s) for evicted BO + * + * Fill in placement data when ttm_bo_evict() is called + */ static void amdgpu_evict_flags(struct ttm_buffer_object *bo, struct ttm_placement *placement) { @@ -204,12 +255,14 @@ static void amdgpu_evict_flags(struct ttm_buffer_object *bo, .flags = TTM_PL_MASK_CACHING | TTM_PL_FLAG_SYSTEM }; + /* Don't handle scatter gather BOs */ if (bo->type == ttm_bo_type_sg) { placement->num_placement = 0; placement->num_busy_placement = 0; return; } + /* Object isn't an AMDGPU object so ignore */ if (!amdgpu_ttm_bo_is_amdgpu_bo(bo)) { placement->placement = &placements; placement->busy_placement = &placements; @@ -217,10 +270,12 @@ static void amdgpu_evict_flags(struct ttm_buffer_object *bo, placement->num_busy_placement = 1; return; } + abo = ttm_to_amdgpu_bo(bo); switch (bo->mem.mem_type) { case TTM_PL_VRAM: if (!adev->mman.buffer_funcs_enabled) { + /* Move to system memory */ amdgpu_ttm_placement_from_domain(abo, AMDGPU_GEM_DOMAIN_CPU); } else if (adev->gmc.visible_vram_size < adev->gmc.real_vram_size && !(abo->flags & AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED) && @@ -238,6 +293,7 @@ static void amdgpu_evict_flags(struct ttm_buffer_object *bo, abo->placement.busy_placement = &abo->placements[1]; abo->placement.num_busy_placement = 1; } else { + /* Move to GTT memory */ amdgpu_ttm_placement_from_domain(abo, AMDGPU_GEM_DOMAIN_GTT); } break; @@ -248,6 +304,15 @@ static void amdgpu_evict_flags(struct ttm_buffer_object *bo, *placement = abo->placement; } +/** + * amdgpu_verify_access - Verify access for a mmap call + * + * @bo: The buffer object to map + * @filp: The file pointer from the process performing the mmap + * + * This is called by ttm_bo_mmap() to verify whether a process + * has the right to mmap a BO to their process space. + */ static int amdgpu_verify_access(struct ttm_buffer_object *bo, struct file *filp) { struct amdgpu_bo *abo = ttm_to_amdgpu_bo(bo); @@ -265,6 +330,15 @@ static int amdgpu_verify_access(struct ttm_buffer_object *bo, struct file *filp) filp->private_data); } +/** + * amdgpu_move_null - Register memory for a buffer object + * + * @bo: The bo to assign the memory to + * @new_mem: The memory to be assigned. + * + * Assign the memory from new_mem to the memory of the buffer object + * bo. + */ static void amdgpu_move_null(struct ttm_buffer_object *bo, struct ttm_mem_reg *new_mem) { @@ -275,6 +349,10 @@ static void amdgpu_move_null(struct ttm_buffer_object *bo, new_mem->mm_node = NULL; } +/** + * amdgpu_mm_node_addr - Compute the GPU relative offset of a GTT + * buffer. + */ static uint64_t amdgpu_mm_node_addr(struct ttm_buffer_object *bo, struct drm_mm_node *mm_node, struct ttm_mem_reg *mem) @@ -289,9 +367,10 @@ static uint64_t amdgpu_mm_node_addr(struct ttm_buffer_object *bo, } /** - * amdgpu_find_mm_node - Helper function finds the drm_mm_node - * corresponding to @offset. It also modifies the offset to be - * within the drm_mm_node returned + * amdgpu_find_mm_node - Helper function finds the drm_mm_node + * corresponding to @offset. It also modifies + * the offset to be within the drm_mm_node + * returned */ static struct drm_mm_node *amdgpu_find_mm_node(struct ttm_mem_reg *mem, unsigned long *offset) @@ -430,7 +509,12 @@ error: return r; } - +/** + * amdgpu_move_blit - Copy an entire buffer to another buffer + * + * This is a helper called by amdgpu_bo_move() and + * amdgpu_move_vram_ram() to help move buffers to and from VRAM. + */ static int amdgpu_move_blit(struct ttm_buffer_object *bo, bool evict, bool no_wait_gpu, struct ttm_mem_reg *new_mem, @@ -465,6 +549,11 @@ error: return r; } +/** + * amdgpu_move_vram_ram - Copy VRAM buffer to RAM buffer + * + * Called by amdgpu_bo_move(). + */ static int amdgpu_move_vram_ram(struct ttm_buffer_object *bo, bool evict, struct ttm_operation_ctx *ctx, struct ttm_mem_reg *new_mem) @@ -477,6 +566,8 @@ static int amdgpu_move_vram_ram(struct ttm_buffer_object *bo, bool evict, int r; adev = amdgpu_ttm_adev(bo->bdev); + + /* create space/pages for new_mem in GTT space */ tmp_mem = *new_mem; tmp_mem.mm_node = NULL; placement.num_placement = 1; @@ -491,25 +582,36 @@ static int amdgpu_move_vram_ram(struct ttm_buffer_object *bo, bool evict, return r; } + /* set caching flags */ r = ttm_tt_set_placement_caching(bo->ttm, tmp_mem.placement); if (unlikely(r)) { goto out_cleanup; } + /* Bind the memory to the GTT space */ r = ttm_tt_bind(bo->ttm, &tmp_mem, ctx); if (unlikely(r)) { goto out_cleanup; } + + /* blit VRAM to GTT */ r = amdgpu_move_blit(bo, true, ctx->no_wait_gpu, &tmp_mem, old_mem); if (unlikely(r)) { goto out_cleanup; } + + /* move BO (in tmp_mem) to new_mem */ r = ttm_bo_move_ttm(bo, ctx, new_mem); out_cleanup: ttm_bo_mem_put(bo, &tmp_mem); return r; } +/** + * amdgpu_move_ram_vram - Copy buffer from RAM to VRAM + * + * Called by amdgpu_bo_move(). + */ static int amdgpu_move_ram_vram(struct ttm_buffer_object *bo, bool evict, struct ttm_operation_ctx *ctx, struct ttm_mem_reg *new_mem) @@ -522,6 +624,8 @@ static int amdgpu_move_ram_vram(struct ttm_buffer_object *bo, bool evict, int r; adev = amdgpu_ttm_adev(bo->bdev); + + /* make space in GTT for old_mem buffer */ tmp_mem = *new_mem; tmp_mem.mm_node = NULL; placement.num_placement = 1; @@ -535,10 +639,14 @@ static int amdgpu_move_ram_vram(struct ttm_buffer_object *bo, bool evict, if (unlikely(r)) { return r; } + + /* move/bind old memory to GTT space */ r = ttm_bo_move_ttm(bo, ctx, &tmp_mem); if (unlikely(r)) { goto out_cleanup; } + + /* copy to VRAM */ r = amdgpu_move_blit(bo, true, ctx->no_wait_gpu, new_mem, old_mem); if (unlikely(r)) { goto out_cleanup; @@ -548,6 +656,11 @@ out_cleanup: return r; } +/** + * amdgpu_bo_move - Move a buffer object to a new memory location + * + * Called by ttm_bo_handle_move_mem() + */ static int amdgpu_bo_move(struct ttm_buffer_object *bo, bool evict, struct ttm_operation_ctx *ctx, struct ttm_mem_reg *new_mem) @@ -613,6 +726,11 @@ memcpy: return 0; } +/** + * amdgpu_ttm_io_mem_reserve - Reserve a block of memory during a fault + * + * Called by ttm_mem_io_reserve() ultimately via ttm_bo_vm_fault() + */ static int amdgpu_ttm_io_mem_reserve(struct ttm_bo_device *bdev, struct ttm_mem_reg *mem) { struct ttm_mem_type_manager *man = &bdev->man[mem->mem_type]; @@ -690,6 +808,14 @@ struct amdgpu_ttm_tt { uint32_t last_set_pages; }; +/** + * amdgpu_ttm_tt_get_user_pages - Pin pages of memory pointed to + * by a USERPTR pointer to memory + * + * Called by amdgpu_gem_userptr_ioctl() and amdgpu_cs_parser_bos(). + * This provides a wrapper around the get_user_pages() call to provide + * device accessible pages that back user memory. + */ int amdgpu_ttm_tt_get_user_pages(struct ttm_tt *ttm, struct page **pages) { struct amdgpu_ttm_tt *gtt = (void *)ttm; @@ -719,6 +845,7 @@ int amdgpu_ttm_tt_get_user_pages(struct ttm_tt *ttm, struct page **pages) } } + /* loop enough times using contiguous pages of memory */ do { unsigned num_pages = ttm->num_pages - pinned; uint64_t userptr = gtt->userptr + pinned * PAGE_SIZE; @@ -757,6 +884,14 @@ release_pages: return r; } +/** + * amdgpu_ttm_tt_set_user_pages - Copy pages in, putting old pages + * as necessary. + * + * Called by amdgpu_cs_list_validate(). This creates the page list + * that backs user memory and will ultimately be mapped into the device + * address space. + */ void amdgpu_ttm_tt_set_user_pages(struct ttm_tt *ttm, struct page **pages) { struct amdgpu_ttm_tt *gtt = (void *)ttm; @@ -771,6 +906,11 @@ void amdgpu_ttm_tt_set_user_pages(struct ttm_tt *ttm, struct page **pages) } } +/** + * amdgpu_ttm_tt_mark_user_page - Mark pages as dirty + * + * Called while unpinning userptr pages + */ void amdgpu_ttm_tt_mark_user_pages(struct ttm_tt *ttm) { struct amdgpu_ttm_tt *gtt = (void *)ttm; @@ -789,7 +929,12 @@ void amdgpu_ttm_tt_mark_user_pages(struct ttm_tt *ttm) } } -/* prepare the sg table with the user pages */ +/** + * amdgpu_ttm_tt_pin_userptr - prepare the sg table with the + * user pages + * + * Called by amdgpu_ttm_backend_bind() + **/ static int amdgpu_ttm_tt_pin_userptr(struct ttm_tt *ttm) { struct amdgpu_device *adev = amdgpu_ttm_adev(ttm->bdev); @@ -801,17 +946,20 @@ static int amdgpu_ttm_tt_pin_userptr(struct ttm_tt *ttm) enum dma_data_direction direction = write ? DMA_BIDIRECTIONAL : DMA_TO_DEVICE; + /* Allocate an SG array and squash pages into it */ r = sg_alloc_table_from_pages(ttm->sg, ttm->pages, ttm->num_pages, 0, ttm->num_pages << PAGE_SHIFT, GFP_KERNEL); if (r) goto release_sg; + /* Map SG to device */ r = -ENOMEM; nents = dma_map_sg(adev->dev, ttm->sg->sgl, ttm->sg->nents, direction); if (nents != ttm->sg->nents) goto release_sg; + /* convert SG to linear array of pages and dma addresses */ drm_prime_sg_to_page_addr_arrays(ttm->sg, ttm->pages, gtt->ttm.dma_address, ttm->num_pages); @@ -822,6 +970,9 @@ release_sg: return r; } +/** + * amdgpu_ttm_tt_unpin_userptr - Unpin and unmap userptr pages + */ static void amdgpu_ttm_tt_unpin_userptr(struct ttm_tt *ttm) { struct amdgpu_device *adev = amdgpu_ttm_adev(ttm->bdev); @@ -835,9 +986,10 @@ static void amdgpu_ttm_tt_unpin_userptr(struct ttm_tt *ttm) if (!ttm->sg->sgl) return; - /* free the sg table and pages again */ + /* unmap the pages mapped to the device */ dma_unmap_sg(adev->dev, ttm->sg->sgl, ttm->sg->nents, direction); + /* mark the pages as dirty */ amdgpu_ttm_tt_mark_user_pages(ttm); sg_free_table(ttm->sg); @@ -882,6 +1034,12 @@ gart_bind_fail: return r; } +/** + * amdgpu_ttm_backend_bind - Bind GTT memory + * + * Called by ttm_tt_bind() on behalf of ttm_bo_handle_move_mem(). + * This handles binding GTT memory to the device address space. + */ static int amdgpu_ttm_backend_bind(struct ttm_tt *ttm, struct ttm_mem_reg *bo_mem) { @@ -912,7 +1070,10 @@ static int amdgpu_ttm_backend_bind(struct ttm_tt *ttm, return 0; } + /* compute PTE flags relevant to this BO memory */ flags = amdgpu_ttm_tt_pte_flags(adev, ttm, bo_mem); + + /* bind pages into GART page tables */ gtt->offset = (u64)bo_mem->start << PAGE_SHIFT; r = amdgpu_gart_bind(adev, gtt->offset, ttm->num_pages, ttm->pages, gtt->ttm.dma_address, flags); @@ -923,6 +1084,9 @@ static int amdgpu_ttm_backend_bind(struct ttm_tt *ttm, return r; } +/** + * amdgpu_ttm_alloc_gart - Allocate GART memory for buffer object + */ int amdgpu_ttm_alloc_gart(struct ttm_buffer_object *bo) { struct amdgpu_device *adev = amdgpu_ttm_adev(bo->bdev); @@ -938,6 +1102,7 @@ int amdgpu_ttm_alloc_gart(struct ttm_buffer_object *bo) amdgpu_gtt_mgr_has_gart_addr(&bo->mem)) return 0; + /* allocate GTT space */ tmp = bo->mem; tmp.mm_node = NULL; placement.num_placement = 1; @@ -953,7 +1118,10 @@ int amdgpu_ttm_alloc_gart(struct ttm_buffer_object *bo) if (unlikely(r)) return r; + /* compute PTE flags for this buffer object */ flags = amdgpu_ttm_tt_pte_flags(adev, bo->ttm, &tmp); + + /* Bind pages */ gtt->offset = (u64)tmp.start << PAGE_SHIFT; r = amdgpu_ttm_gart_bind(adev, bo, flags); if (unlikely(r)) { @@ -969,6 +1137,12 @@ int amdgpu_ttm_alloc_gart(struct ttm_buffer_object *bo) return 0; } +/** + * amdgpu_ttm_recover_gart - Rebind GTT pages + * + * Called by amdgpu_gtt_mgr_recover() from amdgpu_device_reset() to + * rebind GTT pages during a GPU reset. + */ int amdgpu_ttm_recover_gart(struct ttm_buffer_object *tbo) { struct amdgpu_device *adev = amdgpu_ttm_adev(tbo->bdev); @@ -984,12 +1158,19 @@ int amdgpu_ttm_recover_gart(struct ttm_buffer_object *tbo) return r; } +/** + * amdgpu_ttm_backend_unbind - Unbind GTT mapped pages + * + * Called by ttm_tt_unbind() on behalf of ttm_bo_move_ttm() and + * ttm_tt_destroy(). + */ static int amdgpu_ttm_backend_unbind(struct ttm_tt *ttm) { struct amdgpu_device *adev = amdgpu_ttm_adev(ttm->bdev); struct amdgpu_ttm_tt *gtt = (void *)ttm; int r; + /* if the pages have userptr pinning then clear that first */ if (gtt->userptr) amdgpu_ttm_tt_unpin_userptr(ttm); @@ -1021,6 +1202,13 @@ static struct ttm_backend_func amdgpu_backend_func = { .destroy = &amdgpu_ttm_backend_destroy, }; +/** + * amdgpu_ttm_tt_create - Create a ttm_tt object for a given BO + * + * @bo: The buffer object to create a GTT ttm_tt object around + * + * Called by ttm_tt_create(). + */ static struct ttm_tt *amdgpu_ttm_tt_create(struct ttm_buffer_object *bo, uint32_t page_flags) { @@ -1034,6 +1222,8 @@ static struct ttm_tt *amdgpu_ttm_tt_create(struct ttm_buffer_object *bo, return NULL; } gtt->ttm.ttm.func = &amdgpu_backend_func; + + /* allocate space for the uninitialized page entries */ if (ttm_sg_tt_init(>t->ttm, bo, page_flags)) { kfree(gtt); return NULL; @@ -1041,6 +1231,12 @@ static struct ttm_tt *amdgpu_ttm_tt_create(struct ttm_buffer_object *bo, return >t->ttm.ttm; } +/** + * amdgpu_ttm_tt_populate - Map GTT pages visible to the device + * + * Map the pages of a ttm_tt object to an address space visible + * to the underlying device. + */ static int amdgpu_ttm_tt_populate(struct ttm_tt *ttm, struct ttm_operation_ctx *ctx) { @@ -1048,6 +1244,7 @@ static int amdgpu_ttm_tt_populate(struct ttm_tt *ttm, struct amdgpu_ttm_tt *gtt = (void *)ttm; bool slave = !!(ttm->page_flags & TTM_PAGE_FLAG_SG); + /* user pages are bound by amdgpu_ttm_tt_pin_userptr() */ if (gtt && gtt->userptr) { ttm->sg = kzalloc(sizeof(struct sg_table), GFP_KERNEL); if (!ttm->sg) @@ -1072,9 +1269,17 @@ static int amdgpu_ttm_tt_populate(struct ttm_tt *ttm, } #endif + /* fall back to generic helper to populate the page array + * and map them to the device */ return ttm_populate_and_map_pages(adev->dev, >t->ttm, ctx); } +/** + * amdgpu_ttm_tt_unpopulate - unmap GTT pages and unpopulate page arrays + * + * Unmaps pages of a ttm_tt object from the device address space and + * unpopulates the page array backing it. + */ static void amdgpu_ttm_tt_unpopulate(struct ttm_tt *ttm) { struct amdgpu_device *adev; @@ -1100,9 +1305,21 @@ static void amdgpu_ttm_tt_unpopulate(struct ttm_tt *ttm) } #endif + /* fall back to generic helper to unmap and unpopulate array */ ttm_unmap_and_unpopulate_pages(adev->dev, >t->ttm); } +/** + * amdgpu_ttm_tt_set_userptr - Initialize userptr GTT ttm_tt + * for the current task + * + * @ttm: The ttm_tt object to bind this userptr object to + * @addr: The address in the current tasks VM space to use + * @flags: Requirements of userptr object. + * + * Called by amdgpu_gem_userptr_ioctl() to bind userptr pages + * to current task + */ int amdgpu_ttm_tt_set_userptr(struct ttm_tt *ttm, uint64_t addr, uint32_t flags) { @@ -1127,6 +1344,9 @@ int amdgpu_ttm_tt_set_userptr(struct ttm_tt *ttm, uint64_t addr, return 0; } +/** + * amdgpu_ttm_tt_get_usermm - Return memory manager for ttm_tt object + */ struct mm_struct *amdgpu_ttm_tt_get_usermm(struct ttm_tt *ttm) { struct amdgpu_ttm_tt *gtt = (void *)ttm; @@ -1140,6 +1360,12 @@ struct mm_struct *amdgpu_ttm_tt_get_usermm(struct ttm_tt *ttm) return gtt->usertask->mm; } +/** + * amdgpu_ttm_tt_affect_userptr - Determine if a ttm_tt object lays + * inside an address range for the + * current task. + * + */ bool amdgpu_ttm_tt_affect_userptr(struct ttm_tt *ttm, unsigned long start, unsigned long end) { @@ -1150,10 +1376,16 @@ bool amdgpu_ttm_tt_affect_userptr(struct ttm_tt *ttm, unsigned long start, if (gtt == NULL || !gtt->userptr) return false; + /* Return false if no part of the ttm_tt object lies within + * the range + */ size = (unsigned long)gtt->ttm.ttm.num_pages * PAGE_SIZE; if (gtt->userptr > end || gtt->userptr + size <= start) return false; + /* Search the lists of tasks that hold this mapping and see + * if current is one of them. If it is return false. + */ spin_lock(>t->guptasklock); list_for_each_entry(entry, >t->guptasks, list) { if (entry->task == current) { @@ -1168,6 +1400,10 @@ bool amdgpu_ttm_tt_affect_userptr(struct ttm_tt *ttm, unsigned long start, return true; } +/** + * amdgpu_ttm_tt_userptr_invalidated - Has the ttm_tt object been + * invalidated? + */ bool amdgpu_ttm_tt_userptr_invalidated(struct ttm_tt *ttm, int *last_invalidated) { @@ -1178,6 +1414,12 @@ bool amdgpu_ttm_tt_userptr_invalidated(struct ttm_tt *ttm, return prev_invalidated != *last_invalidated; } +/** + * amdgpu_ttm_tt_userptr_needs_pages - Have the pages backing this + * ttm_tt object been invalidated + * since the last time they've + * been set? + */ bool amdgpu_ttm_tt_userptr_needs_pages(struct ttm_tt *ttm) { struct amdgpu_ttm_tt *gtt = (void *)ttm; @@ -1188,6 +1430,9 @@ bool amdgpu_ttm_tt_userptr_needs_pages(struct ttm_tt *ttm) return atomic_read(>t->mmu_invalidations) != gtt->last_set_pages; } +/** + * amdgpu_ttm_tt_is_readonly - Is the ttm_tt object read only? + */ bool amdgpu_ttm_tt_is_readonly(struct ttm_tt *ttm) { struct amdgpu_ttm_tt *gtt = (void *)ttm; @@ -1198,6 +1443,12 @@ bool amdgpu_ttm_tt_is_readonly(struct ttm_tt *ttm) return !!(gtt->userflags & AMDGPU_GEM_USERPTR_READONLY); } +/** + * amdgpu_ttm_tt_pte_flags - Compute PTE flags for ttm_tt object + * + * @ttm: The ttm_tt object to compute the flags for + * @mem: The memory registry backing this ttm_tt object + */ uint64_t amdgpu_ttm_tt_pte_flags(struct amdgpu_device *adev, struct ttm_tt *ttm, struct ttm_mem_reg *mem) { @@ -1222,6 +1473,16 @@ uint64_t amdgpu_ttm_tt_pte_flags(struct amdgpu_device *adev, struct ttm_tt *ttm, return flags; } +/** + * amdgpu_ttm_bo_eviction_valuable - Check to see if we can evict + * a buffer object. + * + * Return true if eviction is sensible. Called by + * ttm_mem_evict_first() on behalf of ttm_bo_mem_force_space() + * which tries to evict buffer objects until it can find space + * for a new object and by ttm_bo_force_list_clean() which is + * used to clean out a memory space. + */ static bool amdgpu_ttm_bo_eviction_valuable(struct ttm_buffer_object *bo, const struct ttm_place *place) { @@ -1268,6 +1529,19 @@ static bool amdgpu_ttm_bo_eviction_valuable(struct ttm_buffer_object *bo, return ttm_bo_eviction_valuable(bo, place); } +/** + * amdgpu_ttm_access_memory - Read or Write memory that backs a + * buffer object. + * + * @bo: The buffer object to read/write + * @offset: Offset into buffer object + * @buf: Secondary buffer to write/read from + * @len: Length in bytes of access + * @write: true if writing + * + * This is used to access VRAM that backs a buffer object via MMIO + * access for debugging purposes. + */ static int amdgpu_ttm_access_memory(struct ttm_buffer_object *bo, unsigned long offset, void *buf, int len, int write) @@ -1444,13 +1718,22 @@ error_create: adev->fw_vram_usage.reserved_bo = NULL; return r; } - +/** + * amdgpu_ttm_init - Init the memory management (ttm) as well as + * various gtt/vram related fields. + * + * This initializes all of the memory space pools that the TTM layer + * will need such as the GTT space (system memory mapped to the device), + * VRAM (on-board memory), and on-chip memories (GDS, GWS, OA) which + * can be mapped per VMID. + */ int amdgpu_ttm_init(struct amdgpu_device *adev) { uint64_t gtt_size; int r; u64 vis_vram_limit; + /* initialize global references for vram/gtt */ r = amdgpu_ttm_global_init(adev); if (r) { return r; @@ -1471,6 +1754,7 @@ int amdgpu_ttm_init(struct amdgpu_device *adev) /* We opt to avoid OOM on system pages allocations */ adev->mman.bdev.no_retry = true; + /* Initialize VRAM pool with all of VRAM divided into pages */ r = ttm_bo_init_mm(&adev->mman.bdev, TTM_PL_VRAM, adev->gmc.real_vram_size >> PAGE_SHIFT); if (r) { @@ -1500,6 +1784,10 @@ int amdgpu_ttm_init(struct amdgpu_device *adev) return r; } + /* allocate memory as required for VGA + * This is used for VGA emulation and pre-OS scanout buffers to + * avoid display artifacts while transitioning between pre-OS + * and driver. */ if (adev->gmc.stolen_size) { r = amdgpu_bo_create_kernel(adev, adev->gmc.stolen_size, PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM, @@ -1511,6 +1799,8 @@ int amdgpu_ttm_init(struct amdgpu_device *adev) DRM_INFO("amdgpu: %uM of VRAM memory ready\n", (unsigned) (adev->gmc.real_vram_size / (1024 * 1024))); + /* Compute GTT size, either bsaed on 3/4th the size of RAM size + * or whatever the user passed on module init */ if (amdgpu_gtt_size == -1) { struct sysinfo si; @@ -1521,6 +1811,8 @@ int amdgpu_ttm_init(struct amdgpu_device *adev) } else gtt_size = (uint64_t)amdgpu_gtt_size << 20; + + /* Initialize GTT memory pool */ r = ttm_bo_init_mm(&adev->mman.bdev, TTM_PL_TT, gtt_size >> PAGE_SHIFT); if (r) { DRM_ERROR("Failed initializing GTT heap.\n"); @@ -1529,6 +1821,7 @@ int amdgpu_ttm_init(struct amdgpu_device *adev) DRM_INFO("amdgpu: %uM of GTT memory ready.\n", (unsigned)(gtt_size / (1024 * 1024))); + /* Initialize various on-chip memory pools */ adev->gds.mem.total_size = adev->gds.mem.total_size << AMDGPU_GDS_SHIFT; adev->gds.mem.gfx_partition_size = adev->gds.mem.gfx_partition_size << AMDGPU_GDS_SHIFT; adev->gds.mem.cs_partition_size = adev->gds.mem.cs_partition_size << AMDGPU_GDS_SHIFT; @@ -1568,6 +1861,7 @@ int amdgpu_ttm_init(struct amdgpu_device *adev) } } + /* Register debugfs entries for amdgpu_ttm */ r = amdgpu_ttm_debugfs_init(adev); if (r) { DRM_ERROR("Failed to init debugfs\n"); @@ -1576,11 +1870,19 @@ int amdgpu_ttm_init(struct amdgpu_device *adev) return 0; } +/** + * amdgpu_ttm_late_init - Handle any late initialization for + * amdgpu_ttm + */ void amdgpu_ttm_late_init(struct amdgpu_device *adev) { + /* return the VGA stolen memory (if any) back to VRAM */ amdgpu_bo_free_kernel(&adev->stolen_vga_memory, NULL, NULL); } +/** + * amdgpu_ttm_fini - De-initialize the TTM memory pools + */ void amdgpu_ttm_fini(struct amdgpu_device *adev) { if (!adev->mman.initialized) @@ -1908,6 +2210,11 @@ static const struct drm_info_list amdgpu_ttm_debugfs_list[] = { #endif }; +/** + * amdgpu_ttm_vram_read - Linear read access to VRAM + * + * Accesses VRAM via MMIO for debugging purposes. + */ static ssize_t amdgpu_ttm_vram_read(struct file *f, char __user *buf, size_t size, loff_t *pos) { @@ -1947,6 +2254,11 @@ static ssize_t amdgpu_ttm_vram_read(struct file *f, char __user *buf, return result; } +/** + * amdgpu_ttm_vram_write - Linear write access to VRAM + * + * Accesses VRAM via MMIO for debugging purposes. + */ static ssize_t amdgpu_ttm_vram_write(struct file *f, const char __user *buf, size_t size, loff_t *pos) { @@ -1995,6 +2307,9 @@ static const struct file_operations amdgpu_ttm_vram_fops = { #ifdef CONFIG_DRM_AMDGPU_GART_DEBUGFS +/** + * amdgpu_ttm_gtt_read - Linear read access to GTT memory + */ static ssize_t amdgpu_ttm_gtt_read(struct file *f, char __user *buf, size_t size, loff_t *pos) { @@ -2042,6 +2357,13 @@ static const struct file_operations amdgpu_ttm_gtt_fops = { #endif +/** + * amdgpu_iomem_read - Virtual read access to GPU mapped memory + * + * This function is used to read memory that has been mapped to the + * GPU and the known addresses are not physical addresses but instead + * bus addresses (e.g., what you'd put in an IB or ring buffer). + */ static ssize_t amdgpu_iomem_read(struct file *f, char __user *buf, size_t size, loff_t *pos) { @@ -2050,6 +2372,7 @@ static ssize_t amdgpu_iomem_read(struct file *f, char __user *buf, ssize_t result = 0; int r; + /* retrieve the IOMMU domain if any for this device */ dom = iommu_get_domain_for_dev(adev->dev); while (size) { @@ -2062,6 +2385,10 @@ static ssize_t amdgpu_iomem_read(struct file *f, char __user *buf, bytes = bytes < size ? bytes : size; + /* Translate the bus address to a physical address. If + * the domain is NULL it means there is no IOMMU active + * and the address translation is the identity + */ addr = dom ? iommu_iova_to_phys(dom, addr) : addr; pfn = addr >> PAGE_SHIFT; @@ -2086,6 +2413,13 @@ static ssize_t amdgpu_iomem_read(struct file *f, char __user *buf, return result; } +/** + * amdgpu_iomem_write - Virtual write access to GPU mapped memory + * + * This function is used to write memory that has been mapped to the + * GPU and the known addresses are not physical addresses but instead + * bus addresses (e.g., what you'd put in an IB or ring buffer). + */ static ssize_t amdgpu_iomem_write(struct file *f, const char __user *buf, size_t size, loff_t *pos) { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c index 75592bd04d6a..f55f72a37ca8 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c @@ -307,6 +307,8 @@ amdgpu_ucode_get_load_type(struct amdgpu_device *adev, int load_type) return AMDGPU_FW_LOAD_DIRECT; else return AMDGPU_FW_LOAD_PSP; + case CHIP_VEGA20: + return AMDGPU_FW_LOAD_DIRECT; default: DRM_ERROR("Unknown firmware load type\n"); } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c index de4d77af02ae..bcf68f80bbf0 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c @@ -70,12 +70,14 @@ #define FIRMWARE_VEGA10 "amdgpu/vega10_uvd.bin" #define FIRMWARE_VEGA12 "amdgpu/vega12_uvd.bin" +#define FIRMWARE_VEGA20 "amdgpu/vega20_uvd.bin" -#define mmUVD_GPCOM_VCPU_DATA0_VEGA10 (0x03c4 + 0x7e00) -#define mmUVD_GPCOM_VCPU_DATA1_VEGA10 (0x03c5 + 0x7e00) -#define mmUVD_GPCOM_VCPU_CMD_VEGA10 (0x03c3 + 0x7e00) -#define mmUVD_NO_OP_VEGA10 (0x03ff + 0x7e00) -#define mmUVD_ENGINE_CNTL_VEGA10 (0x03c6 + 0x7e00) +/* These are common relative offsets for all asics, from uvd_7_0_offset.h, */ +#define UVD_GPCOM_VCPU_CMD 0x03c3 +#define UVD_GPCOM_VCPU_DATA0 0x03c4 +#define UVD_GPCOM_VCPU_DATA1 0x03c5 +#define UVD_NO_OP 0x03ff +#define UVD_BASE_SI 0x3800 /** * amdgpu_uvd_cs_ctx - Command submission parser context @@ -114,6 +116,7 @@ MODULE_FIRMWARE(FIRMWARE_VEGAM); MODULE_FIRMWARE(FIRMWARE_VEGA10); MODULE_FIRMWARE(FIRMWARE_VEGA12); +MODULE_FIRMWARE(FIRMWARE_VEGA20); static void amdgpu_uvd_idle_work_handler(struct work_struct *work); @@ -125,9 +128,9 @@ int amdgpu_uvd_sw_init(struct amdgpu_device *adev) const char *fw_name; const struct common_firmware_header *hdr; unsigned version_major, version_minor, family_id; - int i, r; + int i, j, r; - INIT_DELAYED_WORK(&adev->uvd.idle_work, amdgpu_uvd_idle_work_handler); + INIT_DELAYED_WORK(&adev->uvd.inst->idle_work, amdgpu_uvd_idle_work_handler); switch (adev->asic_type) { #ifdef CONFIG_DRM_AMDGPU_CIK @@ -177,6 +180,9 @@ int amdgpu_uvd_sw_init(struct amdgpu_device *adev) case CHIP_VEGAM: fw_name = FIRMWARE_VEGAM; break; + case CHIP_VEGA20: + fw_name = FIRMWARE_VEGA20; + break; default: return -EINVAL; } @@ -231,28 +237,30 @@ int amdgpu_uvd_sw_init(struct amdgpu_device *adev) if (adev->firmware.load_type != AMDGPU_FW_LOAD_PSP) bo_size += AMDGPU_GPU_PAGE_ALIGN(le32_to_cpu(hdr->ucode_size_bytes) + 8); - r = amdgpu_bo_create_kernel(adev, bo_size, PAGE_SIZE, - AMDGPU_GEM_DOMAIN_VRAM, &adev->uvd.vcpu_bo, - &adev->uvd.gpu_addr, &adev->uvd.cpu_addr); - if (r) { - dev_err(adev->dev, "(%d) failed to allocate UVD bo\n", r); - return r; - } + for (j = 0; j < adev->uvd.num_uvd_inst; j++) { - ring = &adev->uvd.ring; - rq = &ring->sched.sched_rq[DRM_SCHED_PRIORITY_NORMAL]; - r = drm_sched_entity_init(&ring->sched, &adev->uvd.entity, - rq, NULL); - if (r != 0) { - DRM_ERROR("Failed setting up UVD run queue.\n"); - return r; - } + r = amdgpu_bo_create_kernel(adev, bo_size, PAGE_SIZE, + AMDGPU_GEM_DOMAIN_VRAM, &adev->uvd.inst[j].vcpu_bo, + &adev->uvd.inst[j].gpu_addr, &adev->uvd.inst[j].cpu_addr); + if (r) { + dev_err(adev->dev, "(%d) failed to allocate UVD bo\n", r); + return r; + } - for (i = 0; i < adev->uvd.max_handles; ++i) { - atomic_set(&adev->uvd.handles[i], 0); - adev->uvd.filp[i] = NULL; - } + ring = &adev->uvd.inst[j].ring; + rq = &ring->sched.sched_rq[DRM_SCHED_PRIORITY_NORMAL]; + r = drm_sched_entity_init(&ring->sched, &adev->uvd.inst[j].entity, + rq, NULL); + if (r != 0) { + DRM_ERROR("Failed setting up UVD(%d) run queue.\n", j); + return r; + } + for (i = 0; i < adev->uvd.max_handles; ++i) { + atomic_set(&adev->uvd.inst[j].handles[i], 0); + adev->uvd.inst[j].filp[i] = NULL; + } + } /* from uvd v5.0 HW addressing capacity increased to 64 bits */ if (!amdgpu_device_ip_block_version_cmp(adev, AMD_IP_BLOCK_TYPE_UVD, 5, 0)) adev->uvd.address_64_bit = true; @@ -279,20 +287,22 @@ int amdgpu_uvd_sw_init(struct amdgpu_device *adev) int amdgpu_uvd_sw_fini(struct amdgpu_device *adev) { - int i; - kfree(adev->uvd.saved_bo); + int i, j; - drm_sched_entity_fini(&adev->uvd.ring.sched, &adev->uvd.entity); + for (j = 0; j < adev->uvd.num_uvd_inst; ++j) { + kfree(adev->uvd.inst[j].saved_bo); - amdgpu_bo_free_kernel(&adev->uvd.vcpu_bo, - &adev->uvd.gpu_addr, - (void **)&adev->uvd.cpu_addr); + drm_sched_entity_fini(&adev->uvd.inst[j].ring.sched, &adev->uvd.inst[j].entity); - amdgpu_ring_fini(&adev->uvd.ring); + amdgpu_bo_free_kernel(&adev->uvd.inst[j].vcpu_bo, + &adev->uvd.inst[j].gpu_addr, + (void **)&adev->uvd.inst[j].cpu_addr); - for (i = 0; i < AMDGPU_MAX_UVD_ENC_RINGS; ++i) - amdgpu_ring_fini(&adev->uvd.ring_enc[i]); + amdgpu_ring_fini(&adev->uvd.inst[j].ring); + for (i = 0; i < AMDGPU_MAX_UVD_ENC_RINGS; ++i) + amdgpu_ring_fini(&adev->uvd.inst[j].ring_enc[i]); + } release_firmware(adev->uvd.fw); return 0; @@ -302,32 +312,33 @@ int amdgpu_uvd_suspend(struct amdgpu_device *adev) { unsigned size; void *ptr; - int i; - - if (adev->uvd.vcpu_bo == NULL) - return 0; + int i, j; - cancel_delayed_work_sync(&adev->uvd.idle_work); + for (j = 0; j < adev->uvd.num_uvd_inst; ++j) { + if (adev->uvd.inst[j].vcpu_bo == NULL) + continue; - /* only valid for physical mode */ - if (adev->asic_type < CHIP_POLARIS10) { - for (i = 0; i < adev->uvd.max_handles; ++i) - if (atomic_read(&adev->uvd.handles[i])) - break; + cancel_delayed_work_sync(&adev->uvd.inst[j].idle_work); - if (i == adev->uvd.max_handles) - return 0; - } + /* only valid for physical mode */ + if (adev->asic_type < CHIP_POLARIS10) { + for (i = 0; i < adev->uvd.max_handles; ++i) + if (atomic_read(&adev->uvd.inst[j].handles[i])) + break; - size = amdgpu_bo_size(adev->uvd.vcpu_bo); - ptr = adev->uvd.cpu_addr; + if (i == adev->uvd.max_handles) + continue; + } - adev->uvd.saved_bo = kmalloc(size, GFP_KERNEL); - if (!adev->uvd.saved_bo) - return -ENOMEM; + size = amdgpu_bo_size(adev->uvd.inst[j].vcpu_bo); + ptr = adev->uvd.inst[j].cpu_addr; - memcpy_fromio(adev->uvd.saved_bo, ptr, size); + adev->uvd.inst[j].saved_bo = kmalloc(size, GFP_KERNEL); + if (!adev->uvd.inst[j].saved_bo) + return -ENOMEM; + memcpy_fromio(adev->uvd.inst[j].saved_bo, ptr, size); + } return 0; } @@ -335,59 +346,65 @@ int amdgpu_uvd_resume(struct amdgpu_device *adev) { unsigned size; void *ptr; + int i; - if (adev->uvd.vcpu_bo == NULL) - return -EINVAL; + for (i = 0; i < adev->uvd.num_uvd_inst; i++) { + if (adev->uvd.inst[i].vcpu_bo == NULL) + return -EINVAL; - size = amdgpu_bo_size(adev->uvd.vcpu_bo); - ptr = adev->uvd.cpu_addr; + size = amdgpu_bo_size(adev->uvd.inst[i].vcpu_bo); + ptr = adev->uvd.inst[i].cpu_addr; - if (adev->uvd.saved_bo != NULL) { - memcpy_toio(ptr, adev->uvd.saved_bo, size); - kfree(adev->uvd.saved_bo); - adev->uvd.saved_bo = NULL; - } else { - const struct common_firmware_header *hdr; - unsigned offset; - - hdr = (const struct common_firmware_header *)adev->uvd.fw->data; - if (adev->firmware.load_type != AMDGPU_FW_LOAD_PSP) { - offset = le32_to_cpu(hdr->ucode_array_offset_bytes); - memcpy_toio(adev->uvd.cpu_addr, adev->uvd.fw->data + offset, - le32_to_cpu(hdr->ucode_size_bytes)); - size -= le32_to_cpu(hdr->ucode_size_bytes); - ptr += le32_to_cpu(hdr->ucode_size_bytes); + if (adev->uvd.inst[i].saved_bo != NULL) { + memcpy_toio(ptr, adev->uvd.inst[i].saved_bo, size); + kfree(adev->uvd.inst[i].saved_bo); + adev->uvd.inst[i].saved_bo = NULL; + } else { + const struct common_firmware_header *hdr; + unsigned offset; + + hdr = (const struct common_firmware_header *)adev->uvd.fw->data; + if (adev->firmware.load_type != AMDGPU_FW_LOAD_PSP) { + offset = le32_to_cpu(hdr->ucode_array_offset_bytes); + memcpy_toio(adev->uvd.inst[i].cpu_addr, adev->uvd.fw->data + offset, + le32_to_cpu(hdr->ucode_size_bytes)); + size -= le32_to_cpu(hdr->ucode_size_bytes); + ptr += le32_to_cpu(hdr->ucode_size_bytes); + } + memset_io(ptr, 0, size); + /* to restore uvd fence seq */ + amdgpu_fence_driver_force_completion(&adev->uvd.inst[i].ring); } - memset_io(ptr, 0, size); - /* to restore uvd fence seq */ - amdgpu_fence_driver_force_completion(&adev->uvd.ring); } - return 0; } void amdgpu_uvd_free_handles(struct amdgpu_device *adev, struct drm_file *filp) { - struct amdgpu_ring *ring = &adev->uvd.ring; - int i, r; + struct amdgpu_ring *ring; + int i, j, r; - for (i = 0; i < adev->uvd.max_handles; ++i) { - uint32_t handle = atomic_read(&adev->uvd.handles[i]); - if (handle != 0 && adev->uvd.filp[i] == filp) { - struct dma_fence *fence; - - r = amdgpu_uvd_get_destroy_msg(ring, handle, - false, &fence); - if (r) { - DRM_ERROR("Error destroying UVD (%d)!\n", r); - continue; - } + for (j = 0; j < adev->uvd.num_uvd_inst; j++) { + ring = &adev->uvd.inst[j].ring; + + for (i = 0; i < adev->uvd.max_handles; ++i) { + uint32_t handle = atomic_read(&adev->uvd.inst[j].handles[i]); + if (handle != 0 && adev->uvd.inst[j].filp[i] == filp) { + struct dma_fence *fence; + + r = amdgpu_uvd_get_destroy_msg(ring, handle, + false, &fence); + if (r) { + DRM_ERROR("Error destroying UVD(%d) %d!\n", j, r); + continue; + } - dma_fence_wait(fence, false); - dma_fence_put(fence); + dma_fence_wait(fence, false); + dma_fence_put(fence); - adev->uvd.filp[i] = NULL; - atomic_set(&adev->uvd.handles[i], 0); + adev->uvd.inst[j].filp[i] = NULL; + atomic_set(&adev->uvd.inst[j].handles[i], 0); + } } } } @@ -662,15 +679,16 @@ static int amdgpu_uvd_cs_msg(struct amdgpu_uvd_cs_ctx *ctx, void *ptr; long r; int i; + uint32_t ip_instance = ctx->parser->job->ring->me; if (offset & 0x3F) { - DRM_ERROR("UVD messages must be 64 byte aligned!\n"); + DRM_ERROR("UVD(%d) messages must be 64 byte aligned!\n", ip_instance); return -EINVAL; } r = amdgpu_bo_kmap(bo, &ptr); if (r) { - DRM_ERROR("Failed mapping the UVD message (%ld)!\n", r); + DRM_ERROR("Failed mapping the UVD(%d) message (%ld)!\n", ip_instance, r); return r; } @@ -680,7 +698,7 @@ static int amdgpu_uvd_cs_msg(struct amdgpu_uvd_cs_ctx *ctx, handle = msg[2]; if (handle == 0) { - DRM_ERROR("Invalid UVD handle!\n"); + DRM_ERROR("Invalid UVD(%d) handle!\n", ip_instance); return -EINVAL; } @@ -691,18 +709,18 @@ static int amdgpu_uvd_cs_msg(struct amdgpu_uvd_cs_ctx *ctx, /* try to alloc a new handle */ for (i = 0; i < adev->uvd.max_handles; ++i) { - if (atomic_read(&adev->uvd.handles[i]) == handle) { - DRM_ERROR("Handle 0x%x already in use!\n", handle); + if (atomic_read(&adev->uvd.inst[ip_instance].handles[i]) == handle) { + DRM_ERROR("(%d)Handle 0x%x already in use!\n", ip_instance, handle); return -EINVAL; } - if (!atomic_cmpxchg(&adev->uvd.handles[i], 0, handle)) { - adev->uvd.filp[i] = ctx->parser->filp; + if (!atomic_cmpxchg(&adev->uvd.inst[ip_instance].handles[i], 0, handle)) { + adev->uvd.inst[ip_instance].filp[i] = ctx->parser->filp; return 0; } } - DRM_ERROR("No more free UVD handles!\n"); + DRM_ERROR("No more free UVD(%d) handles!\n", ip_instance); return -ENOSPC; case 1: @@ -714,27 +732,27 @@ static int amdgpu_uvd_cs_msg(struct amdgpu_uvd_cs_ctx *ctx, /* validate the handle */ for (i = 0; i < adev->uvd.max_handles; ++i) { - if (atomic_read(&adev->uvd.handles[i]) == handle) { - if (adev->uvd.filp[i] != ctx->parser->filp) { - DRM_ERROR("UVD handle collision detected!\n"); + if (atomic_read(&adev->uvd.inst[ip_instance].handles[i]) == handle) { + if (adev->uvd.inst[ip_instance].filp[i] != ctx->parser->filp) { + DRM_ERROR("UVD(%d) handle collision detected!\n", ip_instance); return -EINVAL; } return 0; } } - DRM_ERROR("Invalid UVD handle 0x%x!\n", handle); + DRM_ERROR("Invalid UVD(%d) handle 0x%x!\n", ip_instance, handle); return -ENOENT; case 2: /* it's a destroy msg, free the handle */ for (i = 0; i < adev->uvd.max_handles; ++i) - atomic_cmpxchg(&adev->uvd.handles[i], handle, 0); + atomic_cmpxchg(&adev->uvd.inst[ip_instance].handles[i], handle, 0); amdgpu_bo_kunmap(bo); return 0; default: - DRM_ERROR("Illegal UVD message type (%d)!\n", msg_type); + DRM_ERROR("Illegal UVD(%d) message type (%d)!\n", ip_instance, msg_type); return -EINVAL; } BUG(); @@ -805,7 +823,7 @@ static int amdgpu_uvd_cs_pass2(struct amdgpu_uvd_cs_ctx *ctx) } if ((cmd == 0 || cmd == 0x3) && - (start >> 28) != (ctx->parser->adev->uvd.gpu_addr >> 28)) { + (start >> 28) != (ctx->parser->adev->uvd.inst->gpu_addr >> 28)) { DRM_ERROR("msg/fb buffer %LX-%LX out of 256MB segment!\n", start, end); return -EINVAL; @@ -973,6 +991,8 @@ static int amdgpu_uvd_send_msg(struct amdgpu_ring *ring, struct amdgpu_bo *bo, uint64_t addr; long r; int i; + unsigned offset_idx = 0; + unsigned offset[3] = { UVD_BASE_SI, 0, 0 }; amdgpu_bo_kunmap(bo); amdgpu_bo_unpin(bo); @@ -992,17 +1012,16 @@ static int amdgpu_uvd_send_msg(struct amdgpu_ring *ring, struct amdgpu_bo *bo, goto err; if (adev->asic_type >= CHIP_VEGA10) { - data[0] = PACKET0(mmUVD_GPCOM_VCPU_DATA0_VEGA10, 0); - data[1] = PACKET0(mmUVD_GPCOM_VCPU_DATA1_VEGA10, 0); - data[2] = PACKET0(mmUVD_GPCOM_VCPU_CMD_VEGA10, 0); - data[3] = PACKET0(mmUVD_NO_OP_VEGA10, 0); - } else { - data[0] = PACKET0(mmUVD_GPCOM_VCPU_DATA0, 0); - data[1] = PACKET0(mmUVD_GPCOM_VCPU_DATA1, 0); - data[2] = PACKET0(mmUVD_GPCOM_VCPU_CMD, 0); - data[3] = PACKET0(mmUVD_NO_OP, 0); + offset_idx = 1 + ring->me; + offset[1] = adev->reg_offset[UVD_HWIP][0][1]; + offset[2] = adev->reg_offset[UVD_HWIP][1][1]; } + data[0] = PACKET0(offset[offset_idx] + UVD_GPCOM_VCPU_DATA0, 0); + data[1] = PACKET0(offset[offset_idx] + UVD_GPCOM_VCPU_DATA1, 0); + data[2] = PACKET0(offset[offset_idx] + UVD_GPCOM_VCPU_CMD, 0); + data[3] = PACKET0(offset[offset_idx] + UVD_NO_OP, 0); + ib = &job->ibs[0]; addr = amdgpu_bo_gpu_offset(bo); ib->ptr[0] = data[0]; @@ -1038,7 +1057,7 @@ static int amdgpu_uvd_send_msg(struct amdgpu_ring *ring, struct amdgpu_bo *bo, if (r) goto err_free; - r = amdgpu_job_submit(job, ring, &adev->uvd.entity, + r = amdgpu_job_submit(job, ring, &adev->uvd.inst[ring->me].entity, AMDGPU_FENCE_OWNER_UNDEFINED, &f); if (r) goto err_free; @@ -1126,8 +1145,15 @@ int amdgpu_uvd_get_destroy_msg(struct amdgpu_ring *ring, uint32_t handle, static void amdgpu_uvd_idle_work_handler(struct work_struct *work) { struct amdgpu_device *adev = - container_of(work, struct amdgpu_device, uvd.idle_work.work); - unsigned fences = amdgpu_fence_count_emitted(&adev->uvd.ring); + container_of(work, struct amdgpu_device, uvd.inst->idle_work.work); + unsigned fences = 0, i, j; + + for (i = 0; i < adev->uvd.num_uvd_inst; ++i) { + fences += amdgpu_fence_count_emitted(&adev->uvd.inst[i].ring); + for (j = 0; j < adev->uvd.num_enc_rings; ++j) { + fences += amdgpu_fence_count_emitted(&adev->uvd.inst[i].ring_enc[j]); + } + } if (fences == 0) { if (adev->pm.dpm_enabled) { @@ -1141,7 +1167,7 @@ static void amdgpu_uvd_idle_work_handler(struct work_struct *work) AMD_CG_STATE_GATE); } } else { - schedule_delayed_work(&adev->uvd.idle_work, UVD_IDLE_TIMEOUT); + schedule_delayed_work(&adev->uvd.inst->idle_work, UVD_IDLE_TIMEOUT); } } @@ -1153,7 +1179,7 @@ void amdgpu_uvd_ring_begin_use(struct amdgpu_ring *ring) if (amdgpu_sriov_vf(adev)) return; - set_clocks = !cancel_delayed_work_sync(&adev->uvd.idle_work); + set_clocks = !cancel_delayed_work_sync(&adev->uvd.inst->idle_work); if (set_clocks) { if (adev->pm.dpm_enabled) { amdgpu_dpm_enable_uvd(adev, true); @@ -1170,7 +1196,7 @@ void amdgpu_uvd_ring_begin_use(struct amdgpu_ring *ring) void amdgpu_uvd_ring_end_use(struct amdgpu_ring *ring) { if (!amdgpu_sriov_vf(ring->adev)) - schedule_delayed_work(&ring->adev->uvd.idle_work, UVD_IDLE_TIMEOUT); + schedule_delayed_work(&ring->adev->uvd.inst->idle_work, UVD_IDLE_TIMEOUT); } /** @@ -1184,27 +1210,28 @@ int amdgpu_uvd_ring_test_ib(struct amdgpu_ring *ring, long timeout) { struct dma_fence *fence; long r; + uint32_t ip_instance = ring->me; r = amdgpu_uvd_get_create_msg(ring, 1, NULL); if (r) { - DRM_ERROR("amdgpu: failed to get create msg (%ld).\n", r); + DRM_ERROR("amdgpu: (%d)failed to get create msg (%ld).\n", ip_instance, r); goto error; } r = amdgpu_uvd_get_destroy_msg(ring, 1, true, &fence); if (r) { - DRM_ERROR("amdgpu: failed to get destroy ib (%ld).\n", r); + DRM_ERROR("amdgpu: (%d)failed to get destroy ib (%ld).\n", ip_instance, r); goto error; } r = dma_fence_wait_timeout(fence, false, timeout); if (r == 0) { - DRM_ERROR("amdgpu: IB test timed out.\n"); + DRM_ERROR("amdgpu: (%d)IB test timed out.\n", ip_instance); r = -ETIMEDOUT; } else if (r < 0) { - DRM_ERROR("amdgpu: fence wait failed (%ld).\n", r); + DRM_ERROR("amdgpu: (%d)fence wait failed (%ld).\n", ip_instance, r); } else { - DRM_DEBUG("ib test on ring %d succeeded\n", ring->idx); + DRM_DEBUG("ib test on (%d)ring %d succeeded\n", ip_instance, ring->idx); r = 0; } @@ -1232,7 +1259,7 @@ uint32_t amdgpu_uvd_used_handles(struct amdgpu_device *adev) * necessarily linear. So we need to count * all non-zero handles. */ - if (atomic_read(&adev->uvd.handles[i])) + if (atomic_read(&adev->uvd.inst->handles[i])) used_handles++; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.h index 32ea20b99e53..b1579fba134c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.h @@ -31,30 +31,37 @@ #define AMDGPU_UVD_SESSION_SIZE (50*1024) #define AMDGPU_UVD_FIRMWARE_OFFSET 256 +#define AMDGPU_MAX_UVD_INSTANCES 2 + #define AMDGPU_UVD_FIRMWARE_SIZE(adev) \ (AMDGPU_GPU_PAGE_ALIGN(le32_to_cpu(((const struct common_firmware_header *)(adev)->uvd.fw->data)->ucode_size_bytes) + \ 8) - AMDGPU_UVD_FIRMWARE_OFFSET) -struct amdgpu_uvd { +struct amdgpu_uvd_inst { struct amdgpu_bo *vcpu_bo; void *cpu_addr; uint64_t gpu_addr; - unsigned fw_version; void *saved_bo; - unsigned max_handles; atomic_t handles[AMDGPU_MAX_UVD_HANDLES]; struct drm_file *filp[AMDGPU_MAX_UVD_HANDLES]; struct delayed_work idle_work; - const struct firmware *fw; /* UVD firmware */ struct amdgpu_ring ring; struct amdgpu_ring ring_enc[AMDGPU_MAX_UVD_ENC_RINGS]; struct amdgpu_irq_src irq; - bool address_64_bit; - bool use_ctx_buf; struct drm_sched_entity entity; struct drm_sched_entity entity_enc; uint32_t srbm_soft_reset; +}; + +struct amdgpu_uvd { + const struct firmware *fw; /* UVD firmware */ + unsigned fw_version; + unsigned max_handles; unsigned num_enc_rings; + uint8_t num_uvd_inst; + bool address_64_bit; + bool use_ctx_buf; + struct amdgpu_uvd_inst inst[AMDGPU_MAX_UVD_INSTANCES]; }; int amdgpu_uvd_sw_init(struct amdgpu_device *adev); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c index a86322f5164f..23d960ec1cf2 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c @@ -57,6 +57,7 @@ #define FIRMWARE_VEGA10 "amdgpu/vega10_vce.bin" #define FIRMWARE_VEGA12 "amdgpu/vega12_vce.bin" +#define FIRMWARE_VEGA20 "amdgpu/vega20_vce.bin" #ifdef CONFIG_DRM_AMDGPU_CIK MODULE_FIRMWARE(FIRMWARE_BONAIRE); @@ -76,6 +77,7 @@ MODULE_FIRMWARE(FIRMWARE_VEGAM); MODULE_FIRMWARE(FIRMWARE_VEGA10); MODULE_FIRMWARE(FIRMWARE_VEGA12); +MODULE_FIRMWARE(FIRMWARE_VEGA20); static void amdgpu_vce_idle_work_handler(struct work_struct *work); @@ -143,6 +145,9 @@ int amdgpu_vce_sw_init(struct amdgpu_device *adev, unsigned long size) case CHIP_VEGA12: fw_name = FIRMWARE_VEGA12; break; + case CHIP_VEGA20: + fw_name = FIRMWARE_VEGA20; + break; default: return -EINVAL; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c index e5d234cf804f..8851bcdfc260 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c @@ -205,13 +205,18 @@ static void amdgpu_vcn_idle_work_handler(struct work_struct *work) struct amdgpu_device *adev = container_of(work, struct amdgpu_device, vcn.idle_work.work); unsigned fences = amdgpu_fence_count_emitted(&adev->vcn.ring_dec); + unsigned i; + + for (i = 0; i < adev->vcn.num_enc_rings; ++i) { + fences += amdgpu_fence_count_emitted(&adev->vcn.ring_enc[i]); + } if (fences == 0) { - if (adev->pm.dpm_enabled) { - /* might be used when with pg/cg + if (adev->pm.dpm_enabled) amdgpu_dpm_enable_uvd(adev, false); - */ - } + else + amdgpu_device_ip_set_powergating_state(adev, AMD_IP_BLOCK_TYPE_VCN, + AMD_PG_STATE_GATE); } else { schedule_delayed_work(&adev->vcn.idle_work, VCN_IDLE_TIMEOUT); } @@ -223,9 +228,11 @@ void amdgpu_vcn_ring_begin_use(struct amdgpu_ring *ring) bool set_clocks = !cancel_delayed_work_sync(&adev->vcn.idle_work); if (set_clocks && adev->pm.dpm_enabled) { - /* might be used when with pg/cg - amdgpu_dpm_enable_uvd(adev, true); - */ + if (adev->pm.dpm_enabled) + amdgpu_dpm_enable_uvd(adev, true); + else + amdgpu_device_ip_set_powergating_state(adev, AMD_IP_BLOCK_TYPE_VCN, + AMD_PG_STATE_UNGATE); } } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h index 2fd7db891689..181e6afa9847 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h @@ -45,6 +45,17 @@ #define VCN_ENC_CMD_REG_WRITE 0x0000000b #define VCN_ENC_CMD_REG_WAIT 0x0000000c +enum engine_status_constants { + UVD_PGFSM_STATUS__UVDM_UVDU_PWR_ON = 0x2AAAA0, + UVD_PGFSM_CONFIG__UVDM_UVDU_PWR_ON = 0x00000002, + UVD_STATUS__UVD_BUSY = 0x00000004, + GB_ADDR_CONFIG_DEFAULT = 0x26010011, + UVD_STATUS__IDLE = 0x2, + UVD_STATUS__BUSY = 0x5, + UVD_POWER_STATUS__UVD_POWER_STATUS_TILES_OFF = 0x1, + UVD_STATUS__RBC_BUSY = 0x1, +}; + struct amdgpu_vcn { struct amdgpu_bo *vcpu_bo; void *cpu_addr; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c index 1a8f4e0dd023..ccba88cc8c54 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c @@ -119,9 +119,7 @@ static void amdgpu_vm_bo_base_init(struct amdgpu_vm_bo_base *base, * is currently evicted. add the bo to the evicted list to make sure it * is validated on next vm use to avoid fault. * */ - spin_lock(&vm->status_lock); list_move_tail(&base->vm_status, &vm->evicted); - spin_unlock(&vm->status_lock); } /** @@ -226,24 +224,16 @@ int amdgpu_vm_validate_pt_bos(struct amdgpu_device *adev, struct amdgpu_vm *vm, void *param) { struct ttm_bo_global *glob = adev->mman.bdev.glob; - int r; - - spin_lock(&vm->status_lock); - while (!list_empty(&vm->evicted)) { - struct amdgpu_vm_bo_base *bo_base; - struct amdgpu_bo *bo; + struct amdgpu_vm_bo_base *bo_base, *tmp; + int r = 0; - bo_base = list_first_entry(&vm->evicted, - struct amdgpu_vm_bo_base, - vm_status); - spin_unlock(&vm->status_lock); + list_for_each_entry_safe(bo_base, tmp, &vm->evicted, vm_status) { + struct amdgpu_bo *bo = bo_base->bo; - bo = bo_base->bo; - BUG_ON(!bo); if (bo->parent) { r = validate(param, bo); if (r) - return r; + break; spin_lock(&glob->lru_lock); ttm_bo_move_to_lru_tail(&bo->tbo); @@ -252,22 +242,29 @@ int amdgpu_vm_validate_pt_bos(struct amdgpu_device *adev, struct amdgpu_vm *vm, spin_unlock(&glob->lru_lock); } - if (bo->tbo.type == ttm_bo_type_kernel && - vm->use_cpu_for_update) { - r = amdgpu_bo_kmap(bo, NULL); - if (r) - return r; - } - - spin_lock(&vm->status_lock); - if (bo->tbo.type != ttm_bo_type_kernel) + if (bo->tbo.type != ttm_bo_type_kernel) { + spin_lock(&vm->moved_lock); list_move(&bo_base->vm_status, &vm->moved); - else + spin_unlock(&vm->moved_lock); + } else { list_move(&bo_base->vm_status, &vm->relocated); + } } - spin_unlock(&vm->status_lock); - return 0; + spin_lock(&glob->lru_lock); + list_for_each_entry(bo_base, &vm->idle, vm_status) { + struct amdgpu_bo *bo = bo_base->bo; + + if (!bo->parent) + continue; + + ttm_bo_move_to_lru_tail(&bo->tbo); + if (bo->shadow) + ttm_bo_move_to_lru_tail(&bo->shadow->tbo); + } + spin_unlock(&glob->lru_lock); + + return r; } /** @@ -279,13 +276,7 @@ int amdgpu_vm_validate_pt_bos(struct amdgpu_device *adev, struct amdgpu_vm *vm, */ bool amdgpu_vm_ready(struct amdgpu_vm *vm) { - bool ready; - - spin_lock(&vm->status_lock); - ready = list_empty(&vm->evicted); - spin_unlock(&vm->status_lock); - - return ready; + return list_empty(&vm->evicted); } /** @@ -477,9 +468,7 @@ static int amdgpu_vm_alloc_levels(struct amdgpu_device *adev, pt->parent = amdgpu_bo_ref(parent->base.bo); amdgpu_vm_bo_base_init(&entry->base, vm, pt); - spin_lock(&vm->status_lock); list_move(&entry->base.vm_status, &vm->relocated); - spin_unlock(&vm->status_lock); } if (level < AMDGPU_VM_PTB) { @@ -926,10 +915,8 @@ static void amdgpu_vm_invalidate_level(struct amdgpu_device *adev, if (!entry->base.bo) continue; - spin_lock(&vm->status_lock); - if (list_empty(&entry->base.vm_status)) - list_add(&entry->base.vm_status, &vm->relocated); - spin_unlock(&vm->status_lock); + if (!entry->base.moved) + list_move(&entry->base.vm_status, &vm->relocated); amdgpu_vm_invalidate_level(adev, vm, entry, level + 1); } } @@ -959,6 +946,14 @@ restart: params.adev = adev; if (vm->use_cpu_for_update) { + struct amdgpu_vm_bo_base *bo_base; + + list_for_each_entry(bo_base, &vm->relocated, vm_status) { + r = amdgpu_bo_kmap(bo_base->bo, NULL); + if (unlikely(r)) + return r; + } + r = amdgpu_vm_wait_pd(adev, vm, AMDGPU_FENCE_OWNER_VM); if (unlikely(r)) return r; @@ -974,7 +969,6 @@ restart: params.func = amdgpu_vm_do_set_ptes; } - spin_lock(&vm->status_lock); while (!list_empty(&vm->relocated)) { struct amdgpu_vm_bo_base *bo_base, *parent; struct amdgpu_vm_pt *pt, *entry; @@ -983,14 +977,12 @@ restart: bo_base = list_first_entry(&vm->relocated, struct amdgpu_vm_bo_base, vm_status); - list_del_init(&bo_base->vm_status); - spin_unlock(&vm->status_lock); + bo_base->moved = false; + list_move(&bo_base->vm_status, &vm->idle); bo = bo_base->bo->parent; - if (!bo) { - spin_lock(&vm->status_lock); + if (!bo) continue; - } parent = list_first_entry(&bo->va, struct amdgpu_vm_bo_base, bo_list); @@ -999,12 +991,10 @@ restart: amdgpu_vm_update_pde(¶ms, vm, pt, entry); - spin_lock(&vm->status_lock); if (!vm->use_cpu_for_update && (ndw - params.ib->length_dw) < 32) break; } - spin_unlock(&vm->status_lock); if (vm->use_cpu_for_update) { /* Flush HDP */ @@ -1107,9 +1097,7 @@ static void amdgpu_vm_handle_huge_pages(struct amdgpu_pte_update_params *p, if (entry->huge) { /* Add the entry to the relocated list to update it. */ entry->huge = false; - spin_lock(&p->vm->status_lock); list_move(&entry->base.vm_status, &p->vm->relocated); - spin_unlock(&p->vm->status_lock); } return; } @@ -1588,18 +1576,22 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev, amdgpu_asic_flush_hdp(adev, NULL); } - spin_lock(&vm->status_lock); + spin_lock(&vm->moved_lock); list_del_init(&bo_va->base.vm_status); + spin_unlock(&vm->moved_lock); /* If the BO is not in its preferred location add it back to * the evicted list so that it gets validated again on the * next command submission. */ - if (bo && bo->tbo.resv == vm->root.base.bo->tbo.resv && - !(bo->preferred_domains & - amdgpu_mem_type_to_domain(bo->tbo.mem.mem_type))) - list_add_tail(&bo_va->base.vm_status, &vm->evicted); - spin_unlock(&vm->status_lock); + if (bo && bo->tbo.resv == vm->root.base.bo->tbo.resv) { + uint32_t mem_type = bo->tbo.mem.mem_type; + + if (!(bo->preferred_domains & amdgpu_mem_type_to_domain(mem_type))) + list_add_tail(&bo_va->base.vm_status, &vm->evicted); + else + list_add(&bo_va->base.vm_status, &vm->idle); + } list_splice_init(&bo_va->invalids, &bo_va->valids); bo_va->cleared = clear; @@ -1808,19 +1800,18 @@ int amdgpu_vm_clear_freed(struct amdgpu_device *adev, int amdgpu_vm_handle_moved(struct amdgpu_device *adev, struct amdgpu_vm *vm) { + struct amdgpu_bo_va *bo_va, *tmp; + struct list_head moved; bool clear; - int r = 0; - - spin_lock(&vm->status_lock); - while (!list_empty(&vm->moved)) { - struct amdgpu_bo_va *bo_va; - struct reservation_object *resv; + int r; - bo_va = list_first_entry(&vm->moved, - struct amdgpu_bo_va, base.vm_status); - spin_unlock(&vm->status_lock); + INIT_LIST_HEAD(&moved); + spin_lock(&vm->moved_lock); + list_splice_init(&vm->moved, &moved); + spin_unlock(&vm->moved_lock); - resv = bo_va->base.bo->tbo.resv; + list_for_each_entry_safe(bo_va, tmp, &moved, base.vm_status) { + struct reservation_object *resv = bo_va->base.bo->tbo.resv; /* Per VM BOs never need to bo cleared in the page tables */ if (resv == vm->root.base.bo->tbo.resv) @@ -1833,17 +1824,19 @@ int amdgpu_vm_handle_moved(struct amdgpu_device *adev, clear = true; r = amdgpu_vm_bo_update(adev, bo_va, clear); - if (r) + if (r) { + spin_lock(&vm->moved_lock); + list_splice(&moved, &vm->moved); + spin_unlock(&vm->moved_lock); return r; + } if (!clear && resv != vm->root.base.bo->tbo.resv) reservation_object_unlock(resv); - spin_lock(&vm->status_lock); } - spin_unlock(&vm->status_lock); - return r; + return 0; } /** @@ -1902,11 +1895,11 @@ static void amdgpu_vm_bo_insert_map(struct amdgpu_device *adev, if (mapping->flags & AMDGPU_PTE_PRT) amdgpu_vm_prt_get(adev); - if (bo && bo->tbo.resv == vm->root.base.bo->tbo.resv) { - spin_lock(&vm->status_lock); - if (list_empty(&bo_va->base.vm_status)) - list_add(&bo_va->base.vm_status, &vm->moved); - spin_unlock(&vm->status_lock); + if (bo && bo->tbo.resv == vm->root.base.bo->tbo.resv && + !bo_va->base.moved) { + spin_lock(&vm->moved_lock); + list_move(&bo_va->base.vm_status, &vm->moved); + spin_unlock(&vm->moved_lock); } trace_amdgpu_vm_bo_map(bo_va, mapping); } @@ -2216,9 +2209,9 @@ void amdgpu_vm_bo_rmv(struct amdgpu_device *adev, list_del(&bo_va->base.bo_list); - spin_lock(&vm->status_lock); + spin_lock(&vm->moved_lock); list_del(&bo_va->base.vm_status); - spin_unlock(&vm->status_lock); + spin_unlock(&vm->moved_lock); list_for_each_entry_safe(mapping, next, &bo_va->valids, list) { list_del(&mapping->list); @@ -2258,31 +2251,28 @@ void amdgpu_vm_bo_invalidate(struct amdgpu_device *adev, list_for_each_entry(bo_base, &bo->va, bo_list) { struct amdgpu_vm *vm = bo_base->vm; + bool was_moved = bo_base->moved; bo_base->moved = true; if (evicted && bo->tbo.resv == vm->root.base.bo->tbo.resv) { - spin_lock(&bo_base->vm->status_lock); if (bo->tbo.type == ttm_bo_type_kernel) list_move(&bo_base->vm_status, &vm->evicted); else list_move_tail(&bo_base->vm_status, &vm->evicted); - spin_unlock(&bo_base->vm->status_lock); continue; } - if (bo->tbo.type == ttm_bo_type_kernel) { - spin_lock(&bo_base->vm->status_lock); - if (list_empty(&bo_base->vm_status)) - list_add(&bo_base->vm_status, &vm->relocated); - spin_unlock(&bo_base->vm->status_lock); + if (was_moved) continue; - } - spin_lock(&bo_base->vm->status_lock); - if (list_empty(&bo_base->vm_status)) - list_add(&bo_base->vm_status, &vm->moved); - spin_unlock(&bo_base->vm->status_lock); + if (bo->tbo.type == ttm_bo_type_kernel) { + list_move(&bo_base->vm_status, &vm->relocated); + } else { + spin_lock(&bo_base->vm->moved_lock); + list_move(&bo_base->vm_status, &vm->moved); + spin_unlock(&bo_base->vm->moved_lock); + } } } @@ -2391,10 +2381,11 @@ int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm, vm->va = RB_ROOT_CACHED; for (i = 0; i < AMDGPU_MAX_VMHUBS; i++) vm->reserved_vmid[i] = NULL; - spin_lock_init(&vm->status_lock); INIT_LIST_HEAD(&vm->evicted); INIT_LIST_HEAD(&vm->relocated); + spin_lock_init(&vm->moved_lock); INIT_LIST_HEAD(&vm->moved); + INIT_LIST_HEAD(&vm->idle); INIT_LIST_HEAD(&vm->freed); /* create scheduler entity for page table updates */ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h index 4cf678684a12..061b99a18cb8 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h @@ -168,9 +168,6 @@ struct amdgpu_vm { /* tree of virtual addresses mapped */ struct rb_root_cached va; - /* protecting invalidated */ - spinlock_t status_lock; - /* BOs who needs a validation */ struct list_head evicted; @@ -179,6 +176,10 @@ struct amdgpu_vm { /* BOs moved, but not yet updated in the PT */ struct list_head moved; + spinlock_t moved_lock; + + /* All BOs of this VM not currently in the state machine */ + struct list_head idle; /* BO mappings freed, but not yet updated in the PT */ struct list_head freed; @@ -187,9 +188,6 @@ struct amdgpu_vm { struct amdgpu_vm_pt root; struct dma_fence *last_update; - /* protecting freed */ - spinlock_t freed_lock; - /* Scheduler entity for page table updates */ struct drm_sched_entity entity; diff --git a/drivers/gpu/drm/amd/amdgpu/dce_virtual.c b/drivers/gpu/drm/amd/amdgpu/dce_virtual.c index de7be3de0f41..dbf2ccd0c744 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_virtual.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_virtual.c @@ -473,6 +473,7 @@ static int dce_virtual_hw_init(void *handle) break; case CHIP_VEGA10: case CHIP_VEGA12: + case CHIP_VEGA20: break; default: DRM_ERROR("Virtual display unsupported ASIC type: 0x%X\n", adev->asic_type); diff --git a/drivers/gpu/drm/amd/amdgpu/df_v1_7.c b/drivers/gpu/drm/amd/amdgpu/df_v1_7.c index 4ffda996660f..9935371db7ce 100644 --- a/drivers/gpu/drm/amd/amdgpu/df_v1_7.c +++ b/drivers/gpu/drm/amd/amdgpu/df_v1_7.c @@ -102,6 +102,13 @@ static void df_v1_7_get_clockgating_state(struct amdgpu_device *adev, *flags |= AMD_CG_SUPPORT_DF_MGCG; } +static void df_v1_7_enable_ecc_force_par_wr_rmw(struct amdgpu_device *adev, + bool enable) +{ + WREG32_FIELD15(DF, 0, DF_CS_AON0_CoherentSlaveModeCtrlA0, + ForceParWrRMW, enable); +} + const struct amdgpu_df_funcs df_v1_7_funcs = { .init = df_v1_7_init, .enable_broadcast_mode = df_v1_7_enable_broadcast_mode, @@ -109,4 +116,5 @@ const struct amdgpu_df_funcs df_v1_7_funcs = { .get_hbm_channel_number = df_v1_7_get_hbm_channel_number, .update_medium_grain_clock_gating = df_v1_7_update_medium_grain_clock_gating, .get_clockgating_state = df_v1_7_get_clockgating_state, + .enable_ecc_force_par_wr_rmw = df_v1_7_enable_ecc_force_par_wr_rmw, }; diff --git a/drivers/gpu/drm/amd/amdgpu/df_v3_6.c b/drivers/gpu/drm/amd/amdgpu/df_v3_6.c new file mode 100644 index 000000000000..60608b3df881 --- /dev/null +++ b/drivers/gpu/drm/amd/amdgpu/df_v3_6.c @@ -0,0 +1,116 @@ +/* + * Copyright 2018 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + */ +#include "amdgpu.h" +#include "df_v3_6.h" + +#include "df/df_3_6_default.h" +#include "df/df_3_6_offset.h" +#include "df/df_3_6_sh_mask.h" + +static u32 df_v3_6_channel_number[] = {1, 2, 0, 4, 0, 8, 0, + 16, 32, 0, 0, 0, 2, 4, 8}; + +static void df_v3_6_init(struct amdgpu_device *adev) +{ +} + +static void df_v3_6_enable_broadcast_mode(struct amdgpu_device *adev, + bool enable) +{ + u32 tmp; + + if (enable) { + tmp = RREG32_SOC15(DF, 0, mmFabricConfigAccessControl); + tmp &= ~FabricConfigAccessControl__CfgRegInstAccEn_MASK; + WREG32_SOC15(DF, 0, mmFabricConfigAccessControl, tmp); + } else + WREG32_SOC15(DF, 0, mmFabricConfigAccessControl, + mmFabricConfigAccessControl_DEFAULT); +} + +static u32 df_v3_6_get_fb_channel_number(struct amdgpu_device *adev) +{ + u32 tmp; + + tmp = RREG32_SOC15(DF, 0, mmDF_CS_UMC_AON0_DramBaseAddress0); + tmp &= DF_CS_UMC_AON0_DramBaseAddress0__IntLvNumChan_MASK; + tmp >>= DF_CS_UMC_AON0_DramBaseAddress0__IntLvNumChan__SHIFT; + + return tmp; +} + +static u32 df_v3_6_get_hbm_channel_number(struct amdgpu_device *adev) +{ + int fb_channel_number; + + fb_channel_number = adev->df_funcs->get_fb_channel_number(adev); + if (fb_channel_number > ARRAY_SIZE(df_v3_6_channel_number)) + fb_channel_number = 0; + + return df_v3_6_channel_number[fb_channel_number]; +} + +static void df_v3_6_update_medium_grain_clock_gating(struct amdgpu_device *adev, + bool enable) +{ + u32 tmp; + + /* Put DF on broadcast mode */ + adev->df_funcs->enable_broadcast_mode(adev, true); + + if (enable && (adev->cg_flags & AMD_CG_SUPPORT_DF_MGCG)) { + tmp = RREG32_SOC15(DF, 0, mmDF_PIE_AON0_DfGlobalClkGater); + tmp &= ~DF_PIE_AON0_DfGlobalClkGater__MGCGMode_MASK; + tmp |= DF_V3_6_MGCG_ENABLE_15_CYCLE_DELAY; + WREG32_SOC15(DF, 0, mmDF_PIE_AON0_DfGlobalClkGater, tmp); + } else { + tmp = RREG32_SOC15(DF, 0, mmDF_PIE_AON0_DfGlobalClkGater); + tmp &= ~DF_PIE_AON0_DfGlobalClkGater__MGCGMode_MASK; + tmp |= DF_V3_6_MGCG_DISABLE; + WREG32_SOC15(DF, 0, mmDF_PIE_AON0_DfGlobalClkGater, tmp); + } + + /* Exit broadcast mode */ + adev->df_funcs->enable_broadcast_mode(adev, false); +} + +static void df_v3_6_get_clockgating_state(struct amdgpu_device *adev, + u32 *flags) +{ + u32 tmp; + + /* AMD_CG_SUPPORT_DF_MGCG */ + tmp = RREG32_SOC15(DF, 0, mmDF_PIE_AON0_DfGlobalClkGater); + if (tmp & DF_V3_6_MGCG_ENABLE_15_CYCLE_DELAY) + *flags |= AMD_CG_SUPPORT_DF_MGCG; +} + +const struct amdgpu_df_funcs df_v3_6_funcs = { + .init = df_v3_6_init, + .enable_broadcast_mode = df_v3_6_enable_broadcast_mode, + .get_fb_channel_number = df_v3_6_get_fb_channel_number, + .get_hbm_channel_number = df_v3_6_get_hbm_channel_number, + .update_medium_grain_clock_gating = + df_v3_6_update_medium_grain_clock_gating, + .get_clockgating_state = df_v3_6_get_clockgating_state, +}; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/basegk104.c b/drivers/gpu/drm/amd/amdgpu/df_v3_6.h index 780a1d973634..e79c58e5efcb 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/basegk104.c +++ b/drivers/gpu/drm/amd/amdgpu/df_v3_6.h @@ -1,5 +1,5 @@ /* - * Copyright 2015 Red Hat Inc. + * Copyright 2018 Advanced Micro Devices, Inc. * * Permission is hereby granted, free of charge, to any person obtaining a * copy of this software and associated documentation files (the "Software"), @@ -19,20 +19,22 @@ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR * OTHER DEALINGS IN THE SOFTWARE. * - * Authors: Ben Skeggs <bskeggs@redhat.com> */ -#include "dmacnv50.h" -#include "rootnv50.h" -#include <nvif/class.h> +#ifndef __DF_V3_6_H__ +#define __DF_V3_6_H__ -const struct nv50_disp_dmac_oclass -gk104_disp_base_oclass = { - .base.oclass = GK104_DISP_BASE_CHANNEL_DMA, - .base.minver = 0, - .base.maxver = 0, - .ctor = nv50_disp_base_new, - .func = &gf119_disp_dmac_func, - .mthd = &gf119_disp_base_chan_mthd, - .chid = 1, +#include "soc15_common.h" + +enum DF_V3_6_MGCG { + DF_V3_6_MGCG_DISABLE = 0, + DF_V3_6_MGCG_ENABLE_00_CYCLE_DELAY = 1, + DF_V3_6_MGCG_ENABLE_01_CYCLE_DELAY = 2, + DF_V3_6_MGCG_ENABLE_15_CYCLE_DELAY = 13, + DF_V3_6_MGCG_ENABLE_31_CYCLE_DELAY = 14, + DF_V3_6_MGCG_ENABLE_63_CYCLE_DELAY = 15 }; + +extern const struct amdgpu_df_funcs df_v3_6_funcs; + +#endif diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c index fc1911834ab5..d7530fdfaad5 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c @@ -27,6 +27,7 @@ #include "amdgpu_gfx.h" #include "soc15.h" #include "soc15d.h" +#include "amdgpu_atomfirmware.h" #include "gc/gc_9_0_offset.h" #include "gc/gc_9_0_sh_mask.h" @@ -63,6 +64,13 @@ MODULE_FIRMWARE("amdgpu/vega12_mec.bin"); MODULE_FIRMWARE("amdgpu/vega12_mec2.bin"); MODULE_FIRMWARE("amdgpu/vega12_rlc.bin"); +MODULE_FIRMWARE("amdgpu/vega20_ce.bin"); +MODULE_FIRMWARE("amdgpu/vega20_pfp.bin"); +MODULE_FIRMWARE("amdgpu/vega20_me.bin"); +MODULE_FIRMWARE("amdgpu/vega20_mec.bin"); +MODULE_FIRMWARE("amdgpu/vega20_mec2.bin"); +MODULE_FIRMWARE("amdgpu/vega20_rlc.bin"); + MODULE_FIRMWARE("amdgpu/raven_ce.bin"); MODULE_FIRMWARE("amdgpu/raven_pfp.bin"); MODULE_FIRMWARE("amdgpu/raven_me.bin"); @@ -72,29 +80,22 @@ MODULE_FIRMWARE("amdgpu/raven_rlc.bin"); static const struct soc15_reg_golden golden_settings_gc_9_0[] = { - SOC15_REG_GOLDEN_VALUE(GC, 0, mmCPC_UTCL1_CNTL, 0x08000000, 0x08000080), - SOC15_REG_GOLDEN_VALUE(GC, 0, mmCPF_UTCL1_CNTL, 0x08000000, 0x08000080), - SOC15_REG_GOLDEN_VALUE(GC, 0, mmCPG_UTCL1_CNTL, 0x08000000, 0x08000080), SOC15_REG_GOLDEN_VALUE(GC, 0, mmDB_DEBUG2, 0xf00fffff, 0x00000420), SOC15_REG_GOLDEN_VALUE(GC, 0, mmGB_GPU_ID, 0x0000000f, 0x00000000), - SOC15_REG_GOLDEN_VALUE(GC, 0, mmIA_UTCL1_CNTL, 0x08000000, 0x08000080), SOC15_REG_GOLDEN_VALUE(GC, 0, mmPA_SC_BINNER_EVENT_CNTL_3, 0x00000003, 0x82400024), SOC15_REG_GOLDEN_VALUE(GC, 0, mmPA_SC_ENHANCE, 0x3fffffff, 0x00000001), SOC15_REG_GOLDEN_VALUE(GC, 0, mmPA_SC_LINE_STIPPLE_STATE, 0x0000ff0f, 0x00000000), - SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_GPM_UTCL1_CNTL_0, 0x08000000, 0x08000080), - SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_GPM_UTCL1_CNTL_1, 0x08000000, 0x08000080), - SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_GPM_UTCL1_CNTL_2, 0x08000000, 0x08000080), - SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_PREWALKER_UTCL1_CNTL, 0x08000000, 0x08000080), - SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_UTCL1_CNTL, 0x08000000, 0x08000080), SOC15_REG_GOLDEN_VALUE(GC, 0, mmSH_MEM_CONFIG, 0x00001000, 0x00001000), - SOC15_REG_GOLDEN_VALUE(GC, 0, mmSPI_CONFIG_CNTL_1, 0x0000000f, 0x01000107), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSPI_RESOURCE_RESERVE_CU_0, 0x0007ffff, 0x00000800), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSPI_RESOURCE_RESERVE_CU_1, 0x0007ffff, 0x00000800), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSPI_RESOURCE_RESERVE_EN_CU_0, 0x01ffffff, 0x0000ff87), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSPI_RESOURCE_RESERVE_EN_CU_1, 0x01ffffff, 0x0000ff8f), SOC15_REG_GOLDEN_VALUE(GC, 0, mmSQC_CONFIG, 0x03000000, 0x020a2000), SOC15_REG_GOLDEN_VALUE(GC, 0, mmTA_CNTL_AUX, 0xfffffeef, 0x010b0000), SOC15_REG_GOLDEN_VALUE(GC, 0, mmTCP_CHAN_STEER_HI, 0xffffffff, 0x4a2c0e68), SOC15_REG_GOLDEN_VALUE(GC, 0, mmTCP_CHAN_STEER_LO, 0xffffffff, 0xb5d3f197), SOC15_REG_GOLDEN_VALUE(GC, 0, mmVGT_CACHE_INVALIDATION, 0x3fff3af3, 0x19200000), - SOC15_REG_GOLDEN_VALUE(GC, 0, mmVGT_GS_MAX_WAVE_ID, 0x00000fff, 0x000003ff), - SOC15_REG_GOLDEN_VALUE(GC, 0, mmWD_UTCL1_CNTL, 0x08000000, 0x08000080) + SOC15_REG_GOLDEN_VALUE(GC, 0, mmVGT_GS_MAX_WAVE_ID, 0x00000fff, 0x000003ff) }; static const struct soc15_reg_golden golden_settings_gc_9_0_vg10[] = @@ -108,6 +109,20 @@ static const struct soc15_reg_golden golden_settings_gc_9_0_vg10[] = SOC15_REG_GOLDEN_VALUE(GC, 0, mmTD_CNTL, 0x00001800, 0x00000800) }; +static const struct soc15_reg_golden golden_settings_gc_9_0_vg20[] = +{ + SOC15_REG_GOLDEN_VALUE(GC, 0, mmCB_HW_CONTROL_2, 0x0f000000, 0x0a000000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmCB_HW_CONTROL_3, 0x30000000, 0x10000000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGB_ADDR_CONFIG, 0xf3e777ff, 0x22014042), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGB_ADDR_CONFIG_READ, 0xf3e777ff, 0x22014042), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmDB_DEBUG2, 0x00003e00, 0x00000400), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmPA_SC_ENHANCE_1, 0xff840000, 0x04040000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRMI_UTCL1_CNTL2, 0x00030000, 0x00030000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSPI_CONFIG_CNTL_1, 0xffff010f, 0x01000107), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmTA_CNTL_AUX, 0x000b0000, 0x000b0000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmTD_CNTL, 0x01000000, 0x01000000) +}; + static const struct soc15_reg_golden golden_settings_gc_9_1[] = { SOC15_REG_GOLDEN_VALUE(GC, 0, mmCB_HW_CONTROL, 0xfffdf3cf, 0x00014104), @@ -241,6 +256,14 @@ static void gfx_v9_0_init_golden_registers(struct amdgpu_device *adev) golden_settings_gc_9_2_1_vg12, ARRAY_SIZE(golden_settings_gc_9_2_1_vg12)); break; + case CHIP_VEGA20: + soc15_program_register_sequence(adev, + golden_settings_gc_9_0, + ARRAY_SIZE(golden_settings_gc_9_0)); + soc15_program_register_sequence(adev, + golden_settings_gc_9_0_vg20, + ARRAY_SIZE(golden_settings_gc_9_0_vg20)); + break; case CHIP_RAVEN: soc15_program_register_sequence(adev, golden_settings_gc_9_1, @@ -468,6 +491,9 @@ static int gfx_v9_0_init_microcode(struct amdgpu_device *adev) case CHIP_VEGA12: chip_name = "vega12"; break; + case CHIP_VEGA20: + chip_name = "vega20"; + break; case CHIP_RAVEN: chip_name = "raven"; break; @@ -1088,9 +1114,10 @@ static const struct amdgpu_gfx_funcs gfx_v9_0_gfx_funcs = { .select_me_pipe_q = &gfx_v9_0_select_me_pipe_q }; -static void gfx_v9_0_gpu_early_init(struct amdgpu_device *adev) +static int gfx_v9_0_gpu_early_init(struct amdgpu_device *adev) { u32 gb_addr_config; + int err; adev->gfx.funcs = &gfx_v9_0_gfx_funcs; @@ -1112,6 +1139,20 @@ static void gfx_v9_0_gpu_early_init(struct amdgpu_device *adev) gb_addr_config = VEGA12_GB_ADDR_CONFIG_GOLDEN; DRM_INFO("fix gfx.config for vega12\n"); break; + case CHIP_VEGA20: + adev->gfx.config.max_hw_contexts = 8; + adev->gfx.config.sc_prim_fifo_size_frontend = 0x20; + adev->gfx.config.sc_prim_fifo_size_backend = 0x100; + adev->gfx.config.sc_hiz_tile_fifo_size = 0x30; + adev->gfx.config.sc_earlyz_tile_fifo_size = 0x4C0; + gb_addr_config = RREG32_SOC15(GC, 0, mmGB_ADDR_CONFIG); + gb_addr_config &= ~0xf3e777ff; + gb_addr_config |= 0x22014042; + /* check vbios table if gpu info is not available */ + err = amdgpu_atomfirmware_get_gfx_info(adev); + if (err) + return err; + break; case CHIP_RAVEN: adev->gfx.config.max_hw_contexts = 8; adev->gfx.config.sc_prim_fifo_size_frontend = 0x20; @@ -1161,6 +1202,8 @@ static void gfx_v9_0_gpu_early_init(struct amdgpu_device *adev) adev->gfx.config.gb_addr_config, GB_ADDR_CONFIG, PIPE_INTERLEAVE_SIZE)); + + return 0; } static int gfx_v9_0_ngg_create_buf(struct amdgpu_device *adev, @@ -1394,6 +1437,7 @@ static int gfx_v9_0_sw_init(void *handle) switch (adev->asic_type) { case CHIP_VEGA10: case CHIP_VEGA12: + case CHIP_VEGA20: case CHIP_RAVEN: adev->gfx.mec.num_mec = 2; break; @@ -1521,7 +1565,9 @@ static int gfx_v9_0_sw_init(void *handle) adev->gfx.ce_ram_size = 0x8000; - gfx_v9_0_gpu_early_init(adev); + r = gfx_v9_0_gpu_early_init(adev); + if (r) + return r; r = gfx_v9_0_ngg_init(adev); if (r) @@ -3688,6 +3734,7 @@ static int gfx_v9_0_set_clockgating_state(void *handle, switch (adev->asic_type) { case CHIP_VEGA10: case CHIP_VEGA12: + case CHIP_VEGA20: case CHIP_RAVEN: gfx_v9_0_update_gfx_clock_gating(adev, state == AMD_CG_STATE_GATE ? true : false); @@ -4680,6 +4727,7 @@ static void gfx_v9_0_set_rlc_funcs(struct amdgpu_device *adev) switch (adev->asic_type) { case CHIP_VEGA10: case CHIP_VEGA12: + case CHIP_VEGA20: case CHIP_RAVEN: adev->gfx.rlc.funcs = &gfx_v9_0_rlc_funcs; break; diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c index 6cccf0e0acd7..3c0a85d4e4ab 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c @@ -675,6 +675,7 @@ static int gmc_v9_0_late_init(void *handle) DRM_INFO("ECC is active.\n"); } else if (r == 0) { DRM_INFO("ECC is not present.\n"); + adev->df_funcs->enable_ecc_force_par_wr_rmw(adev, false); } else { DRM_ERROR("gmc_v9_0_ecc_available() failed. r: %d\n", r); return r; @@ -693,10 +694,7 @@ static void gmc_v9_0_vram_gtt_location(struct amdgpu_device *adev, amdgpu_device_vram_location(adev, &adev->gmc, base); amdgpu_device_gart_location(adev, mc); /* base offset of vram pages */ - if (adev->flags & AMD_IS_APU) - adev->vm_manager.vram_base_offset = gfxhub_v1_0_get_mc_fb_offset(adev); - else - adev->vm_manager.vram_base_offset = 0; + adev->vm_manager.vram_base_offset = gfxhub_v1_0_get_mc_fb_offset(adev); } /** @@ -755,6 +753,7 @@ static int gmc_v9_0_mc_init(struct amdgpu_device *adev) switch (adev->asic_type) { case CHIP_VEGA10: /* all engines support GPUVM */ case CHIP_VEGA12: /* all engines support GPUVM */ + case CHIP_VEGA20: default: adev->gmc.gart_size = 512ULL << 20; break; @@ -860,6 +859,7 @@ static int gmc_v9_0_sw_init(void *handle) break; case CHIP_VEGA10: case CHIP_VEGA12: + case CHIP_VEGA20: /* * To fulfill 4-level page support, * vm size is 256TB (48bit), maximum size of Vega10, @@ -977,6 +977,7 @@ static void gmc_v9_0_init_golden_registers(struct amdgpu_device *adev) switch (adev->asic_type) { case CHIP_VEGA10: + case CHIP_VEGA20: soc15_program_register_sequence(adev, golden_settings_mmhub_1_0_0, ARRAY_SIZE(golden_settings_mmhub_1_0_0)); diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c index 43f925773b57..3d53c4413f13 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c @@ -734,6 +734,7 @@ int mmhub_v1_0_set_clockgating(struct amdgpu_device *adev, switch (adev->asic_type) { case CHIP_VEGA10: case CHIP_VEGA12: + case CHIP_VEGA20: case CHIP_RAVEN: mmhub_v1_0_update_medium_grain_clock_gating(adev, state == AMD_CG_STATE_GATE ? true : false); diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_0.c b/drivers/gpu/drm/amd/amdgpu/nbio_v7_0.c index df34dc79d444..365517c0121e 100644 --- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_0.c +++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_0.c @@ -34,10 +34,19 @@ #define smnCPM_CONTROL 0x11180460 #define smnPCIE_CNTL2 0x11180070 +/* vega20 */ +#define mmRCC_DEV0_EPF0_STRAP0_VG20 0x0011 +#define mmRCC_DEV0_EPF0_STRAP0_VG20_BASE_IDX 2 + static u32 nbio_v7_0_get_rev_id(struct amdgpu_device *adev) { u32 tmp = RREG32_SOC15(NBIO, 0, mmRCC_DEV0_EPF0_STRAP0); + if (adev->asic_type == CHIP_VEGA20) + tmp = RREG32_SOC15(NBIO, 0, mmRCC_DEV0_EPF0_STRAP0_VG20); + else + tmp = RREG32_SOC15(NBIO, 0, mmRCC_DEV0_EPF0_STRAP0); + tmp &= RCC_DEV0_EPF0_STRAP0__STRAP_ATI_REV_ID_DEV0_F0_MASK; tmp >>= RCC_DEV0_EPF0_STRAP0__STRAP_ATI_REV_ID_DEV0_F0__SHIFT; @@ -75,10 +84,14 @@ static void nbio_v7_0_sdma_doorbell_range(struct amdgpu_device *adev, int instan SOC15_REG_OFFSET(NBIO, 0, mmBIF_SDMA1_DOORBELL_RANGE); u32 doorbell_range = RREG32(reg); + u32 range = 2; + + if (adev->asic_type == CHIP_VEGA20) + range = 8; if (use_doorbell) { doorbell_range = REG_SET_FIELD(doorbell_range, BIF_SDMA0_DOORBELL_RANGE, OFFSET, doorbell_index); - doorbell_range = REG_SET_FIELD(doorbell_range, BIF_SDMA0_DOORBELL_RANGE, SIZE, 2); + doorbell_range = REG_SET_FIELD(doorbell_range, BIF_SDMA0_DOORBELL_RANGE, SIZE, range); } else doorbell_range = REG_SET_FIELD(doorbell_range, BIF_SDMA0_DOORBELL_RANGE, SIZE, 0); @@ -133,6 +146,9 @@ static void nbio_v7_0_update_medium_grain_clock_gating(struct amdgpu_device *ade { uint32_t def, data; + if (adev->asic_type == CHIP_VEGA20) + return; + /* NBIF_MGCG_CTRL_LCLK */ def = data = RREG32_PCIE(smnNBIF_MGCG_CTRL_LCLK); diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v3_1.c b/drivers/gpu/drm/amd/amdgpu/psp_v3_1.c index 196e75def1f2..0c768e388ace 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v3_1.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v3_1.c @@ -41,6 +41,9 @@ MODULE_FIRMWARE("amdgpu/vega10_sos.bin"); MODULE_FIRMWARE("amdgpu/vega10_asd.bin"); MODULE_FIRMWARE("amdgpu/vega12_sos.bin"); MODULE_FIRMWARE("amdgpu/vega12_asd.bin"); +MODULE_FIRMWARE("amdgpu/vega20_sos.bin"); +MODULE_FIRMWARE("amdgpu/vega20_asd.bin"); + #define smnMP1_FIRMWARE_FLAGS 0x3010028 diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c index 03a36cbe7557..ca53b3fba422 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c @@ -42,6 +42,8 @@ MODULE_FIRMWARE("amdgpu/vega10_sdma.bin"); MODULE_FIRMWARE("amdgpu/vega10_sdma1.bin"); MODULE_FIRMWARE("amdgpu/vega12_sdma.bin"); MODULE_FIRMWARE("amdgpu/vega12_sdma1.bin"); +MODULE_FIRMWARE("amdgpu/vega20_sdma.bin"); +MODULE_FIRMWARE("amdgpu/vega20_sdma1.bin"); MODULE_FIRMWARE("amdgpu/raven_sdma.bin"); #define SDMA0_POWER_CNTL__ON_OFF_CONDITION_HOLD_TIME_MASK 0x000000F8L @@ -107,6 +109,28 @@ static const struct soc15_reg_golden golden_settings_sdma_4_1[] = SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_UTCL1_PAGE, 0x000003ff, 0x000003c0) }; +static const struct soc15_reg_golden golden_settings_sdma_4_2[] = +{ + SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_CHICKEN_BITS, 0xfe931f07, 0x02831d07), + SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_CLK_CTRL, 0xffffffff, 0x3f000100), + SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_GB_ADDR_CONFIG, 0x0000773f, 0x00004002), + SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_GB_ADDR_CONFIG_READ, 0x0000773f, 0x00004002), + SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_GFX_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_PAGE_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_RLC0_RB_WPTR_POLL_CNTL, 0xfffffff0, 0x00403000), + SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_RLC1_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_UTCL1_PAGE, 0x000003ff, 0x000003c0), + SOC15_REG_GOLDEN_VALUE(SDMA1, 0, mmSDMA1_CHICKEN_BITS, 0xfe931f07, 0x02831d07), + SOC15_REG_GOLDEN_VALUE(SDMA1, 0, mmSDMA1_CLK_CTRL, 0xffffffff, 0x3f000100), + SOC15_REG_GOLDEN_VALUE(SDMA1, 0, mmSDMA1_GB_ADDR_CONFIG, 0x0000773f, 0x00004002), + SOC15_REG_GOLDEN_VALUE(SDMA1, 0, mmSDMA1_GB_ADDR_CONFIG_READ, 0x0000773f, 0x00004002), + SOC15_REG_GOLDEN_VALUE(SDMA1, 0, mmSDMA1_GFX_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(SDMA1, 0, mmSDMA1_PAGE_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(SDMA1, 0, mmSDMA1_RLC0_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(SDMA1, 0, mmSDMA1_RLC1_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(SDMA1, 0, mmSDMA1_UTCL1_PAGE, 0x000003ff, 0x000003c0) +}; + static const struct soc15_reg_golden golden_settings_sdma_rv1[] = { SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_GB_ADDR_CONFIG, 0x0018773f, 0x00000002), @@ -139,6 +163,11 @@ static void sdma_v4_0_init_golden_registers(struct amdgpu_device *adev) golden_settings_sdma_vg12, ARRAY_SIZE(golden_settings_sdma_vg12)); break; + case CHIP_VEGA20: + soc15_program_register_sequence(adev, + golden_settings_sdma_4_2, + ARRAY_SIZE(golden_settings_sdma_4_2)); + break; case CHIP_RAVEN: soc15_program_register_sequence(adev, golden_settings_sdma_4_1, @@ -182,6 +211,9 @@ static int sdma_v4_0_init_microcode(struct amdgpu_device *adev) case CHIP_VEGA12: chip_name = "vega12"; break; + case CHIP_VEGA20: + chip_name = "vega20"; + break; case CHIP_RAVEN: chip_name = "raven"; break; @@ -1516,6 +1548,7 @@ static int sdma_v4_0_set_clockgating_state(void *handle, switch (adev->asic_type) { case CHIP_VEGA10: case CHIP_VEGA12: + case CHIP_VEGA20: case CHIP_RAVEN: sdma_v4_0_update_medium_grain_clock_gating(adev, state == AMD_CG_STATE_GATE ? true : false); diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c b/drivers/gpu/drm/amd/amdgpu/soc15.c index 90065766fffb..68b4a22a8892 100644 --- a/drivers/gpu/drm/amd/amdgpu/soc15.c +++ b/drivers/gpu/drm/amd/amdgpu/soc15.c @@ -41,8 +41,6 @@ #include "sdma1/sdma1_4_0_offset.h" #include "hdp/hdp_4_0_offset.h" #include "hdp/hdp_4_0_sh_mask.h" -#include "mp/mp_9_0_offset.h" -#include "mp/mp_9_0_sh_mask.h" #include "smuio/smuio_9_0_offset.h" #include "smuio/smuio_9_0_sh_mask.h" @@ -53,6 +51,7 @@ #include "gfxhub_v1_0.h" #include "mmhub_v1_0.h" #include "df_v1_7.h" +#include "df_v3_6.h" #include "vega10_ih.h" #include "sdma_v4_0.h" #include "uvd_v7_0.h" @@ -489,16 +488,24 @@ int soc15_set_ip_blocks(struct amdgpu_device *adev) case CHIP_RAVEN: vega10_reg_base_init(adev); break; + case CHIP_VEGA20: + vega20_reg_base_init(adev); + break; default: return -EINVAL; } if (adev->flags & AMD_IS_APU) adev->nbio_funcs = &nbio_v7_0_funcs; + else if (adev->asic_type == CHIP_VEGA20) + adev->nbio_funcs = &nbio_v7_0_funcs; else adev->nbio_funcs = &nbio_v6_1_funcs; - adev->df_funcs = &df_v1_7_funcs; + if (adev->asic_type == CHIP_VEGA20) + adev->df_funcs = &df_v3_6_funcs; + else + adev->df_funcs = &df_v1_7_funcs; adev->nbio_funcs->detect_hw_virt(adev); if (amdgpu_sriov_vf(adev)) @@ -507,12 +514,15 @@ int soc15_set_ip_blocks(struct amdgpu_device *adev) switch (adev->asic_type) { case CHIP_VEGA10: case CHIP_VEGA12: + case CHIP_VEGA20: amdgpu_device_ip_block_add(adev, &vega10_common_ip_block); amdgpu_device_ip_block_add(adev, &gmc_v9_0_ip_block); amdgpu_device_ip_block_add(adev, &vega10_ih_ip_block); - amdgpu_device_ip_block_add(adev, &psp_v3_1_ip_block); - if (!amdgpu_sriov_vf(adev)) - amdgpu_device_ip_block_add(adev, &pp_smu_ip_block); + if (adev->asic_type != CHIP_VEGA20) { + amdgpu_device_ip_block_add(adev, &psp_v3_1_ip_block); + if (!amdgpu_sriov_vf(adev)) + amdgpu_device_ip_block_add(adev, &pp_smu_ip_block); + } if (adev->enable_virtual_display || amdgpu_sriov_vf(adev)) amdgpu_device_ip_block_add(adev, &dce_virtual_ip_block); #if defined(CONFIG_DRM_AMD_DC) @@ -660,6 +670,27 @@ static int soc15_common_early_init(void *handle) adev->pg_flags = 0; adev->external_rev_id = adev->rev_id + 0x14; break; + case CHIP_VEGA20: + adev->cg_flags = AMD_CG_SUPPORT_GFX_MGCG | + AMD_CG_SUPPORT_GFX_MGLS | + AMD_CG_SUPPORT_GFX_CGCG | + AMD_CG_SUPPORT_GFX_CGLS | + AMD_CG_SUPPORT_GFX_3D_CGCG | + AMD_CG_SUPPORT_GFX_3D_CGLS | + AMD_CG_SUPPORT_GFX_CP_LS | + AMD_CG_SUPPORT_MC_LS | + AMD_CG_SUPPORT_MC_MGCG | + AMD_CG_SUPPORT_SDMA_MGCG | + AMD_CG_SUPPORT_SDMA_LS | + AMD_CG_SUPPORT_BIF_MGCG | + AMD_CG_SUPPORT_BIF_LS | + AMD_CG_SUPPORT_HDP_MGCG | + AMD_CG_SUPPORT_ROM_MGCG | + AMD_CG_SUPPORT_VCE_MGCG | + AMD_CG_SUPPORT_UVD_MGCG; + adev->pg_flags = 0; + adev->external_rev_id = adev->rev_id + 0x28; + break; case CHIP_RAVEN: adev->cg_flags = AMD_CG_SUPPORT_GFX_MGCG | AMD_CG_SUPPORT_GFX_MGLS | @@ -679,8 +710,10 @@ static int soc15_common_early_init(void *handle) AMD_CG_SUPPORT_MC_MGCG | AMD_CG_SUPPORT_MC_LS | AMD_CG_SUPPORT_SDMA_MGCG | - AMD_CG_SUPPORT_SDMA_LS; - adev->pg_flags = AMD_PG_SUPPORT_SDMA; + AMD_CG_SUPPORT_SDMA_LS | + AMD_CG_SUPPORT_VCN_MGCG; + + adev->pg_flags = AMD_PG_SUPPORT_SDMA | AMD_PG_SUPPORT_VCN; if (adev->powerplay.pp_feature & PP_GFXOFF_MASK) adev->pg_flags |= AMD_PG_SUPPORT_GFX_PG | @@ -872,6 +905,7 @@ static int soc15_common_set_clockgating_state(void *handle, switch (adev->asic_type) { case CHIP_VEGA10: case CHIP_VEGA12: + case CHIP_VEGA20: adev->nbio_funcs->update_medium_grain_clock_gating(adev, state == AMD_CG_STATE_GATE ? true : false); adev->nbio_funcs->update_medium_grain_light_sleep(adev, diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.h b/drivers/gpu/drm/amd/amdgpu/soc15.h index f70da8a29f86..1f714b7af520 100644 --- a/drivers/gpu/drm/amd/amdgpu/soc15.h +++ b/drivers/gpu/drm/amd/amdgpu/soc15.h @@ -55,5 +55,6 @@ void soc15_program_register_sequence(struct amdgpu_device *adev, const u32 array_size); int vega10_reg_base_init(struct amdgpu_device *adev); +int vega20_reg_base_init(struct amdgpu_device *adev); #endif diff --git a/drivers/gpu/drm/amd/amdgpu/soc15_common.h b/drivers/gpu/drm/amd/amdgpu/soc15_common.h index def865067edd..0942f492d2e1 100644 --- a/drivers/gpu/drm/amd/amdgpu/soc15_common.h +++ b/drivers/gpu/drm/amd/amdgpu/soc15_common.h @@ -47,6 +47,21 @@ #define WREG32_SOC15_OFFSET(ip, inst, reg, offset, value) \ WREG32((adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + reg) + offset, value) +#define SOC15_WAIT_ON_RREG(ip, inst, reg, expected_value, mask, ret) \ + do { \ + uint32_t tmp_ = RREG32(adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + reg); \ + uint32_t loop = adev->usec_timeout; \ + while ((tmp_ & (mask)) != (expected_value)) { \ + udelay(2); \ + tmp_ = RREG32(adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + reg); \ + loop--; \ + if (!loop) { \ + ret = -ETIMEDOUT; \ + break; \ + } \ + } \ + } while (0) + #endif diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c b/drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c index 87cbb142dd0b..6fed3d7797a8 100644 --- a/drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c +++ b/drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c @@ -93,6 +93,7 @@ static void uvd_v4_2_ring_set_wptr(struct amdgpu_ring *ring) static int uvd_v4_2_early_init(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; + adev->uvd.num_uvd_inst = 1; uvd_v4_2_set_ring_funcs(adev); uvd_v4_2_set_irq_funcs(adev); @@ -107,7 +108,7 @@ static int uvd_v4_2_sw_init(void *handle) int r; /* UVD TRAP */ - r = amdgpu_irq_add_id(adev, AMDGPU_IH_CLIENTID_LEGACY, 124, &adev->uvd.irq); + r = amdgpu_irq_add_id(adev, AMDGPU_IH_CLIENTID_LEGACY, 124, &adev->uvd.inst->irq); if (r) return r; @@ -119,9 +120,9 @@ static int uvd_v4_2_sw_init(void *handle) if (r) return r; - ring = &adev->uvd.ring; + ring = &adev->uvd.inst->ring; sprintf(ring->name, "uvd"); - r = amdgpu_ring_init(adev, ring, 512, &adev->uvd.irq, 0); + r = amdgpu_ring_init(adev, ring, 512, &adev->uvd.inst->irq, 0); return r; } @@ -150,7 +151,7 @@ static void uvd_v4_2_enable_mgcg(struct amdgpu_device *adev, static int uvd_v4_2_hw_init(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - struct amdgpu_ring *ring = &adev->uvd.ring; + struct amdgpu_ring *ring = &adev->uvd.inst->ring; uint32_t tmp; int r; @@ -208,7 +209,7 @@ done: static int uvd_v4_2_hw_fini(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - struct amdgpu_ring *ring = &adev->uvd.ring; + struct amdgpu_ring *ring = &adev->uvd.inst->ring; if (RREG32(mmUVD_STATUS) != 0) uvd_v4_2_stop(adev); @@ -251,7 +252,7 @@ static int uvd_v4_2_resume(void *handle) */ static int uvd_v4_2_start(struct amdgpu_device *adev) { - struct amdgpu_ring *ring = &adev->uvd.ring; + struct amdgpu_ring *ring = &adev->uvd.inst->ring; uint32_t rb_bufsz; int i, j, r; u32 tmp; @@ -523,6 +524,18 @@ static void uvd_v4_2_ring_emit_ib(struct amdgpu_ring *ring, amdgpu_ring_write(ring, ib->length_dw); } +static void uvd_v4_2_ring_insert_nop(struct amdgpu_ring *ring, uint32_t count) +{ + int i; + + WARN_ON(ring->wptr % 2 || count % 2); + + for (i = 0; i < count / 2; i++) { + amdgpu_ring_write(ring, PACKET0(mmUVD_NO_OP, 0)); + amdgpu_ring_write(ring, 0); + } +} + /** * uvd_v4_2_mc_resume - memory controller programming * @@ -536,7 +549,7 @@ static void uvd_v4_2_mc_resume(struct amdgpu_device *adev) uint32_t size; /* programm the VCPU memory controller bits 0-27 */ - addr = (adev->uvd.gpu_addr + AMDGPU_UVD_FIRMWARE_OFFSET) >> 3; + addr = (adev->uvd.inst->gpu_addr + AMDGPU_UVD_FIRMWARE_OFFSET) >> 3; size = AMDGPU_UVD_FIRMWARE_SIZE(adev) >> 3; WREG32(mmUVD_VCPU_CACHE_OFFSET0, addr); WREG32(mmUVD_VCPU_CACHE_SIZE0, size); @@ -553,11 +566,11 @@ static void uvd_v4_2_mc_resume(struct amdgpu_device *adev) WREG32(mmUVD_VCPU_CACHE_SIZE2, size); /* bits 28-31 */ - addr = (adev->uvd.gpu_addr >> 28) & 0xF; + addr = (adev->uvd.inst->gpu_addr >> 28) & 0xF; WREG32(mmUVD_LMI_ADDR_EXT, (addr << 12) | (addr << 0)); /* bits 32-39 */ - addr = (adev->uvd.gpu_addr >> 32) & 0xFF; + addr = (adev->uvd.inst->gpu_addr >> 32) & 0xFF; WREG32(mmUVD_LMI_EXT40_ADDR, addr | (0x9 << 16) | (0x1 << 31)); WREG32(mmUVD_UDEC_ADDR_CONFIG, adev->gfx.config.gb_addr_config); @@ -664,7 +677,7 @@ static int uvd_v4_2_process_interrupt(struct amdgpu_device *adev, struct amdgpu_iv_entry *entry) { DRM_DEBUG("IH: UVD TRAP\n"); - amdgpu_fence_process(&adev->uvd.ring); + amdgpu_fence_process(&adev->uvd.inst->ring); return 0; } @@ -732,7 +745,6 @@ static const struct amd_ip_funcs uvd_v4_2_ip_funcs = { static const struct amdgpu_ring_funcs uvd_v4_2_ring_funcs = { .type = AMDGPU_RING_TYPE_UVD, .align_mask = 0xf, - .nop = PACKET0(mmUVD_NO_OP, 0), .support_64bit_ptrs = false, .get_rptr = uvd_v4_2_ring_get_rptr, .get_wptr = uvd_v4_2_ring_get_wptr, @@ -745,7 +757,7 @@ static const struct amdgpu_ring_funcs uvd_v4_2_ring_funcs = { .emit_fence = uvd_v4_2_ring_emit_fence, .test_ring = uvd_v4_2_ring_test_ring, .test_ib = amdgpu_uvd_ring_test_ib, - .insert_nop = amdgpu_ring_insert_nop, + .insert_nop = uvd_v4_2_ring_insert_nop, .pad_ib = amdgpu_ring_generic_pad_ib, .begin_use = amdgpu_uvd_ring_begin_use, .end_use = amdgpu_uvd_ring_end_use, @@ -753,7 +765,7 @@ static const struct amdgpu_ring_funcs uvd_v4_2_ring_funcs = { static void uvd_v4_2_set_ring_funcs(struct amdgpu_device *adev) { - adev->uvd.ring.funcs = &uvd_v4_2_ring_funcs; + adev->uvd.inst->ring.funcs = &uvd_v4_2_ring_funcs; } static const struct amdgpu_irq_src_funcs uvd_v4_2_irq_funcs = { @@ -763,8 +775,8 @@ static const struct amdgpu_irq_src_funcs uvd_v4_2_irq_funcs = { static void uvd_v4_2_set_irq_funcs(struct amdgpu_device *adev) { - adev->uvd.irq.num_types = 1; - adev->uvd.irq.funcs = &uvd_v4_2_irq_funcs; + adev->uvd.inst->irq.num_types = 1; + adev->uvd.inst->irq.funcs = &uvd_v4_2_irq_funcs; } const struct amdgpu_ip_block_version uvd_v4_2_ip_block = diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c b/drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c index 6445d55e7d5a..341ee6d55ce8 100644 --- a/drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c +++ b/drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c @@ -89,6 +89,7 @@ static void uvd_v5_0_ring_set_wptr(struct amdgpu_ring *ring) static int uvd_v5_0_early_init(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; + adev->uvd.num_uvd_inst = 1; uvd_v5_0_set_ring_funcs(adev); uvd_v5_0_set_irq_funcs(adev); @@ -103,7 +104,7 @@ static int uvd_v5_0_sw_init(void *handle) int r; /* UVD TRAP */ - r = amdgpu_irq_add_id(adev, AMDGPU_IH_CLIENTID_LEGACY, 124, &adev->uvd.irq); + r = amdgpu_irq_add_id(adev, AMDGPU_IH_CLIENTID_LEGACY, 124, &adev->uvd.inst->irq); if (r) return r; @@ -115,9 +116,9 @@ static int uvd_v5_0_sw_init(void *handle) if (r) return r; - ring = &adev->uvd.ring; + ring = &adev->uvd.inst->ring; sprintf(ring->name, "uvd"); - r = amdgpu_ring_init(adev, ring, 512, &adev->uvd.irq, 0); + r = amdgpu_ring_init(adev, ring, 512, &adev->uvd.inst->irq, 0); return r; } @@ -144,7 +145,7 @@ static int uvd_v5_0_sw_fini(void *handle) static int uvd_v5_0_hw_init(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - struct amdgpu_ring *ring = &adev->uvd.ring; + struct amdgpu_ring *ring = &adev->uvd.inst->ring; uint32_t tmp; int r; @@ -204,7 +205,7 @@ done: static int uvd_v5_0_hw_fini(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - struct amdgpu_ring *ring = &adev->uvd.ring; + struct amdgpu_ring *ring = &adev->uvd.inst->ring; if (RREG32(mmUVD_STATUS) != 0) uvd_v5_0_stop(adev); @@ -253,9 +254,9 @@ static void uvd_v5_0_mc_resume(struct amdgpu_device *adev) /* programm memory controller bits 0-27 */ WREG32(mmUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW, - lower_32_bits(adev->uvd.gpu_addr)); + lower_32_bits(adev->uvd.inst->gpu_addr)); WREG32(mmUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH, - upper_32_bits(adev->uvd.gpu_addr)); + upper_32_bits(adev->uvd.inst->gpu_addr)); offset = AMDGPU_UVD_FIRMWARE_OFFSET; size = AMDGPU_UVD_FIRMWARE_SIZE(adev); @@ -287,7 +288,7 @@ static void uvd_v5_0_mc_resume(struct amdgpu_device *adev) */ static int uvd_v5_0_start(struct amdgpu_device *adev) { - struct amdgpu_ring *ring = &adev->uvd.ring; + struct amdgpu_ring *ring = &adev->uvd.inst->ring; uint32_t rb_bufsz, tmp; uint32_t lmi_swap_cntl; uint32_t mp_swap_cntl; @@ -540,6 +541,18 @@ static void uvd_v5_0_ring_emit_ib(struct amdgpu_ring *ring, amdgpu_ring_write(ring, ib->length_dw); } +static void uvd_v5_0_ring_insert_nop(struct amdgpu_ring *ring, uint32_t count) +{ + int i; + + WARN_ON(ring->wptr % 2 || count % 2); + + for (i = 0; i < count / 2; i++) { + amdgpu_ring_write(ring, PACKET0(mmUVD_NO_OP, 0)); + amdgpu_ring_write(ring, 0); + } +} + static bool uvd_v5_0_is_idle(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; @@ -586,7 +599,7 @@ static int uvd_v5_0_process_interrupt(struct amdgpu_device *adev, struct amdgpu_iv_entry *entry) { DRM_DEBUG("IH: UVD TRAP\n"); - amdgpu_fence_process(&adev->uvd.ring); + amdgpu_fence_process(&adev->uvd.inst->ring); return 0; } @@ -840,7 +853,6 @@ static const struct amd_ip_funcs uvd_v5_0_ip_funcs = { static const struct amdgpu_ring_funcs uvd_v5_0_ring_funcs = { .type = AMDGPU_RING_TYPE_UVD, .align_mask = 0xf, - .nop = PACKET0(mmUVD_NO_OP, 0), .support_64bit_ptrs = false, .get_rptr = uvd_v5_0_ring_get_rptr, .get_wptr = uvd_v5_0_ring_get_wptr, @@ -853,7 +865,7 @@ static const struct amdgpu_ring_funcs uvd_v5_0_ring_funcs = { .emit_fence = uvd_v5_0_ring_emit_fence, .test_ring = uvd_v5_0_ring_test_ring, .test_ib = amdgpu_uvd_ring_test_ib, - .insert_nop = amdgpu_ring_insert_nop, + .insert_nop = uvd_v5_0_ring_insert_nop, .pad_ib = amdgpu_ring_generic_pad_ib, .begin_use = amdgpu_uvd_ring_begin_use, .end_use = amdgpu_uvd_ring_end_use, @@ -861,7 +873,7 @@ static const struct amdgpu_ring_funcs uvd_v5_0_ring_funcs = { static void uvd_v5_0_set_ring_funcs(struct amdgpu_device *adev) { - adev->uvd.ring.funcs = &uvd_v5_0_ring_funcs; + adev->uvd.inst->ring.funcs = &uvd_v5_0_ring_funcs; } static const struct amdgpu_irq_src_funcs uvd_v5_0_irq_funcs = { @@ -871,8 +883,8 @@ static const struct amdgpu_irq_src_funcs uvd_v5_0_irq_funcs = { static void uvd_v5_0_set_irq_funcs(struct amdgpu_device *adev) { - adev->uvd.irq.num_types = 1; - adev->uvd.irq.funcs = &uvd_v5_0_irq_funcs; + adev->uvd.inst->irq.num_types = 1; + adev->uvd.inst->irq.funcs = &uvd_v5_0_irq_funcs; } const struct amdgpu_ip_block_version uvd_v5_0_ip_block = diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c b/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c index ca6ab56357b5..bfddf97dd13e 100644 --- a/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c +++ b/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c @@ -91,7 +91,7 @@ static uint64_t uvd_v6_0_enc_ring_get_rptr(struct amdgpu_ring *ring) { struct amdgpu_device *adev = ring->adev; - if (ring == &adev->uvd.ring_enc[0]) + if (ring == &adev->uvd.inst->ring_enc[0]) return RREG32(mmUVD_RB_RPTR); else return RREG32(mmUVD_RB_RPTR2); @@ -121,7 +121,7 @@ static uint64_t uvd_v6_0_enc_ring_get_wptr(struct amdgpu_ring *ring) { struct amdgpu_device *adev = ring->adev; - if (ring == &adev->uvd.ring_enc[0]) + if (ring == &adev->uvd.inst->ring_enc[0]) return RREG32(mmUVD_RB_WPTR); else return RREG32(mmUVD_RB_WPTR2); @@ -152,7 +152,7 @@ static void uvd_v6_0_enc_ring_set_wptr(struct amdgpu_ring *ring) { struct amdgpu_device *adev = ring->adev; - if (ring == &adev->uvd.ring_enc[0]) + if (ring == &adev->uvd.inst->ring_enc[0]) WREG32(mmUVD_RB_WPTR, lower_32_bits(ring->wptr)); else @@ -375,6 +375,7 @@ error: static int uvd_v6_0_early_init(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; + adev->uvd.num_uvd_inst = 1; if (!(adev->flags & AMD_IS_APU) && (RREG32_SMC(ixCC_HARVEST_FUSES) & CC_HARVEST_FUSES__UVD_DISABLE_MASK)) @@ -399,14 +400,14 @@ static int uvd_v6_0_sw_init(void *handle) struct amdgpu_device *adev = (struct amdgpu_device *)handle; /* UVD TRAP */ - r = amdgpu_irq_add_id(adev, AMDGPU_IH_CLIENTID_LEGACY, 124, &adev->uvd.irq); + r = amdgpu_irq_add_id(adev, AMDGPU_IH_CLIENTID_LEGACY, 124, &adev->uvd.inst->irq); if (r) return r; /* UVD ENC TRAP */ if (uvd_v6_0_enc_support(adev)) { for (i = 0; i < adev->uvd.num_enc_rings; ++i) { - r = amdgpu_irq_add_id(adev, AMDGPU_IH_CLIENTID_LEGACY, i + 119, &adev->uvd.irq); + r = amdgpu_irq_add_id(adev, AMDGPU_IH_CLIENTID_LEGACY, i + 119, &adev->uvd.inst->irq); if (r) return r; } @@ -418,17 +419,17 @@ static int uvd_v6_0_sw_init(void *handle) if (!uvd_v6_0_enc_support(adev)) { for (i = 0; i < adev->uvd.num_enc_rings; ++i) - adev->uvd.ring_enc[i].funcs = NULL; + adev->uvd.inst->ring_enc[i].funcs = NULL; - adev->uvd.irq.num_types = 1; + adev->uvd.inst->irq.num_types = 1; adev->uvd.num_enc_rings = 0; DRM_INFO("UVD ENC is disabled\n"); } else { struct drm_sched_rq *rq; - ring = &adev->uvd.ring_enc[0]; + ring = &adev->uvd.inst->ring_enc[0]; rq = &ring->sched.sched_rq[DRM_SCHED_PRIORITY_NORMAL]; - r = drm_sched_entity_init(&ring->sched, &adev->uvd.entity_enc, + r = drm_sched_entity_init(&ring->sched, &adev->uvd.inst->entity_enc, rq, NULL); if (r) { DRM_ERROR("Failed setting up UVD ENC run queue.\n"); @@ -440,17 +441,17 @@ static int uvd_v6_0_sw_init(void *handle) if (r) return r; - ring = &adev->uvd.ring; + ring = &adev->uvd.inst->ring; sprintf(ring->name, "uvd"); - r = amdgpu_ring_init(adev, ring, 512, &adev->uvd.irq, 0); + r = amdgpu_ring_init(adev, ring, 512, &adev->uvd.inst->irq, 0); if (r) return r; if (uvd_v6_0_enc_support(adev)) { for (i = 0; i < adev->uvd.num_enc_rings; ++i) { - ring = &adev->uvd.ring_enc[i]; + ring = &adev->uvd.inst->ring_enc[i]; sprintf(ring->name, "uvd_enc%d", i); - r = amdgpu_ring_init(adev, ring, 512, &adev->uvd.irq, 0); + r = amdgpu_ring_init(adev, ring, 512, &adev->uvd.inst->irq, 0); if (r) return r; } @@ -469,10 +470,10 @@ static int uvd_v6_0_sw_fini(void *handle) return r; if (uvd_v6_0_enc_support(adev)) { - drm_sched_entity_fini(&adev->uvd.ring_enc[0].sched, &adev->uvd.entity_enc); + drm_sched_entity_fini(&adev->uvd.inst->ring_enc[0].sched, &adev->uvd.inst->entity_enc); for (i = 0; i < adev->uvd.num_enc_rings; ++i) - amdgpu_ring_fini(&adev->uvd.ring_enc[i]); + amdgpu_ring_fini(&adev->uvd.inst->ring_enc[i]); } return amdgpu_uvd_sw_fini(adev); @@ -488,7 +489,7 @@ static int uvd_v6_0_sw_fini(void *handle) static int uvd_v6_0_hw_init(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - struct amdgpu_ring *ring = &adev->uvd.ring; + struct amdgpu_ring *ring = &adev->uvd.inst->ring; uint32_t tmp; int i, r; @@ -532,7 +533,7 @@ static int uvd_v6_0_hw_init(void *handle) if (uvd_v6_0_enc_support(adev)) { for (i = 0; i < adev->uvd.num_enc_rings; ++i) { - ring = &adev->uvd.ring_enc[i]; + ring = &adev->uvd.inst->ring_enc[i]; ring->ready = true; r = amdgpu_ring_test_ring(ring); if (r) { @@ -563,7 +564,7 @@ done: static int uvd_v6_0_hw_fini(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - struct amdgpu_ring *ring = &adev->uvd.ring; + struct amdgpu_ring *ring = &adev->uvd.inst->ring; if (RREG32(mmUVD_STATUS) != 0) uvd_v6_0_stop(adev); @@ -611,9 +612,9 @@ static void uvd_v6_0_mc_resume(struct amdgpu_device *adev) /* programm memory controller bits 0-27 */ WREG32(mmUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW, - lower_32_bits(adev->uvd.gpu_addr)); + lower_32_bits(adev->uvd.inst->gpu_addr)); WREG32(mmUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH, - upper_32_bits(adev->uvd.gpu_addr)); + upper_32_bits(adev->uvd.inst->gpu_addr)); offset = AMDGPU_UVD_FIRMWARE_OFFSET; size = AMDGPU_UVD_FIRMWARE_SIZE(adev); @@ -726,7 +727,7 @@ static void cz_set_uvd_clock_gating_branches(struct amdgpu_device *adev, */ static int uvd_v6_0_start(struct amdgpu_device *adev) { - struct amdgpu_ring *ring = &adev->uvd.ring; + struct amdgpu_ring *ring = &adev->uvd.inst->ring; uint32_t rb_bufsz, tmp; uint32_t lmi_swap_cntl; uint32_t mp_swap_cntl; @@ -866,14 +867,14 @@ static int uvd_v6_0_start(struct amdgpu_device *adev) WREG32_FIELD(UVD_RBC_RB_CNTL, RB_NO_FETCH, 0); if (uvd_v6_0_enc_support(adev)) { - ring = &adev->uvd.ring_enc[0]; + ring = &adev->uvd.inst->ring_enc[0]; WREG32(mmUVD_RB_RPTR, lower_32_bits(ring->wptr)); WREG32(mmUVD_RB_WPTR, lower_32_bits(ring->wptr)); WREG32(mmUVD_RB_BASE_LO, ring->gpu_addr); WREG32(mmUVD_RB_BASE_HI, upper_32_bits(ring->gpu_addr)); WREG32(mmUVD_RB_SIZE, ring->ring_size / 4); - ring = &adev->uvd.ring_enc[1]; + ring = &adev->uvd.inst->ring_enc[1]; WREG32(mmUVD_RB_RPTR2, lower_32_bits(ring->wptr)); WREG32(mmUVD_RB_WPTR2, lower_32_bits(ring->wptr)); WREG32(mmUVD_RB_BASE_LO2, ring->gpu_addr); @@ -1099,6 +1100,18 @@ static void uvd_v6_0_ring_emit_pipeline_sync(struct amdgpu_ring *ring) amdgpu_ring_write(ring, 0xE); } +static void uvd_v6_0_ring_insert_nop(struct amdgpu_ring *ring, uint32_t count) +{ + int i; + + WARN_ON(ring->wptr % 2 || count % 2); + + for (i = 0; i < count / 2; i++) { + amdgpu_ring_write(ring, PACKET0(mmUVD_NO_OP, 0)); + amdgpu_ring_write(ring, 0); + } +} + static void uvd_v6_0_enc_ring_emit_pipeline_sync(struct amdgpu_ring *ring) { uint32_t seq = ring->fence_drv.sync_seq; @@ -1158,10 +1171,10 @@ static bool uvd_v6_0_check_soft_reset(void *handle) srbm_soft_reset = REG_SET_FIELD(srbm_soft_reset, SRBM_SOFT_RESET, SOFT_RESET_UVD, 1); if (srbm_soft_reset) { - adev->uvd.srbm_soft_reset = srbm_soft_reset; + adev->uvd.inst->srbm_soft_reset = srbm_soft_reset; return true; } else { - adev->uvd.srbm_soft_reset = 0; + adev->uvd.inst->srbm_soft_reset = 0; return false; } } @@ -1170,7 +1183,7 @@ static int uvd_v6_0_pre_soft_reset(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - if (!adev->uvd.srbm_soft_reset) + if (!adev->uvd.inst->srbm_soft_reset) return 0; uvd_v6_0_stop(adev); @@ -1182,9 +1195,9 @@ static int uvd_v6_0_soft_reset(void *handle) struct amdgpu_device *adev = (struct amdgpu_device *)handle; u32 srbm_soft_reset; - if (!adev->uvd.srbm_soft_reset) + if (!adev->uvd.inst->srbm_soft_reset) return 0; - srbm_soft_reset = adev->uvd.srbm_soft_reset; + srbm_soft_reset = adev->uvd.inst->srbm_soft_reset; if (srbm_soft_reset) { u32 tmp; @@ -1212,7 +1225,7 @@ static int uvd_v6_0_post_soft_reset(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - if (!adev->uvd.srbm_soft_reset) + if (!adev->uvd.inst->srbm_soft_reset) return 0; mdelay(5); @@ -1238,17 +1251,17 @@ static int uvd_v6_0_process_interrupt(struct amdgpu_device *adev, switch (entry->src_id) { case 124: - amdgpu_fence_process(&adev->uvd.ring); + amdgpu_fence_process(&adev->uvd.inst->ring); break; case 119: if (likely(uvd_v6_0_enc_support(adev))) - amdgpu_fence_process(&adev->uvd.ring_enc[0]); + amdgpu_fence_process(&adev->uvd.inst->ring_enc[0]); else int_handled = false; break; case 120: if (likely(uvd_v6_0_enc_support(adev))) - amdgpu_fence_process(&adev->uvd.ring_enc[1]); + amdgpu_fence_process(&adev->uvd.inst->ring_enc[1]); else int_handled = false; break; @@ -1531,7 +1544,6 @@ static const struct amd_ip_funcs uvd_v6_0_ip_funcs = { static const struct amdgpu_ring_funcs uvd_v6_0_ring_phys_funcs = { .type = AMDGPU_RING_TYPE_UVD, .align_mask = 0xf, - .nop = PACKET0(mmUVD_NO_OP, 0), .support_64bit_ptrs = false, .get_rptr = uvd_v6_0_ring_get_rptr, .get_wptr = uvd_v6_0_ring_get_wptr, @@ -1547,7 +1559,7 @@ static const struct amdgpu_ring_funcs uvd_v6_0_ring_phys_funcs = { .emit_hdp_flush = uvd_v6_0_ring_emit_hdp_flush, .test_ring = uvd_v6_0_ring_test_ring, .test_ib = amdgpu_uvd_ring_test_ib, - .insert_nop = amdgpu_ring_insert_nop, + .insert_nop = uvd_v6_0_ring_insert_nop, .pad_ib = amdgpu_ring_generic_pad_ib, .begin_use = amdgpu_uvd_ring_begin_use, .end_use = amdgpu_uvd_ring_end_use, @@ -1612,10 +1624,10 @@ static const struct amdgpu_ring_funcs uvd_v6_0_enc_ring_vm_funcs = { static void uvd_v6_0_set_ring_funcs(struct amdgpu_device *adev) { if (adev->asic_type >= CHIP_POLARIS10) { - adev->uvd.ring.funcs = &uvd_v6_0_ring_vm_funcs; + adev->uvd.inst->ring.funcs = &uvd_v6_0_ring_vm_funcs; DRM_INFO("UVD is enabled in VM mode\n"); } else { - adev->uvd.ring.funcs = &uvd_v6_0_ring_phys_funcs; + adev->uvd.inst->ring.funcs = &uvd_v6_0_ring_phys_funcs; DRM_INFO("UVD is enabled in physical mode\n"); } } @@ -1625,7 +1637,7 @@ static void uvd_v6_0_set_enc_ring_funcs(struct amdgpu_device *adev) int i; for (i = 0; i < adev->uvd.num_enc_rings; ++i) - adev->uvd.ring_enc[i].funcs = &uvd_v6_0_enc_ring_vm_funcs; + adev->uvd.inst->ring_enc[i].funcs = &uvd_v6_0_enc_ring_vm_funcs; DRM_INFO("UVD ENC is enabled in VM mode\n"); } @@ -1638,11 +1650,11 @@ static const struct amdgpu_irq_src_funcs uvd_v6_0_irq_funcs = { static void uvd_v6_0_set_irq_funcs(struct amdgpu_device *adev) { if (uvd_v6_0_enc_support(adev)) - adev->uvd.irq.num_types = adev->uvd.num_enc_rings + 1; + adev->uvd.inst->irq.num_types = adev->uvd.num_enc_rings + 1; else - adev->uvd.irq.num_types = 1; + adev->uvd.inst->irq.num_types = 1; - adev->uvd.irq.funcs = &uvd_v6_0_irq_funcs; + adev->uvd.inst->irq.funcs = &uvd_v6_0_irq_funcs; } const struct amdgpu_ip_block_version uvd_v6_0_ip_block = diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c b/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c index 0ca63d588670..57d32f21b3a6 100644 --- a/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c +++ b/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c @@ -40,6 +40,8 @@ #include "mmhub/mmhub_1_0_offset.h" #include "mmhub/mmhub_1_0_sh_mask.h" +#define UVD7_MAX_HW_INSTANCES_VEGA20 2 + static void uvd_v7_0_set_ring_funcs(struct amdgpu_device *adev); static void uvd_v7_0_set_enc_ring_funcs(struct amdgpu_device *adev); static void uvd_v7_0_set_irq_funcs(struct amdgpu_device *adev); @@ -47,6 +49,11 @@ static int uvd_v7_0_start(struct amdgpu_device *adev); static void uvd_v7_0_stop(struct amdgpu_device *adev); static int uvd_v7_0_sriov_start(struct amdgpu_device *adev); +static int amdgpu_ih_clientid_uvds[] = { + SOC15_IH_CLIENTID_UVD, + SOC15_IH_CLIENTID_UVD1 +}; + /** * uvd_v7_0_ring_get_rptr - get read pointer * @@ -58,7 +65,7 @@ static uint64_t uvd_v7_0_ring_get_rptr(struct amdgpu_ring *ring) { struct amdgpu_device *adev = ring->adev; - return RREG32_SOC15(UVD, 0, mmUVD_RBC_RB_RPTR); + return RREG32_SOC15(UVD, ring->me, mmUVD_RBC_RB_RPTR); } /** @@ -72,10 +79,10 @@ static uint64_t uvd_v7_0_enc_ring_get_rptr(struct amdgpu_ring *ring) { struct amdgpu_device *adev = ring->adev; - if (ring == &adev->uvd.ring_enc[0]) - return RREG32_SOC15(UVD, 0, mmUVD_RB_RPTR); + if (ring == &adev->uvd.inst[ring->me].ring_enc[0]) + return RREG32_SOC15(UVD, ring->me, mmUVD_RB_RPTR); else - return RREG32_SOC15(UVD, 0, mmUVD_RB_RPTR2); + return RREG32_SOC15(UVD, ring->me, mmUVD_RB_RPTR2); } /** @@ -89,7 +96,7 @@ static uint64_t uvd_v7_0_ring_get_wptr(struct amdgpu_ring *ring) { struct amdgpu_device *adev = ring->adev; - return RREG32_SOC15(UVD, 0, mmUVD_RBC_RB_WPTR); + return RREG32_SOC15(UVD, ring->me, mmUVD_RBC_RB_WPTR); } /** @@ -106,10 +113,10 @@ static uint64_t uvd_v7_0_enc_ring_get_wptr(struct amdgpu_ring *ring) if (ring->use_doorbell) return adev->wb.wb[ring->wptr_offs]; - if (ring == &adev->uvd.ring_enc[0]) - return RREG32_SOC15(UVD, 0, mmUVD_RB_WPTR); + if (ring == &adev->uvd.inst[ring->me].ring_enc[0]) + return RREG32_SOC15(UVD, ring->me, mmUVD_RB_WPTR); else - return RREG32_SOC15(UVD, 0, mmUVD_RB_WPTR2); + return RREG32_SOC15(UVD, ring->me, mmUVD_RB_WPTR2); } /** @@ -123,7 +130,7 @@ static void uvd_v7_0_ring_set_wptr(struct amdgpu_ring *ring) { struct amdgpu_device *adev = ring->adev; - WREG32_SOC15(UVD, 0, mmUVD_RBC_RB_WPTR, lower_32_bits(ring->wptr)); + WREG32_SOC15(UVD, ring->me, mmUVD_RBC_RB_WPTR, lower_32_bits(ring->wptr)); } /** @@ -144,11 +151,11 @@ static void uvd_v7_0_enc_ring_set_wptr(struct amdgpu_ring *ring) return; } - if (ring == &adev->uvd.ring_enc[0]) - WREG32_SOC15(UVD, 0, mmUVD_RB_WPTR, + if (ring == &adev->uvd.inst[ring->me].ring_enc[0]) + WREG32_SOC15(UVD, ring->me, mmUVD_RB_WPTR, lower_32_bits(ring->wptr)); else - WREG32_SOC15(UVD, 0, mmUVD_RB_WPTR2, + WREG32_SOC15(UVD, ring->me, mmUVD_RB_WPTR2, lower_32_bits(ring->wptr)); } @@ -170,8 +177,8 @@ static int uvd_v7_0_enc_ring_test_ring(struct amdgpu_ring *ring) r = amdgpu_ring_alloc(ring, 16); if (r) { - DRM_ERROR("amdgpu: uvd enc failed to lock ring %d (%d).\n", - ring->idx, r); + DRM_ERROR("amdgpu: uvd enc failed to lock (%d)ring %d (%d).\n", + ring->me, ring->idx, r); return r; } amdgpu_ring_write(ring, HEVC_ENC_CMD_END); @@ -184,11 +191,11 @@ static int uvd_v7_0_enc_ring_test_ring(struct amdgpu_ring *ring) } if (i < adev->usec_timeout) { - DRM_DEBUG("ring test on %d succeeded in %d usecs\n", - ring->idx, i); + DRM_DEBUG("(%d)ring test on %d succeeded in %d usecs\n", + ring->me, ring->idx, i); } else { - DRM_ERROR("amdgpu: ring %d test failed\n", - ring->idx); + DRM_ERROR("amdgpu: (%d)ring %d test failed\n", + ring->me, ring->idx); r = -ETIMEDOUT; } @@ -342,24 +349,24 @@ static int uvd_v7_0_enc_ring_test_ib(struct amdgpu_ring *ring, long timeout) r = uvd_v7_0_enc_get_create_msg(ring, 1, NULL); if (r) { - DRM_ERROR("amdgpu: failed to get create msg (%ld).\n", r); + DRM_ERROR("amdgpu: (%d)failed to get create msg (%ld).\n", ring->me, r); goto error; } r = uvd_v7_0_enc_get_destroy_msg(ring, 1, true, &fence); if (r) { - DRM_ERROR("amdgpu: failed to get destroy ib (%ld).\n", r); + DRM_ERROR("amdgpu: (%d)failed to get destroy ib (%ld).\n", ring->me, r); goto error; } r = dma_fence_wait_timeout(fence, false, timeout); if (r == 0) { - DRM_ERROR("amdgpu: IB test timed out.\n"); + DRM_ERROR("amdgpu: (%d)IB test timed out.\n", ring->me); r = -ETIMEDOUT; } else if (r < 0) { - DRM_ERROR("amdgpu: fence wait failed (%ld).\n", r); + DRM_ERROR("amdgpu: (%d)fence wait failed (%ld).\n", ring->me, r); } else { - DRM_DEBUG("ib test on ring %d succeeded\n", ring->idx); + DRM_DEBUG("ib test on (%d)ring %d succeeded\n", ring->me, ring->idx); r = 0; } error: @@ -370,6 +377,10 @@ error: static int uvd_v7_0_early_init(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; + if (adev->asic_type == CHIP_VEGA20) + adev->uvd.num_uvd_inst = UVD7_MAX_HW_INSTANCES_VEGA20; + else + adev->uvd.num_uvd_inst = 1; if (amdgpu_sriov_vf(adev)) adev->uvd.num_enc_rings = 1; @@ -386,19 +397,21 @@ static int uvd_v7_0_sw_init(void *handle) { struct amdgpu_ring *ring; struct drm_sched_rq *rq; - int i, r; + int i, j, r; struct amdgpu_device *adev = (struct amdgpu_device *)handle; - /* UVD TRAP */ - r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_UVD, 124, &adev->uvd.irq); - if (r) - return r; - - /* UVD ENC TRAP */ - for (i = 0; i < adev->uvd.num_enc_rings; ++i) { - r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_UVD, i + 119, &adev->uvd.irq); + for (j = 0; j < adev->uvd.num_uvd_inst; j++) { + /* UVD TRAP */ + r = amdgpu_irq_add_id(adev, amdgpu_ih_clientid_uvds[j], 124, &adev->uvd.inst[j].irq); if (r) return r; + + /* UVD ENC TRAP */ + for (i = 0; i < adev->uvd.num_enc_rings; ++i) { + r = amdgpu_irq_add_id(adev, amdgpu_ih_clientid_uvds[j], i + 119, &adev->uvd.inst[j].irq); + if (r) + return r; + } } r = amdgpu_uvd_sw_init(adev); @@ -415,43 +428,48 @@ static int uvd_v7_0_sw_init(void *handle) DRM_INFO("PSP loading UVD firmware\n"); } - ring = &adev->uvd.ring_enc[0]; - rq = &ring->sched.sched_rq[DRM_SCHED_PRIORITY_NORMAL]; - r = drm_sched_entity_init(&ring->sched, &adev->uvd.entity_enc, - rq, NULL); - if (r) { - DRM_ERROR("Failed setting up UVD ENC run queue.\n"); - return r; + for (j = 0; j < adev->uvd.num_uvd_inst; j++) { + ring = &adev->uvd.inst[j].ring_enc[0]; + rq = &ring->sched.sched_rq[DRM_SCHED_PRIORITY_NORMAL]; + r = drm_sched_entity_init(&ring->sched, &adev->uvd.inst[j].entity_enc, + rq, NULL); + if (r) { + DRM_ERROR("(%d)Failed setting up UVD ENC run queue.\n", j); + return r; + } } r = amdgpu_uvd_resume(adev); if (r) return r; - if (!amdgpu_sriov_vf(adev)) { - ring = &adev->uvd.ring; - sprintf(ring->name, "uvd"); - r = amdgpu_ring_init(adev, ring, 512, &adev->uvd.irq, 0); - if (r) - return r; - } - for (i = 0; i < adev->uvd.num_enc_rings; ++i) { - ring = &adev->uvd.ring_enc[i]; - sprintf(ring->name, "uvd_enc%d", i); - if (amdgpu_sriov_vf(adev)) { - ring->use_doorbell = true; - - /* currently only use the first enconding ring for - * sriov, so set unused location for other unused rings. - */ - if (i == 0) - ring->doorbell_index = AMDGPU_DOORBELL64_UVD_RING0_1 * 2; - else - ring->doorbell_index = AMDGPU_DOORBELL64_UVD_RING2_3 * 2 + 1; + for (j = 0; j < adev->uvd.num_uvd_inst; j++) { + if (!amdgpu_sriov_vf(adev)) { + ring = &adev->uvd.inst[j].ring; + sprintf(ring->name, "uvd<%d>", j); + r = amdgpu_ring_init(adev, ring, 512, &adev->uvd.inst[j].irq, 0); + if (r) + return r; + } + + for (i = 0; i < adev->uvd.num_enc_rings; ++i) { + ring = &adev->uvd.inst[j].ring_enc[i]; + sprintf(ring->name, "uvd_enc%d<%d>", i, j); + if (amdgpu_sriov_vf(adev)) { + ring->use_doorbell = true; + + /* currently only use the first enconding ring for + * sriov, so set unused location for other unused rings. + */ + if (i == 0) + ring->doorbell_index = AMDGPU_DOORBELL64_UVD_RING0_1 * 2; + else + ring->doorbell_index = AMDGPU_DOORBELL64_UVD_RING2_3 * 2 + 1; + } + r = amdgpu_ring_init(adev, ring, 512, &adev->uvd.inst[j].irq, 0); + if (r) + return r; } - r = amdgpu_ring_init(adev, ring, 512, &adev->uvd.irq, 0); - if (r) - return r; } r = amdgpu_virt_alloc_mm_table(adev); @@ -463,7 +481,7 @@ static int uvd_v7_0_sw_init(void *handle) static int uvd_v7_0_sw_fini(void *handle) { - int i, r; + int i, j, r; struct amdgpu_device *adev = (struct amdgpu_device *)handle; amdgpu_virt_free_mm_table(adev); @@ -472,11 +490,12 @@ static int uvd_v7_0_sw_fini(void *handle) if (r) return r; - drm_sched_entity_fini(&adev->uvd.ring_enc[0].sched, &adev->uvd.entity_enc); - - for (i = 0; i < adev->uvd.num_enc_rings; ++i) - amdgpu_ring_fini(&adev->uvd.ring_enc[i]); + for (j = 0; j < adev->uvd.num_uvd_inst; ++j) { + drm_sched_entity_fini(&adev->uvd.inst[j].ring_enc[0].sched, &adev->uvd.inst[j].entity_enc); + for (i = 0; i < adev->uvd.num_enc_rings; ++i) + amdgpu_ring_fini(&adev->uvd.inst[j].ring_enc[i]); + } return amdgpu_uvd_sw_fini(adev); } @@ -490,9 +509,9 @@ static int uvd_v7_0_sw_fini(void *handle) static int uvd_v7_0_hw_init(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - struct amdgpu_ring *ring = &adev->uvd.ring; + struct amdgpu_ring *ring; uint32_t tmp; - int i, r; + int i, j, r; if (amdgpu_sriov_vf(adev)) r = uvd_v7_0_sriov_start(adev); @@ -501,57 +520,60 @@ static int uvd_v7_0_hw_init(void *handle) if (r) goto done; - if (!amdgpu_sriov_vf(adev)) { - ring->ready = true; - r = amdgpu_ring_test_ring(ring); - if (r) { - ring->ready = false; - goto done; - } - - r = amdgpu_ring_alloc(ring, 10); - if (r) { - DRM_ERROR("amdgpu: ring failed to lock UVD ring (%d).\n", r); - goto done; + for (j = 0; j < adev->uvd.num_uvd_inst; ++j) { + ring = &adev->uvd.inst[j].ring; + + if (!amdgpu_sriov_vf(adev)) { + ring->ready = true; + r = amdgpu_ring_test_ring(ring); + if (r) { + ring->ready = false; + goto done; + } + + r = amdgpu_ring_alloc(ring, 10); + if (r) { + DRM_ERROR("amdgpu: (%d)ring failed to lock UVD ring (%d).\n", j, r); + goto done; + } + + tmp = PACKET0(SOC15_REG_OFFSET(UVD, j, + mmUVD_SEMA_WAIT_FAULT_TIMEOUT_CNTL), 0); + amdgpu_ring_write(ring, tmp); + amdgpu_ring_write(ring, 0xFFFFF); + + tmp = PACKET0(SOC15_REG_OFFSET(UVD, j, + mmUVD_SEMA_WAIT_INCOMPLETE_TIMEOUT_CNTL), 0); + amdgpu_ring_write(ring, tmp); + amdgpu_ring_write(ring, 0xFFFFF); + + tmp = PACKET0(SOC15_REG_OFFSET(UVD, j, + mmUVD_SEMA_SIGNAL_INCOMPLETE_TIMEOUT_CNTL), 0); + amdgpu_ring_write(ring, tmp); + amdgpu_ring_write(ring, 0xFFFFF); + + /* Clear timeout status bits */ + amdgpu_ring_write(ring, PACKET0(SOC15_REG_OFFSET(UVD, j, + mmUVD_SEMA_TIMEOUT_STATUS), 0)); + amdgpu_ring_write(ring, 0x8); + + amdgpu_ring_write(ring, PACKET0(SOC15_REG_OFFSET(UVD, j, + mmUVD_SEMA_CNTL), 0)); + amdgpu_ring_write(ring, 3); + + amdgpu_ring_commit(ring); } - tmp = PACKET0(SOC15_REG_OFFSET(UVD, 0, - mmUVD_SEMA_WAIT_FAULT_TIMEOUT_CNTL), 0); - amdgpu_ring_write(ring, tmp); - amdgpu_ring_write(ring, 0xFFFFF); - - tmp = PACKET0(SOC15_REG_OFFSET(UVD, 0, - mmUVD_SEMA_WAIT_INCOMPLETE_TIMEOUT_CNTL), 0); - amdgpu_ring_write(ring, tmp); - amdgpu_ring_write(ring, 0xFFFFF); - - tmp = PACKET0(SOC15_REG_OFFSET(UVD, 0, - mmUVD_SEMA_SIGNAL_INCOMPLETE_TIMEOUT_CNTL), 0); - amdgpu_ring_write(ring, tmp); - amdgpu_ring_write(ring, 0xFFFFF); - - /* Clear timeout status bits */ - amdgpu_ring_write(ring, PACKET0(SOC15_REG_OFFSET(UVD, 0, - mmUVD_SEMA_TIMEOUT_STATUS), 0)); - amdgpu_ring_write(ring, 0x8); - - amdgpu_ring_write(ring, PACKET0(SOC15_REG_OFFSET(UVD, 0, - mmUVD_SEMA_CNTL), 0)); - amdgpu_ring_write(ring, 3); - - amdgpu_ring_commit(ring); - } - - for (i = 0; i < adev->uvd.num_enc_rings; ++i) { - ring = &adev->uvd.ring_enc[i]; - ring->ready = true; - r = amdgpu_ring_test_ring(ring); - if (r) { - ring->ready = false; - goto done; + for (i = 0; i < adev->uvd.num_enc_rings; ++i) { + ring = &adev->uvd.inst[j].ring_enc[i]; + ring->ready = true; + r = amdgpu_ring_test_ring(ring); + if (r) { + ring->ready = false; + goto done; + } } } - done: if (!r) DRM_INFO("UVD and UVD ENC initialized successfully.\n"); @@ -569,7 +591,7 @@ done: static int uvd_v7_0_hw_fini(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - struct amdgpu_ring *ring = &adev->uvd.ring; + int i; if (!amdgpu_sriov_vf(adev)) uvd_v7_0_stop(adev); @@ -578,7 +600,8 @@ static int uvd_v7_0_hw_fini(void *handle) DRM_DEBUG("For SRIOV client, shouldn't do anything.\n"); } - ring->ready = false; + for (i = 0; i < adev->uvd.num_uvd_inst; ++i) + adev->uvd.inst[i].ring.ready = false; return 0; } @@ -618,48 +641,51 @@ static void uvd_v7_0_mc_resume(struct amdgpu_device *adev) { uint32_t size = AMDGPU_UVD_FIRMWARE_SIZE(adev); uint32_t offset; + int i; - if (adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) { - WREG32_SOC15(UVD, 0, mmUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW, - lower_32_bits(adev->firmware.ucode[AMDGPU_UCODE_ID_UVD].mc_addr)); - WREG32_SOC15(UVD, 0, mmUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH, - upper_32_bits(adev->firmware.ucode[AMDGPU_UCODE_ID_UVD].mc_addr)); - offset = 0; - } else { - WREG32_SOC15(UVD, 0, mmUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW, - lower_32_bits(adev->uvd.gpu_addr)); - WREG32_SOC15(UVD, 0, mmUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH, - upper_32_bits(adev->uvd.gpu_addr)); - offset = size; - } + for (i = 0; i < adev->uvd.num_uvd_inst; ++i) { + if (adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) { + WREG32_SOC15(UVD, i, mmUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW, + lower_32_bits(adev->firmware.ucode[AMDGPU_UCODE_ID_UVD].mc_addr)); + WREG32_SOC15(UVD, i, mmUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH, + upper_32_bits(adev->firmware.ucode[AMDGPU_UCODE_ID_UVD].mc_addr)); + offset = 0; + } else { + WREG32_SOC15(UVD, i, mmUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW, + lower_32_bits(adev->uvd.inst[i].gpu_addr)); + WREG32_SOC15(UVD, i, mmUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH, + upper_32_bits(adev->uvd.inst[i].gpu_addr)); + offset = size; + } - WREG32_SOC15(UVD, 0, mmUVD_VCPU_CACHE_OFFSET0, - AMDGPU_UVD_FIRMWARE_OFFSET >> 3); - WREG32_SOC15(UVD, 0, mmUVD_VCPU_CACHE_SIZE0, size); - - WREG32_SOC15(UVD, 0, mmUVD_LMI_VCPU_CACHE1_64BIT_BAR_LOW, - lower_32_bits(adev->uvd.gpu_addr + offset)); - WREG32_SOC15(UVD, 0, mmUVD_LMI_VCPU_CACHE1_64BIT_BAR_HIGH, - upper_32_bits(adev->uvd.gpu_addr + offset)); - WREG32_SOC15(UVD, 0, mmUVD_VCPU_CACHE_OFFSET1, (1 << 21)); - WREG32_SOC15(UVD, 0, mmUVD_VCPU_CACHE_SIZE1, AMDGPU_UVD_HEAP_SIZE); - - WREG32_SOC15(UVD, 0, mmUVD_LMI_VCPU_CACHE2_64BIT_BAR_LOW, - lower_32_bits(adev->uvd.gpu_addr + offset + AMDGPU_UVD_HEAP_SIZE)); - WREG32_SOC15(UVD, 0, mmUVD_LMI_VCPU_CACHE2_64BIT_BAR_HIGH, - upper_32_bits(adev->uvd.gpu_addr + offset + AMDGPU_UVD_HEAP_SIZE)); - WREG32_SOC15(UVD, 0, mmUVD_VCPU_CACHE_OFFSET2, (2 << 21)); - WREG32_SOC15(UVD, 0, mmUVD_VCPU_CACHE_SIZE2, - AMDGPU_UVD_STACK_SIZE + (AMDGPU_UVD_SESSION_SIZE * 40)); - - WREG32_SOC15(UVD, 0, mmUVD_UDEC_ADDR_CONFIG, - adev->gfx.config.gb_addr_config); - WREG32_SOC15(UVD, 0, mmUVD_UDEC_DB_ADDR_CONFIG, - adev->gfx.config.gb_addr_config); - WREG32_SOC15(UVD, 0, mmUVD_UDEC_DBW_ADDR_CONFIG, - adev->gfx.config.gb_addr_config); - - WREG32_SOC15(UVD, 0, mmUVD_GP_SCRATCH4, adev->uvd.max_handles); + WREG32_SOC15(UVD, i, mmUVD_VCPU_CACHE_OFFSET0, + AMDGPU_UVD_FIRMWARE_OFFSET >> 3); + WREG32_SOC15(UVD, i, mmUVD_VCPU_CACHE_SIZE0, size); + + WREG32_SOC15(UVD, i, mmUVD_LMI_VCPU_CACHE1_64BIT_BAR_LOW, + lower_32_bits(adev->uvd.inst[i].gpu_addr + offset)); + WREG32_SOC15(UVD, i, mmUVD_LMI_VCPU_CACHE1_64BIT_BAR_HIGH, + upper_32_bits(adev->uvd.inst[i].gpu_addr + offset)); + WREG32_SOC15(UVD, i, mmUVD_VCPU_CACHE_OFFSET1, (1 << 21)); + WREG32_SOC15(UVD, i, mmUVD_VCPU_CACHE_SIZE1, AMDGPU_UVD_HEAP_SIZE); + + WREG32_SOC15(UVD, i, mmUVD_LMI_VCPU_CACHE2_64BIT_BAR_LOW, + lower_32_bits(adev->uvd.inst[i].gpu_addr + offset + AMDGPU_UVD_HEAP_SIZE)); + WREG32_SOC15(UVD, i, mmUVD_LMI_VCPU_CACHE2_64BIT_BAR_HIGH, + upper_32_bits(adev->uvd.inst[i].gpu_addr + offset + AMDGPU_UVD_HEAP_SIZE)); + WREG32_SOC15(UVD, i, mmUVD_VCPU_CACHE_OFFSET2, (2 << 21)); + WREG32_SOC15(UVD, i, mmUVD_VCPU_CACHE_SIZE2, + AMDGPU_UVD_STACK_SIZE + (AMDGPU_UVD_SESSION_SIZE * 40)); + + WREG32_SOC15(UVD, i, mmUVD_UDEC_ADDR_CONFIG, + adev->gfx.config.gb_addr_config); + WREG32_SOC15(UVD, i, mmUVD_UDEC_DB_ADDR_CONFIG, + adev->gfx.config.gb_addr_config); + WREG32_SOC15(UVD, i, mmUVD_UDEC_DBW_ADDR_CONFIG, + adev->gfx.config.gb_addr_config); + + WREG32_SOC15(UVD, i, mmUVD_GP_SCRATCH4, adev->uvd.max_handles); + } } static int uvd_v7_0_mmsch_start(struct amdgpu_device *adev, @@ -669,6 +695,7 @@ static int uvd_v7_0_mmsch_start(struct amdgpu_device *adev, uint64_t addr = table->gpu_addr; struct mmsch_v1_0_init_header *header = (struct mmsch_v1_0_init_header *)table->cpu_addr; uint32_t size; + int i; size = header->header_size + header->vce_table_size + header->uvd_table_size; @@ -688,11 +715,12 @@ static int uvd_v7_0_mmsch_start(struct amdgpu_device *adev, /* 4, set resp to zero */ WREG32_SOC15(VCE, 0, mmVCE_MMSCH_VF_MAILBOX_RESP, 0); - WDOORBELL32(adev->uvd.ring_enc[0].doorbell_index, 0); - adev->wb.wb[adev->uvd.ring_enc[0].wptr_offs] = 0; - adev->uvd.ring_enc[0].wptr = 0; - adev->uvd.ring_enc[0].wptr_old = 0; - + for (i = 0; i < adev->uvd.num_uvd_inst; ++i) { + WDOORBELL32(adev->uvd.inst[i].ring_enc[0].doorbell_index, 0); + adev->wb.wb[adev->uvd.inst[i].ring_enc[0].wptr_offs] = 0; + adev->uvd.inst[i].ring_enc[0].wptr = 0; + adev->uvd.inst[i].ring_enc[0].wptr_old = 0; + } /* 5, kick off the initialization and wait until VCE_MMSCH_VF_MAILBOX_RESP becomes non-zero */ WREG32_SOC15(VCE, 0, mmVCE_MMSCH_VF_MAILBOX_HOST, 0x10000001); @@ -725,6 +753,7 @@ static int uvd_v7_0_sriov_start(struct amdgpu_device *adev) struct mmsch_v1_0_cmd_end end = { {0} }; uint32_t *init_table = adev->virt.mm_table.cpu_addr; struct mmsch_v1_0_init_header *header = (struct mmsch_v1_0_init_header *)init_table; + uint8_t i = 0; direct_wt.cmd_header.command_type = MMSCH_COMMAND__DIRECT_REG_WRITE; direct_rd_mod_wt.cmd_header.command_type = MMSCH_COMMAND__DIRECT_REG_READ_MODIFY_WRITE; @@ -742,120 +771,121 @@ static int uvd_v7_0_sriov_start(struct amdgpu_device *adev) init_table += header->uvd_table_offset; - ring = &adev->uvd.ring; - ring->wptr = 0; - size = AMDGPU_GPU_PAGE_ALIGN(adev->uvd.fw->size + 4); - - MMSCH_V1_0_INSERT_DIRECT_RD_MOD_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_STATUS), - 0xFFFFFFFF, 0x00000004); - /* mc resume*/ - if (adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) { - MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW), - lower_32_bits(adev->firmware.ucode[AMDGPU_UCODE_ID_UVD].mc_addr)); - MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH), - upper_32_bits(adev->firmware.ucode[AMDGPU_UCODE_ID_UVD].mc_addr)); - offset = 0; - } else { - MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW), - lower_32_bits(adev->uvd.gpu_addr)); - MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH), - upper_32_bits(adev->uvd.gpu_addr)); - offset = size; + for (i = 0; i < adev->uvd.num_uvd_inst; ++i) { + ring = &adev->uvd.inst[i].ring; + ring->wptr = 0; + size = AMDGPU_GPU_PAGE_ALIGN(adev->uvd.fw->size + 4); + + MMSCH_V1_0_INSERT_DIRECT_RD_MOD_WT(SOC15_REG_OFFSET(UVD, i, mmUVD_STATUS), + 0xFFFFFFFF, 0x00000004); + /* mc resume*/ + if (adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) { + MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, i, mmUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW), + lower_32_bits(adev->firmware.ucode[AMDGPU_UCODE_ID_UVD].mc_addr)); + MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, i, mmUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH), + upper_32_bits(adev->firmware.ucode[AMDGPU_UCODE_ID_UVD].mc_addr)); + offset = 0; + } else { + MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, i, mmUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW), + lower_32_bits(adev->uvd.inst[i].gpu_addr)); + MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, i, mmUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH), + upper_32_bits(adev->uvd.inst[i].gpu_addr)); + offset = size; + } + + MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, i, mmUVD_VCPU_CACHE_OFFSET0), + AMDGPU_UVD_FIRMWARE_OFFSET >> 3); + MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, i, mmUVD_VCPU_CACHE_SIZE0), size); + + MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, i, mmUVD_LMI_VCPU_CACHE1_64BIT_BAR_LOW), + lower_32_bits(adev->uvd.inst[i].gpu_addr + offset)); + MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, i, mmUVD_LMI_VCPU_CACHE1_64BIT_BAR_HIGH), + upper_32_bits(adev->uvd.inst[i].gpu_addr + offset)); + MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, i, mmUVD_VCPU_CACHE_OFFSET1), (1 << 21)); + MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, i, mmUVD_VCPU_CACHE_SIZE1), AMDGPU_UVD_HEAP_SIZE); + + MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, i, mmUVD_LMI_VCPU_CACHE2_64BIT_BAR_LOW), + lower_32_bits(adev->uvd.inst[i].gpu_addr + offset + AMDGPU_UVD_HEAP_SIZE)); + MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, i, mmUVD_LMI_VCPU_CACHE2_64BIT_BAR_HIGH), + upper_32_bits(adev->uvd.inst[i].gpu_addr + offset + AMDGPU_UVD_HEAP_SIZE)); + MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, i, mmUVD_VCPU_CACHE_OFFSET2), (2 << 21)); + MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, i, mmUVD_VCPU_CACHE_SIZE2), + AMDGPU_UVD_STACK_SIZE + (AMDGPU_UVD_SESSION_SIZE * 40)); + + MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, i, mmUVD_GP_SCRATCH4), adev->uvd.max_handles); + /* mc resume end*/ + + /* disable clock gating */ + MMSCH_V1_0_INSERT_DIRECT_RD_MOD_WT(SOC15_REG_OFFSET(UVD, i, mmUVD_CGC_CTRL), + ~UVD_CGC_CTRL__DYN_CLOCK_MODE_MASK, 0); + + /* disable interupt */ + MMSCH_V1_0_INSERT_DIRECT_RD_MOD_WT(SOC15_REG_OFFSET(UVD, i, mmUVD_MASTINT_EN), + ~UVD_MASTINT_EN__VCPU_EN_MASK, 0); + + /* stall UMC and register bus before resetting VCPU */ + MMSCH_V1_0_INSERT_DIRECT_RD_MOD_WT(SOC15_REG_OFFSET(UVD, i, mmUVD_LMI_CTRL2), + ~UVD_LMI_CTRL2__STALL_ARB_UMC_MASK, + UVD_LMI_CTRL2__STALL_ARB_UMC_MASK); + + /* put LMI, VCPU, RBC etc... into reset */ + MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, i, mmUVD_SOFT_RESET), + (uint32_t)(UVD_SOFT_RESET__LMI_SOFT_RESET_MASK | + UVD_SOFT_RESET__VCPU_SOFT_RESET_MASK | + UVD_SOFT_RESET__LBSI_SOFT_RESET_MASK | + UVD_SOFT_RESET__RBC_SOFT_RESET_MASK | + UVD_SOFT_RESET__CSM_SOFT_RESET_MASK | + UVD_SOFT_RESET__CXW_SOFT_RESET_MASK | + UVD_SOFT_RESET__TAP_SOFT_RESET_MASK | + UVD_SOFT_RESET__LMI_UMC_SOFT_RESET_MASK)); + + /* initialize UVD memory controller */ + MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, i, mmUVD_LMI_CTRL), + (uint32_t)((0x40 << UVD_LMI_CTRL__WRITE_CLEAN_TIMER__SHIFT) | + UVD_LMI_CTRL__WRITE_CLEAN_TIMER_EN_MASK | + UVD_LMI_CTRL__DATA_COHERENCY_EN_MASK | + UVD_LMI_CTRL__VCPU_DATA_COHERENCY_EN_MASK | + UVD_LMI_CTRL__REQ_MODE_MASK | + 0x00100000L)); + + /* take all subblocks out of reset, except VCPU */ + MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, i, mmUVD_SOFT_RESET), + UVD_SOFT_RESET__VCPU_SOFT_RESET_MASK); + + /* enable VCPU clock */ + MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, i, mmUVD_VCPU_CNTL), + UVD_VCPU_CNTL__CLK_EN_MASK); + + /* enable master interrupt */ + MMSCH_V1_0_INSERT_DIRECT_RD_MOD_WT(SOC15_REG_OFFSET(UVD, i, mmUVD_MASTINT_EN), + ~(UVD_MASTINT_EN__VCPU_EN_MASK|UVD_MASTINT_EN__SYS_EN_MASK), + (UVD_MASTINT_EN__VCPU_EN_MASK|UVD_MASTINT_EN__SYS_EN_MASK)); + + /* clear the bit 4 of UVD_STATUS */ + MMSCH_V1_0_INSERT_DIRECT_RD_MOD_WT(SOC15_REG_OFFSET(UVD, i, mmUVD_STATUS), + ~(2 << UVD_STATUS__VCPU_REPORT__SHIFT), 0); + + /* force RBC into idle state */ + size = order_base_2(ring->ring_size); + tmp = REG_SET_FIELD(0, UVD_RBC_RB_CNTL, RB_BUFSZ, size); + tmp = REG_SET_FIELD(tmp, UVD_RBC_RB_CNTL, RB_NO_FETCH, 1); + MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, i, mmUVD_RBC_RB_CNTL), tmp); + + ring = &adev->uvd.inst[i].ring_enc[0]; + ring->wptr = 0; + MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, i, mmUVD_RB_BASE_LO), ring->gpu_addr); + MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, i, mmUVD_RB_BASE_HI), upper_32_bits(ring->gpu_addr)); + MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, i, mmUVD_RB_SIZE), ring->ring_size / 4); + + /* boot up the VCPU */ + MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, i, mmUVD_SOFT_RESET), 0); + + /* enable UMC */ + MMSCH_V1_0_INSERT_DIRECT_RD_MOD_WT(SOC15_REG_OFFSET(UVD, i, mmUVD_LMI_CTRL2), + ~UVD_LMI_CTRL2__STALL_ARB_UMC_MASK, 0); + + MMSCH_V1_0_INSERT_DIRECT_POLL(SOC15_REG_OFFSET(UVD, i, mmUVD_STATUS), 0x02, 0x02); } - - MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_VCPU_CACHE_OFFSET0), - AMDGPU_UVD_FIRMWARE_OFFSET >> 3); - MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_VCPU_CACHE_SIZE0), size); - - MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_LMI_VCPU_CACHE1_64BIT_BAR_LOW), - lower_32_bits(adev->uvd.gpu_addr + offset)); - MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_LMI_VCPU_CACHE1_64BIT_BAR_HIGH), - upper_32_bits(adev->uvd.gpu_addr + offset)); - MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_VCPU_CACHE_OFFSET1), (1 << 21)); - MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_VCPU_CACHE_SIZE1), AMDGPU_UVD_HEAP_SIZE); - - MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_LMI_VCPU_CACHE2_64BIT_BAR_LOW), - lower_32_bits(adev->uvd.gpu_addr + offset + AMDGPU_UVD_HEAP_SIZE)); - MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_LMI_VCPU_CACHE2_64BIT_BAR_HIGH), - upper_32_bits(adev->uvd.gpu_addr + offset + AMDGPU_UVD_HEAP_SIZE)); - MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_VCPU_CACHE_OFFSET2), (2 << 21)); - MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_VCPU_CACHE_SIZE2), - AMDGPU_UVD_STACK_SIZE + (AMDGPU_UVD_SESSION_SIZE * 40)); - - MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_GP_SCRATCH4), adev->uvd.max_handles); - /* mc resume end*/ - - /* disable clock gating */ - MMSCH_V1_0_INSERT_DIRECT_RD_MOD_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_CGC_CTRL), - ~UVD_CGC_CTRL__DYN_CLOCK_MODE_MASK, 0); - - /* disable interupt */ - MMSCH_V1_0_INSERT_DIRECT_RD_MOD_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_MASTINT_EN), - ~UVD_MASTINT_EN__VCPU_EN_MASK, 0); - - /* stall UMC and register bus before resetting VCPU */ - MMSCH_V1_0_INSERT_DIRECT_RD_MOD_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_LMI_CTRL2), - ~UVD_LMI_CTRL2__STALL_ARB_UMC_MASK, - UVD_LMI_CTRL2__STALL_ARB_UMC_MASK); - - /* put LMI, VCPU, RBC etc... into reset */ - MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_SOFT_RESET), - (uint32_t)(UVD_SOFT_RESET__LMI_SOFT_RESET_MASK | - UVD_SOFT_RESET__VCPU_SOFT_RESET_MASK | - UVD_SOFT_RESET__LBSI_SOFT_RESET_MASK | - UVD_SOFT_RESET__RBC_SOFT_RESET_MASK | - UVD_SOFT_RESET__CSM_SOFT_RESET_MASK | - UVD_SOFT_RESET__CXW_SOFT_RESET_MASK | - UVD_SOFT_RESET__TAP_SOFT_RESET_MASK | - UVD_SOFT_RESET__LMI_UMC_SOFT_RESET_MASK)); - - /* initialize UVD memory controller */ - MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_LMI_CTRL), - (uint32_t)((0x40 << UVD_LMI_CTRL__WRITE_CLEAN_TIMER__SHIFT) | - UVD_LMI_CTRL__WRITE_CLEAN_TIMER_EN_MASK | - UVD_LMI_CTRL__DATA_COHERENCY_EN_MASK | - UVD_LMI_CTRL__VCPU_DATA_COHERENCY_EN_MASK | - UVD_LMI_CTRL__REQ_MODE_MASK | - 0x00100000L)); - - /* take all subblocks out of reset, except VCPU */ - MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_SOFT_RESET), - UVD_SOFT_RESET__VCPU_SOFT_RESET_MASK); - - /* enable VCPU clock */ - MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_VCPU_CNTL), - UVD_VCPU_CNTL__CLK_EN_MASK); - - /* enable master interrupt */ - MMSCH_V1_0_INSERT_DIRECT_RD_MOD_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_MASTINT_EN), - ~(UVD_MASTINT_EN__VCPU_EN_MASK|UVD_MASTINT_EN__SYS_EN_MASK), - (UVD_MASTINT_EN__VCPU_EN_MASK|UVD_MASTINT_EN__SYS_EN_MASK)); - - /* clear the bit 4 of UVD_STATUS */ - MMSCH_V1_0_INSERT_DIRECT_RD_MOD_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_STATUS), - ~(2 << UVD_STATUS__VCPU_REPORT__SHIFT), 0); - - /* force RBC into idle state */ - size = order_base_2(ring->ring_size); - tmp = REG_SET_FIELD(0, UVD_RBC_RB_CNTL, RB_BUFSZ, size); - tmp = REG_SET_FIELD(tmp, UVD_RBC_RB_CNTL, RB_NO_FETCH, 1); - MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_RBC_RB_CNTL), tmp); - - ring = &adev->uvd.ring_enc[0]; - ring->wptr = 0; - MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_RB_BASE_LO), ring->gpu_addr); - MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_RB_BASE_HI), upper_32_bits(ring->gpu_addr)); - MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_RB_SIZE), ring->ring_size / 4); - - /* boot up the VCPU */ - MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_SOFT_RESET), 0); - - /* enable UMC */ - MMSCH_V1_0_INSERT_DIRECT_RD_MOD_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_LMI_CTRL2), - ~UVD_LMI_CTRL2__STALL_ARB_UMC_MASK, 0); - - MMSCH_V1_0_INSERT_DIRECT_POLL(SOC15_REG_OFFSET(UVD, 0, mmUVD_STATUS), 0x02, 0x02); - /* add end packet */ memcpy((void *)init_table, &end, sizeof(struct mmsch_v1_0_cmd_end)); table_size += sizeof(struct mmsch_v1_0_cmd_end) / 4; @@ -874,15 +904,17 @@ static int uvd_v7_0_sriov_start(struct amdgpu_device *adev) */ static int uvd_v7_0_start(struct amdgpu_device *adev) { - struct amdgpu_ring *ring = &adev->uvd.ring; + struct amdgpu_ring *ring; uint32_t rb_bufsz, tmp; uint32_t lmi_swap_cntl; uint32_t mp_swap_cntl; - int i, j, r; + int i, j, k, r; - /* disable DPG */ - WREG32_P(SOC15_REG_OFFSET(UVD, 0, mmUVD_POWER_STATUS), 0, - ~UVD_POWER_STATUS__UVD_PG_MODE_MASK); + for (k = 0; k < adev->uvd.num_uvd_inst; ++k) { + /* disable DPG */ + WREG32_P(SOC15_REG_OFFSET(UVD, k, mmUVD_POWER_STATUS), 0, + ~UVD_POWER_STATUS__UVD_PG_MODE_MASK); + } /* disable byte swapping */ lmi_swap_cntl = 0; @@ -890,157 +922,159 @@ static int uvd_v7_0_start(struct amdgpu_device *adev) uvd_v7_0_mc_resume(adev); - /* disable clock gating */ - WREG32_P(SOC15_REG_OFFSET(UVD, 0, mmUVD_CGC_CTRL), 0, - ~UVD_CGC_CTRL__DYN_CLOCK_MODE_MASK); - - /* disable interupt */ - WREG32_P(SOC15_REG_OFFSET(UVD, 0, mmUVD_MASTINT_EN), 0, - ~UVD_MASTINT_EN__VCPU_EN_MASK); - - /* stall UMC and register bus before resetting VCPU */ - WREG32_P(SOC15_REG_OFFSET(UVD, 0, mmUVD_LMI_CTRL2), - UVD_LMI_CTRL2__STALL_ARB_UMC_MASK, - ~UVD_LMI_CTRL2__STALL_ARB_UMC_MASK); - mdelay(1); - - /* put LMI, VCPU, RBC etc... into reset */ - WREG32_SOC15(UVD, 0, mmUVD_SOFT_RESET, - UVD_SOFT_RESET__LMI_SOFT_RESET_MASK | - UVD_SOFT_RESET__VCPU_SOFT_RESET_MASK | - UVD_SOFT_RESET__LBSI_SOFT_RESET_MASK | - UVD_SOFT_RESET__RBC_SOFT_RESET_MASK | - UVD_SOFT_RESET__CSM_SOFT_RESET_MASK | - UVD_SOFT_RESET__CXW_SOFT_RESET_MASK | - UVD_SOFT_RESET__TAP_SOFT_RESET_MASK | - UVD_SOFT_RESET__LMI_UMC_SOFT_RESET_MASK); - mdelay(5); + for (k = 0; k < adev->uvd.num_uvd_inst; ++k) { + ring = &adev->uvd.inst[k].ring; + /* disable clock gating */ + WREG32_P(SOC15_REG_OFFSET(UVD, k, mmUVD_CGC_CTRL), 0, + ~UVD_CGC_CTRL__DYN_CLOCK_MODE_MASK); - /* initialize UVD memory controller */ - WREG32_SOC15(UVD, 0, mmUVD_LMI_CTRL, - (0x40 << UVD_LMI_CTRL__WRITE_CLEAN_TIMER__SHIFT) | - UVD_LMI_CTRL__WRITE_CLEAN_TIMER_EN_MASK | - UVD_LMI_CTRL__DATA_COHERENCY_EN_MASK | - UVD_LMI_CTRL__VCPU_DATA_COHERENCY_EN_MASK | - UVD_LMI_CTRL__REQ_MODE_MASK | - 0x00100000L); + /* disable interupt */ + WREG32_P(SOC15_REG_OFFSET(UVD, k, mmUVD_MASTINT_EN), 0, + ~UVD_MASTINT_EN__VCPU_EN_MASK); + + /* stall UMC and register bus before resetting VCPU */ + WREG32_P(SOC15_REG_OFFSET(UVD, k, mmUVD_LMI_CTRL2), + UVD_LMI_CTRL2__STALL_ARB_UMC_MASK, + ~UVD_LMI_CTRL2__STALL_ARB_UMC_MASK); + mdelay(1); + + /* put LMI, VCPU, RBC etc... into reset */ + WREG32_SOC15(UVD, k, mmUVD_SOFT_RESET, + UVD_SOFT_RESET__LMI_SOFT_RESET_MASK | + UVD_SOFT_RESET__VCPU_SOFT_RESET_MASK | + UVD_SOFT_RESET__LBSI_SOFT_RESET_MASK | + UVD_SOFT_RESET__RBC_SOFT_RESET_MASK | + UVD_SOFT_RESET__CSM_SOFT_RESET_MASK | + UVD_SOFT_RESET__CXW_SOFT_RESET_MASK | + UVD_SOFT_RESET__TAP_SOFT_RESET_MASK | + UVD_SOFT_RESET__LMI_UMC_SOFT_RESET_MASK); + mdelay(5); + + /* initialize UVD memory controller */ + WREG32_SOC15(UVD, k, mmUVD_LMI_CTRL, + (0x40 << UVD_LMI_CTRL__WRITE_CLEAN_TIMER__SHIFT) | + UVD_LMI_CTRL__WRITE_CLEAN_TIMER_EN_MASK | + UVD_LMI_CTRL__DATA_COHERENCY_EN_MASK | + UVD_LMI_CTRL__VCPU_DATA_COHERENCY_EN_MASK | + UVD_LMI_CTRL__REQ_MODE_MASK | + 0x00100000L); #ifdef __BIG_ENDIAN - /* swap (8 in 32) RB and IB */ - lmi_swap_cntl = 0xa; - mp_swap_cntl = 0; + /* swap (8 in 32) RB and IB */ + lmi_swap_cntl = 0xa; + mp_swap_cntl = 0; #endif - WREG32_SOC15(UVD, 0, mmUVD_LMI_SWAP_CNTL, lmi_swap_cntl); - WREG32_SOC15(UVD, 0, mmUVD_MP_SWAP_CNTL, mp_swap_cntl); - - WREG32_SOC15(UVD, 0, mmUVD_MPC_SET_MUXA0, 0x40c2040); - WREG32_SOC15(UVD, 0, mmUVD_MPC_SET_MUXA1, 0x0); - WREG32_SOC15(UVD, 0, mmUVD_MPC_SET_MUXB0, 0x40c2040); - WREG32_SOC15(UVD, 0, mmUVD_MPC_SET_MUXB1, 0x0); - WREG32_SOC15(UVD, 0, mmUVD_MPC_SET_ALU, 0); - WREG32_SOC15(UVD, 0, mmUVD_MPC_SET_MUX, 0x88); - - /* take all subblocks out of reset, except VCPU */ - WREG32_SOC15(UVD, 0, mmUVD_SOFT_RESET, - UVD_SOFT_RESET__VCPU_SOFT_RESET_MASK); - mdelay(5); + WREG32_SOC15(UVD, k, mmUVD_LMI_SWAP_CNTL, lmi_swap_cntl); + WREG32_SOC15(UVD, k, mmUVD_MP_SWAP_CNTL, mp_swap_cntl); - /* enable VCPU clock */ - WREG32_SOC15(UVD, 0, mmUVD_VCPU_CNTL, - UVD_VCPU_CNTL__CLK_EN_MASK); + WREG32_SOC15(UVD, k, mmUVD_MPC_SET_MUXA0, 0x40c2040); + WREG32_SOC15(UVD, k, mmUVD_MPC_SET_MUXA1, 0x0); + WREG32_SOC15(UVD, k, mmUVD_MPC_SET_MUXB0, 0x40c2040); + WREG32_SOC15(UVD, k, mmUVD_MPC_SET_MUXB1, 0x0); + WREG32_SOC15(UVD, k, mmUVD_MPC_SET_ALU, 0); + WREG32_SOC15(UVD, k, mmUVD_MPC_SET_MUX, 0x88); - /* enable UMC */ - WREG32_P(SOC15_REG_OFFSET(UVD, 0, mmUVD_LMI_CTRL2), 0, - ~UVD_LMI_CTRL2__STALL_ARB_UMC_MASK); + /* take all subblocks out of reset, except VCPU */ + WREG32_SOC15(UVD, k, mmUVD_SOFT_RESET, + UVD_SOFT_RESET__VCPU_SOFT_RESET_MASK); + mdelay(5); + + /* enable VCPU clock */ + WREG32_SOC15(UVD, k, mmUVD_VCPU_CNTL, + UVD_VCPU_CNTL__CLK_EN_MASK); + + /* enable UMC */ + WREG32_P(SOC15_REG_OFFSET(UVD, k, mmUVD_LMI_CTRL2), 0, + ~UVD_LMI_CTRL2__STALL_ARB_UMC_MASK); - /* boot up the VCPU */ - WREG32_SOC15(UVD, 0, mmUVD_SOFT_RESET, 0); - mdelay(10); + /* boot up the VCPU */ + WREG32_SOC15(UVD, k, mmUVD_SOFT_RESET, 0); + mdelay(10); - for (i = 0; i < 10; ++i) { - uint32_t status; + for (i = 0; i < 10; ++i) { + uint32_t status; - for (j = 0; j < 100; ++j) { - status = RREG32_SOC15(UVD, 0, mmUVD_STATUS); + for (j = 0; j < 100; ++j) { + status = RREG32_SOC15(UVD, k, mmUVD_STATUS); + if (status & 2) + break; + mdelay(10); + } + r = 0; if (status & 2) break; + + DRM_ERROR("UVD(%d) not responding, trying to reset the VCPU!!!\n", k); + WREG32_P(SOC15_REG_OFFSET(UVD, k, mmUVD_SOFT_RESET), + UVD_SOFT_RESET__VCPU_SOFT_RESET_MASK, + ~UVD_SOFT_RESET__VCPU_SOFT_RESET_MASK); mdelay(10); + WREG32_P(SOC15_REG_OFFSET(UVD, k, mmUVD_SOFT_RESET), 0, + ~UVD_SOFT_RESET__VCPU_SOFT_RESET_MASK); + mdelay(10); + r = -1; } - r = 0; - if (status & 2) - break; - - DRM_ERROR("UVD not responding, trying to reset the VCPU!!!\n"); - WREG32_P(SOC15_REG_OFFSET(UVD, 0, mmUVD_SOFT_RESET), - UVD_SOFT_RESET__VCPU_SOFT_RESET_MASK, - ~UVD_SOFT_RESET__VCPU_SOFT_RESET_MASK); - mdelay(10); - WREG32_P(SOC15_REG_OFFSET(UVD, 0, mmUVD_SOFT_RESET), 0, - ~UVD_SOFT_RESET__VCPU_SOFT_RESET_MASK); - mdelay(10); - r = -1; - } - if (r) { - DRM_ERROR("UVD not responding, giving up!!!\n"); - return r; - } - /* enable master interrupt */ - WREG32_P(SOC15_REG_OFFSET(UVD, 0, mmUVD_MASTINT_EN), - (UVD_MASTINT_EN__VCPU_EN_MASK|UVD_MASTINT_EN__SYS_EN_MASK), - ~(UVD_MASTINT_EN__VCPU_EN_MASK|UVD_MASTINT_EN__SYS_EN_MASK)); - - /* clear the bit 4 of UVD_STATUS */ - WREG32_P(SOC15_REG_OFFSET(UVD, 0, mmUVD_STATUS), 0, - ~(2 << UVD_STATUS__VCPU_REPORT__SHIFT)); - - /* force RBC into idle state */ - rb_bufsz = order_base_2(ring->ring_size); - tmp = REG_SET_FIELD(0, UVD_RBC_RB_CNTL, RB_BUFSZ, rb_bufsz); - tmp = REG_SET_FIELD(tmp, UVD_RBC_RB_CNTL, RB_BLKSZ, 1); - tmp = REG_SET_FIELD(tmp, UVD_RBC_RB_CNTL, RB_NO_FETCH, 1); - tmp = REG_SET_FIELD(tmp, UVD_RBC_RB_CNTL, RB_WPTR_POLL_EN, 0); - tmp = REG_SET_FIELD(tmp, UVD_RBC_RB_CNTL, RB_NO_UPDATE, 1); - tmp = REG_SET_FIELD(tmp, UVD_RBC_RB_CNTL, RB_RPTR_WR_EN, 1); - WREG32_SOC15(UVD, 0, mmUVD_RBC_RB_CNTL, tmp); - - /* set the write pointer delay */ - WREG32_SOC15(UVD, 0, mmUVD_RBC_RB_WPTR_CNTL, 0); - - /* set the wb address */ - WREG32_SOC15(UVD, 0, mmUVD_RBC_RB_RPTR_ADDR, - (upper_32_bits(ring->gpu_addr) >> 2)); - - /* programm the RB_BASE for ring buffer */ - WREG32_SOC15(UVD, 0, mmUVD_LMI_RBC_RB_64BIT_BAR_LOW, - lower_32_bits(ring->gpu_addr)); - WREG32_SOC15(UVD, 0, mmUVD_LMI_RBC_RB_64BIT_BAR_HIGH, - upper_32_bits(ring->gpu_addr)); - - /* Initialize the ring buffer's read and write pointers */ - WREG32_SOC15(UVD, 0, mmUVD_RBC_RB_RPTR, 0); - - ring->wptr = RREG32_SOC15(UVD, 0, mmUVD_RBC_RB_RPTR); - WREG32_SOC15(UVD, 0, mmUVD_RBC_RB_WPTR, - lower_32_bits(ring->wptr)); - - WREG32_P(SOC15_REG_OFFSET(UVD, 0, mmUVD_RBC_RB_CNTL), 0, - ~UVD_RBC_RB_CNTL__RB_NO_FETCH_MASK); - - ring = &adev->uvd.ring_enc[0]; - WREG32_SOC15(UVD, 0, mmUVD_RB_RPTR, lower_32_bits(ring->wptr)); - WREG32_SOC15(UVD, 0, mmUVD_RB_WPTR, lower_32_bits(ring->wptr)); - WREG32_SOC15(UVD, 0, mmUVD_RB_BASE_LO, ring->gpu_addr); - WREG32_SOC15(UVD, 0, mmUVD_RB_BASE_HI, upper_32_bits(ring->gpu_addr)); - WREG32_SOC15(UVD, 0, mmUVD_RB_SIZE, ring->ring_size / 4); + if (r) { + DRM_ERROR("UVD(%d) not responding, giving up!!!\n", k); + return r; + } + /* enable master interrupt */ + WREG32_P(SOC15_REG_OFFSET(UVD, k, mmUVD_MASTINT_EN), + (UVD_MASTINT_EN__VCPU_EN_MASK|UVD_MASTINT_EN__SYS_EN_MASK), + ~(UVD_MASTINT_EN__VCPU_EN_MASK|UVD_MASTINT_EN__SYS_EN_MASK)); - ring = &adev->uvd.ring_enc[1]; - WREG32_SOC15(UVD, 0, mmUVD_RB_RPTR2, lower_32_bits(ring->wptr)); - WREG32_SOC15(UVD, 0, mmUVD_RB_WPTR2, lower_32_bits(ring->wptr)); - WREG32_SOC15(UVD, 0, mmUVD_RB_BASE_LO2, ring->gpu_addr); - WREG32_SOC15(UVD, 0, mmUVD_RB_BASE_HI2, upper_32_bits(ring->gpu_addr)); - WREG32_SOC15(UVD, 0, mmUVD_RB_SIZE2, ring->ring_size / 4); + /* clear the bit 4 of UVD_STATUS */ + WREG32_P(SOC15_REG_OFFSET(UVD, k, mmUVD_STATUS), 0, + ~(2 << UVD_STATUS__VCPU_REPORT__SHIFT)); + /* force RBC into idle state */ + rb_bufsz = order_base_2(ring->ring_size); + tmp = REG_SET_FIELD(0, UVD_RBC_RB_CNTL, RB_BUFSZ, rb_bufsz); + tmp = REG_SET_FIELD(tmp, UVD_RBC_RB_CNTL, RB_BLKSZ, 1); + tmp = REG_SET_FIELD(tmp, UVD_RBC_RB_CNTL, RB_NO_FETCH, 1); + tmp = REG_SET_FIELD(tmp, UVD_RBC_RB_CNTL, RB_WPTR_POLL_EN, 0); + tmp = REG_SET_FIELD(tmp, UVD_RBC_RB_CNTL, RB_NO_UPDATE, 1); + tmp = REG_SET_FIELD(tmp, UVD_RBC_RB_CNTL, RB_RPTR_WR_EN, 1); + WREG32_SOC15(UVD, k, mmUVD_RBC_RB_CNTL, tmp); + + /* set the write pointer delay */ + WREG32_SOC15(UVD, k, mmUVD_RBC_RB_WPTR_CNTL, 0); + + /* set the wb address */ + WREG32_SOC15(UVD, k, mmUVD_RBC_RB_RPTR_ADDR, + (upper_32_bits(ring->gpu_addr) >> 2)); + + /* programm the RB_BASE for ring buffer */ + WREG32_SOC15(UVD, k, mmUVD_LMI_RBC_RB_64BIT_BAR_LOW, + lower_32_bits(ring->gpu_addr)); + WREG32_SOC15(UVD, k, mmUVD_LMI_RBC_RB_64BIT_BAR_HIGH, + upper_32_bits(ring->gpu_addr)); + + /* Initialize the ring buffer's read and write pointers */ + WREG32_SOC15(UVD, k, mmUVD_RBC_RB_RPTR, 0); + + ring->wptr = RREG32_SOC15(UVD, k, mmUVD_RBC_RB_RPTR); + WREG32_SOC15(UVD, k, mmUVD_RBC_RB_WPTR, + lower_32_bits(ring->wptr)); + + WREG32_P(SOC15_REG_OFFSET(UVD, k, mmUVD_RBC_RB_CNTL), 0, + ~UVD_RBC_RB_CNTL__RB_NO_FETCH_MASK); + + ring = &adev->uvd.inst[k].ring_enc[0]; + WREG32_SOC15(UVD, k, mmUVD_RB_RPTR, lower_32_bits(ring->wptr)); + WREG32_SOC15(UVD, k, mmUVD_RB_WPTR, lower_32_bits(ring->wptr)); + WREG32_SOC15(UVD, k, mmUVD_RB_BASE_LO, ring->gpu_addr); + WREG32_SOC15(UVD, k, mmUVD_RB_BASE_HI, upper_32_bits(ring->gpu_addr)); + WREG32_SOC15(UVD, k, mmUVD_RB_SIZE, ring->ring_size / 4); + + ring = &adev->uvd.inst[k].ring_enc[1]; + WREG32_SOC15(UVD, k, mmUVD_RB_RPTR2, lower_32_bits(ring->wptr)); + WREG32_SOC15(UVD, k, mmUVD_RB_WPTR2, lower_32_bits(ring->wptr)); + WREG32_SOC15(UVD, k, mmUVD_RB_BASE_LO2, ring->gpu_addr); + WREG32_SOC15(UVD, k, mmUVD_RB_BASE_HI2, upper_32_bits(ring->gpu_addr)); + WREG32_SOC15(UVD, k, mmUVD_RB_SIZE2, ring->ring_size / 4); + } return 0; } @@ -1053,26 +1087,30 @@ static int uvd_v7_0_start(struct amdgpu_device *adev) */ static void uvd_v7_0_stop(struct amdgpu_device *adev) { - /* force RBC into idle state */ - WREG32_SOC15(UVD, 0, mmUVD_RBC_RB_CNTL, 0x11010101); - - /* Stall UMC and register bus before resetting VCPU */ - WREG32_P(SOC15_REG_OFFSET(UVD, 0, mmUVD_LMI_CTRL2), - UVD_LMI_CTRL2__STALL_ARB_UMC_MASK, - ~UVD_LMI_CTRL2__STALL_ARB_UMC_MASK); - mdelay(1); - - /* put VCPU into reset */ - WREG32_SOC15(UVD, 0, mmUVD_SOFT_RESET, - UVD_SOFT_RESET__VCPU_SOFT_RESET_MASK); - mdelay(5); + uint8_t i = 0; - /* disable VCPU clock */ - WREG32_SOC15(UVD, 0, mmUVD_VCPU_CNTL, 0x0); + for (i = 0; i < adev->uvd.num_uvd_inst; ++i) { + /* force RBC into idle state */ + WREG32_SOC15(UVD, i, mmUVD_RBC_RB_CNTL, 0x11010101); + + /* Stall UMC and register bus before resetting VCPU */ + WREG32_P(SOC15_REG_OFFSET(UVD, i, mmUVD_LMI_CTRL2), + UVD_LMI_CTRL2__STALL_ARB_UMC_MASK, + ~UVD_LMI_CTRL2__STALL_ARB_UMC_MASK); + mdelay(1); + + /* put VCPU into reset */ + WREG32_SOC15(UVD, i, mmUVD_SOFT_RESET, + UVD_SOFT_RESET__VCPU_SOFT_RESET_MASK); + mdelay(5); - /* Unstall UMC and register bus */ - WREG32_P(SOC15_REG_OFFSET(UVD, 0, mmUVD_LMI_CTRL2), 0, - ~UVD_LMI_CTRL2__STALL_ARB_UMC_MASK); + /* disable VCPU clock */ + WREG32_SOC15(UVD, i, mmUVD_VCPU_CNTL, 0x0); + + /* Unstall UMC and register bus */ + WREG32_P(SOC15_REG_OFFSET(UVD, i, mmUVD_LMI_CTRL2), 0, + ~UVD_LMI_CTRL2__STALL_ARB_UMC_MASK); + } } /** @@ -1091,26 +1129,26 @@ static void uvd_v7_0_ring_emit_fence(struct amdgpu_ring *ring, u64 addr, u64 seq WARN_ON(flags & AMDGPU_FENCE_FLAG_64BIT); amdgpu_ring_write(ring, - PACKET0(SOC15_REG_OFFSET(UVD, 0, mmUVD_CONTEXT_ID), 0)); + PACKET0(SOC15_REG_OFFSET(UVD, ring->me, mmUVD_CONTEXT_ID), 0)); amdgpu_ring_write(ring, seq); amdgpu_ring_write(ring, - PACKET0(SOC15_REG_OFFSET(UVD, 0, mmUVD_GPCOM_VCPU_DATA0), 0)); + PACKET0(SOC15_REG_OFFSET(UVD, ring->me, mmUVD_GPCOM_VCPU_DATA0), 0)); amdgpu_ring_write(ring, addr & 0xffffffff); amdgpu_ring_write(ring, - PACKET0(SOC15_REG_OFFSET(UVD, 0, mmUVD_GPCOM_VCPU_DATA1), 0)); + PACKET0(SOC15_REG_OFFSET(UVD, ring->me, mmUVD_GPCOM_VCPU_DATA1), 0)); amdgpu_ring_write(ring, upper_32_bits(addr) & 0xff); amdgpu_ring_write(ring, - PACKET0(SOC15_REG_OFFSET(UVD, 0, mmUVD_GPCOM_VCPU_CMD), 0)); + PACKET0(SOC15_REG_OFFSET(UVD, ring->me, mmUVD_GPCOM_VCPU_CMD), 0)); amdgpu_ring_write(ring, 0); amdgpu_ring_write(ring, - PACKET0(SOC15_REG_OFFSET(UVD, 0, mmUVD_GPCOM_VCPU_DATA0), 0)); + PACKET0(SOC15_REG_OFFSET(UVD, ring->me, mmUVD_GPCOM_VCPU_DATA0), 0)); amdgpu_ring_write(ring, 0); amdgpu_ring_write(ring, - PACKET0(SOC15_REG_OFFSET(UVD, 0, mmUVD_GPCOM_VCPU_DATA1), 0)); + PACKET0(SOC15_REG_OFFSET(UVD, ring->me, mmUVD_GPCOM_VCPU_DATA1), 0)); amdgpu_ring_write(ring, 0); amdgpu_ring_write(ring, - PACKET0(SOC15_REG_OFFSET(UVD, 0, mmUVD_GPCOM_VCPU_CMD), 0)); + PACKET0(SOC15_REG_OFFSET(UVD, ring->me, mmUVD_GPCOM_VCPU_CMD), 0)); amdgpu_ring_write(ring, 2); } @@ -1159,30 +1197,30 @@ static int uvd_v7_0_ring_test_ring(struct amdgpu_ring *ring) unsigned i; int r; - WREG32_SOC15(UVD, 0, mmUVD_CONTEXT_ID, 0xCAFEDEAD); + WREG32_SOC15(UVD, ring->me, mmUVD_CONTEXT_ID, 0xCAFEDEAD); r = amdgpu_ring_alloc(ring, 3); if (r) { - DRM_ERROR("amdgpu: cp failed to lock ring %d (%d).\n", - ring->idx, r); + DRM_ERROR("amdgpu: (%d)cp failed to lock ring %d (%d).\n", + ring->me, ring->idx, r); return r; } amdgpu_ring_write(ring, - PACKET0(SOC15_REG_OFFSET(UVD, 0, mmUVD_CONTEXT_ID), 0)); + PACKET0(SOC15_REG_OFFSET(UVD, ring->me, mmUVD_CONTEXT_ID), 0)); amdgpu_ring_write(ring, 0xDEADBEEF); amdgpu_ring_commit(ring); for (i = 0; i < adev->usec_timeout; i++) { - tmp = RREG32_SOC15(UVD, 0, mmUVD_CONTEXT_ID); + tmp = RREG32_SOC15(UVD, ring->me, mmUVD_CONTEXT_ID); if (tmp == 0xDEADBEEF) break; DRM_UDELAY(1); } if (i < adev->usec_timeout) { - DRM_DEBUG("ring test on %d succeeded in %d usecs\n", - ring->idx, i); + DRM_DEBUG("(%d)ring test on %d succeeded in %d usecs\n", + ring->me, ring->idx, i); } else { - DRM_ERROR("amdgpu: ring %d test failed (0x%08X)\n", - ring->idx, tmp); + DRM_ERROR("(%d)amdgpu: ring %d test failed (0x%08X)\n", + ring->me, ring->idx, tmp); r = -EINVAL; } return r; @@ -1203,17 +1241,17 @@ static void uvd_v7_0_ring_emit_ib(struct amdgpu_ring *ring, struct amdgpu_device *adev = ring->adev; amdgpu_ring_write(ring, - PACKET0(SOC15_REG_OFFSET(UVD, 0, mmUVD_LMI_RBC_IB_VMID), 0)); + PACKET0(SOC15_REG_OFFSET(UVD, ring->me, mmUVD_LMI_RBC_IB_VMID), 0)); amdgpu_ring_write(ring, vmid); amdgpu_ring_write(ring, - PACKET0(SOC15_REG_OFFSET(UVD, 0, mmUVD_LMI_RBC_IB_64BIT_BAR_LOW), 0)); + PACKET0(SOC15_REG_OFFSET(UVD, ring->me, mmUVD_LMI_RBC_IB_64BIT_BAR_LOW), 0)); amdgpu_ring_write(ring, lower_32_bits(ib->gpu_addr)); amdgpu_ring_write(ring, - PACKET0(SOC15_REG_OFFSET(UVD, 0, mmUVD_LMI_RBC_IB_64BIT_BAR_HIGH), 0)); + PACKET0(SOC15_REG_OFFSET(UVD, ring->me, mmUVD_LMI_RBC_IB_64BIT_BAR_HIGH), 0)); amdgpu_ring_write(ring, upper_32_bits(ib->gpu_addr)); amdgpu_ring_write(ring, - PACKET0(SOC15_REG_OFFSET(UVD, 0, mmUVD_RBC_IB_SIZE), 0)); + PACKET0(SOC15_REG_OFFSET(UVD, ring->me, mmUVD_RBC_IB_SIZE), 0)); amdgpu_ring_write(ring, ib->length_dw); } @@ -1241,13 +1279,13 @@ static void uvd_v7_0_ring_emit_wreg(struct amdgpu_ring *ring, struct amdgpu_device *adev = ring->adev; amdgpu_ring_write(ring, - PACKET0(SOC15_REG_OFFSET(UVD, 0, mmUVD_GPCOM_VCPU_DATA0), 0)); + PACKET0(SOC15_REG_OFFSET(UVD, ring->me, mmUVD_GPCOM_VCPU_DATA0), 0)); amdgpu_ring_write(ring, reg << 2); amdgpu_ring_write(ring, - PACKET0(SOC15_REG_OFFSET(UVD, 0, mmUVD_GPCOM_VCPU_DATA1), 0)); + PACKET0(SOC15_REG_OFFSET(UVD, ring->me, mmUVD_GPCOM_VCPU_DATA1), 0)); amdgpu_ring_write(ring, val); amdgpu_ring_write(ring, - PACKET0(SOC15_REG_OFFSET(UVD, 0, mmUVD_GPCOM_VCPU_CMD), 0)); + PACKET0(SOC15_REG_OFFSET(UVD, ring->me, mmUVD_GPCOM_VCPU_CMD), 0)); amdgpu_ring_write(ring, 8); } @@ -1257,16 +1295,16 @@ static void uvd_v7_0_ring_emit_reg_wait(struct amdgpu_ring *ring, uint32_t reg, struct amdgpu_device *adev = ring->adev; amdgpu_ring_write(ring, - PACKET0(SOC15_REG_OFFSET(UVD, 0, mmUVD_GPCOM_VCPU_DATA0), 0)); + PACKET0(SOC15_REG_OFFSET(UVD, ring->me, mmUVD_GPCOM_VCPU_DATA0), 0)); amdgpu_ring_write(ring, reg << 2); amdgpu_ring_write(ring, - PACKET0(SOC15_REG_OFFSET(UVD, 0, mmUVD_GPCOM_VCPU_DATA1), 0)); + PACKET0(SOC15_REG_OFFSET(UVD, ring->me, mmUVD_GPCOM_VCPU_DATA1), 0)); amdgpu_ring_write(ring, val); amdgpu_ring_write(ring, - PACKET0(SOC15_REG_OFFSET(UVD, 0, mmUVD_GP_SCRATCH8), 0)); + PACKET0(SOC15_REG_OFFSET(UVD, ring->me, mmUVD_GP_SCRATCH8), 0)); amdgpu_ring_write(ring, mask); amdgpu_ring_write(ring, - PACKET0(SOC15_REG_OFFSET(UVD, 0, mmUVD_GPCOM_VCPU_CMD), 0)); + PACKET0(SOC15_REG_OFFSET(UVD, ring->me, mmUVD_GPCOM_VCPU_CMD), 0)); amdgpu_ring_write(ring, 12); } @@ -1287,12 +1325,15 @@ static void uvd_v7_0_ring_emit_vm_flush(struct amdgpu_ring *ring, static void uvd_v7_0_ring_insert_nop(struct amdgpu_ring *ring, uint32_t count) { - int i; struct amdgpu_device *adev = ring->adev; + int i; - for (i = 0; i < count; i++) - amdgpu_ring_write(ring, PACKET0(SOC15_REG_OFFSET(UVD, 0, mmUVD_NO_OP), 0)); + WARN_ON(ring->wptr % 2 || count % 2); + for (i = 0; i < count / 2; i++) { + amdgpu_ring_write(ring, PACKET0(SOC15_REG_OFFSET(UVD, ring->me, mmUVD_NO_OP), 0)); + amdgpu_ring_write(ring, 0); + } } static void uvd_v7_0_enc_ring_insert_end(struct amdgpu_ring *ring) @@ -1359,16 +1400,16 @@ static bool uvd_v7_0_check_soft_reset(void *handle) if (REG_GET_FIELD(tmp, SRBM_STATUS, UVD_RQ_PENDING) || REG_GET_FIELD(tmp, SRBM_STATUS, UVD_BUSY) || - (RREG32_SOC15(UVD, 0, mmUVD_STATUS) & + (RREG32_SOC15(UVD, ring->me, mmUVD_STATUS) & AMDGPU_UVD_STATUS_BUSY_MASK)) srbm_soft_reset = REG_SET_FIELD(srbm_soft_reset, SRBM_SOFT_RESET, SOFT_RESET_UVD, 1); if (srbm_soft_reset) { - adev->uvd.srbm_soft_reset = srbm_soft_reset; + adev->uvd.inst[ring->me].srbm_soft_reset = srbm_soft_reset; return true; } else { - adev->uvd.srbm_soft_reset = 0; + adev->uvd.inst[ring->me].srbm_soft_reset = 0; return false; } } @@ -1377,7 +1418,7 @@ static int uvd_v7_0_pre_soft_reset(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - if (!adev->uvd.srbm_soft_reset) + if (!adev->uvd.inst[ring->me].srbm_soft_reset) return 0; uvd_v7_0_stop(adev); @@ -1389,9 +1430,9 @@ static int uvd_v7_0_soft_reset(void *handle) struct amdgpu_device *adev = (struct amdgpu_device *)handle; u32 srbm_soft_reset; - if (!adev->uvd.srbm_soft_reset) + if (!adev->uvd.inst[ring->me].srbm_soft_reset) return 0; - srbm_soft_reset = adev->uvd.srbm_soft_reset; + srbm_soft_reset = adev->uvd.inst[ring->me].srbm_soft_reset; if (srbm_soft_reset) { u32 tmp; @@ -1419,7 +1460,7 @@ static int uvd_v7_0_post_soft_reset(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - if (!adev->uvd.srbm_soft_reset) + if (!adev->uvd.inst[ring->me].srbm_soft_reset) return 0; mdelay(5); @@ -1441,17 +1482,32 @@ static int uvd_v7_0_process_interrupt(struct amdgpu_device *adev, struct amdgpu_irq_src *source, struct amdgpu_iv_entry *entry) { + uint32_t ip_instance; + + switch (entry->client_id) { + case SOC15_IH_CLIENTID_UVD: + ip_instance = 0; + break; + case SOC15_IH_CLIENTID_UVD1: + ip_instance = 1; + break; + default: + DRM_ERROR("Unhandled client id: %d\n", entry->client_id); + return 0; + } + DRM_DEBUG("IH: UVD TRAP\n"); + switch (entry->src_id) { case 124: - amdgpu_fence_process(&adev->uvd.ring); + amdgpu_fence_process(&adev->uvd.inst[ip_instance].ring); break; case 119: - amdgpu_fence_process(&adev->uvd.ring_enc[0]); + amdgpu_fence_process(&adev->uvd.inst[ip_instance].ring_enc[0]); break; case 120: if (!amdgpu_sriov_vf(adev)) - amdgpu_fence_process(&adev->uvd.ring_enc[1]); + amdgpu_fence_process(&adev->uvd.inst[ip_instance].ring_enc[1]); break; default: DRM_ERROR("Unhandled interrupt: %d %d\n", @@ -1467,9 +1523,9 @@ static void uvd_v7_0_set_sw_clock_gating(struct amdgpu_device *adev) { uint32_t data, data1, data2, suvd_flags; - data = RREG32_SOC15(UVD, 0, mmUVD_CGC_CTRL); - data1 = RREG32_SOC15(UVD, 0, mmUVD_SUVD_CGC_GATE); - data2 = RREG32_SOC15(UVD, 0, mmUVD_SUVD_CGC_CTRL); + data = RREG32_SOC15(UVD, ring->me, mmUVD_CGC_CTRL); + data1 = RREG32_SOC15(UVD, ring->me, mmUVD_SUVD_CGC_GATE); + data2 = RREG32_SOC15(UVD, ring->me, mmUVD_SUVD_CGC_CTRL); data &= ~(UVD_CGC_CTRL__CLK_OFF_DELAY_MASK | UVD_CGC_CTRL__CLK_GATE_DLY_TIMER_MASK); @@ -1513,18 +1569,18 @@ static void uvd_v7_0_set_sw_clock_gating(struct amdgpu_device *adev) UVD_SUVD_CGC_CTRL__SDB_MODE_MASK); data1 |= suvd_flags; - WREG32_SOC15(UVD, 0, mmUVD_CGC_CTRL, data); - WREG32_SOC15(UVD, 0, mmUVD_CGC_GATE, 0); - WREG32_SOC15(UVD, 0, mmUVD_SUVD_CGC_GATE, data1); - WREG32_SOC15(UVD, 0, mmUVD_SUVD_CGC_CTRL, data2); + WREG32_SOC15(UVD, ring->me, mmUVD_CGC_CTRL, data); + WREG32_SOC15(UVD, ring->me, mmUVD_CGC_GATE, 0); + WREG32_SOC15(UVD, ring->me, mmUVD_SUVD_CGC_GATE, data1); + WREG32_SOC15(UVD, ring->me, mmUVD_SUVD_CGC_CTRL, data2); } static void uvd_v7_0_set_hw_clock_gating(struct amdgpu_device *adev) { uint32_t data, data1, cgc_flags, suvd_flags; - data = RREG32_SOC15(UVD, 0, mmUVD_CGC_GATE); - data1 = RREG32_SOC15(UVD, 0, mmUVD_SUVD_CGC_GATE); + data = RREG32_SOC15(UVD, ring->me, mmUVD_CGC_GATE); + data1 = RREG32_SOC15(UVD, ring->me, mmUVD_SUVD_CGC_GATE); cgc_flags = UVD_CGC_GATE__SYS_MASK | UVD_CGC_GATE__UDEC_MASK | @@ -1556,8 +1612,8 @@ static void uvd_v7_0_set_hw_clock_gating(struct amdgpu_device *adev) data |= cgc_flags; data1 |= suvd_flags; - WREG32_SOC15(UVD, 0, mmUVD_CGC_GATE, data); - WREG32_SOC15(UVD, 0, mmUVD_SUVD_CGC_GATE, data1); + WREG32_SOC15(UVD, ring->me, mmUVD_CGC_GATE, data); + WREG32_SOC15(UVD, ring->me, mmUVD_SUVD_CGC_GATE, data1); } static void uvd_v7_0_set_bypass_mode(struct amdgpu_device *adev, bool enable) @@ -1616,7 +1672,7 @@ static int uvd_v7_0_set_powergating_state(void *handle, if (!(adev->pg_flags & AMD_PG_SUPPORT_UVD)) return 0; - WREG32_SOC15(UVD, 0, mmUVD_POWER_STATUS, UVD_POWER_STATUS__UVD_PG_EN_MASK); + WREG32_SOC15(UVD, ring->me, mmUVD_POWER_STATUS, UVD_POWER_STATUS__UVD_PG_EN_MASK); if (state == AMD_PG_STATE_GATE) { uvd_v7_0_stop(adev); @@ -1657,7 +1713,6 @@ const struct amd_ip_funcs uvd_v7_0_ip_funcs = { static const struct amdgpu_ring_funcs uvd_v7_0_ring_vm_funcs = { .type = AMDGPU_RING_TYPE_UVD, .align_mask = 0xf, - .nop = PACKET0(0x81ff, 0), .support_64bit_ptrs = false, .vmhub = AMDGPU_MMHUB, .get_rptr = uvd_v7_0_ring_get_rptr, @@ -1719,18 +1774,27 @@ static const struct amdgpu_ring_funcs uvd_v7_0_enc_ring_vm_funcs = { static void uvd_v7_0_set_ring_funcs(struct amdgpu_device *adev) { - adev->uvd.ring.funcs = &uvd_v7_0_ring_vm_funcs; - DRM_INFO("UVD is enabled in VM mode\n"); + int i; + + for (i = 0; i < adev->uvd.num_uvd_inst; i++) { + adev->uvd.inst[i].ring.funcs = &uvd_v7_0_ring_vm_funcs; + adev->uvd.inst[i].ring.me = i; + DRM_INFO("UVD(%d) is enabled in VM mode\n", i); + } } static void uvd_v7_0_set_enc_ring_funcs(struct amdgpu_device *adev) { - int i; + int i, j; - for (i = 0; i < adev->uvd.num_enc_rings; ++i) - adev->uvd.ring_enc[i].funcs = &uvd_v7_0_enc_ring_vm_funcs; + for (j = 0; j < adev->uvd.num_uvd_inst; j++) { + for (i = 0; i < adev->uvd.num_enc_rings; ++i) { + adev->uvd.inst[j].ring_enc[i].funcs = &uvd_v7_0_enc_ring_vm_funcs; + adev->uvd.inst[j].ring_enc[i].me = j; + } - DRM_INFO("UVD ENC is enabled in VM mode\n"); + DRM_INFO("UVD(%d) ENC is enabled in VM mode\n", j); + } } static const struct amdgpu_irq_src_funcs uvd_v7_0_irq_funcs = { @@ -1740,8 +1804,12 @@ static const struct amdgpu_irq_src_funcs uvd_v7_0_irq_funcs = { static void uvd_v7_0_set_irq_funcs(struct amdgpu_device *adev) { - adev->uvd.irq.num_types = adev->uvd.num_enc_rings + 1; - adev->uvd.irq.funcs = &uvd_v7_0_irq_funcs; + int i; + + for (i = 0; i < adev->uvd.num_uvd_inst; i++) { + adev->uvd.inst[i].irq.num_types = adev->uvd.num_enc_rings + 1; + adev->uvd.inst[i].irq.funcs = &uvd_v7_0_irq_funcs; + } } const struct amdgpu_ip_block_version uvd_v7_0_ip_block = diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c index 0501746b6c2c..110b294ebed3 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c @@ -35,7 +35,6 @@ #include "mmhub/mmhub_9_1_offset.h" #include "mmhub/mmhub_9_1_sh_mask.h" -static int vcn_v1_0_start(struct amdgpu_device *adev); static int vcn_v1_0_stop(struct amdgpu_device *adev); static void vcn_v1_0_set_dec_ring_funcs(struct amdgpu_device *adev); static void vcn_v1_0_set_enc_ring_funcs(struct amdgpu_device *adev); @@ -146,10 +145,6 @@ static int vcn_v1_0_hw_init(void *handle) struct amdgpu_ring *ring = &adev->vcn.ring_dec; int i, r; - r = vcn_v1_0_start(adev); - if (r) - goto done; - ring->ready = true; r = amdgpu_ring_test_ring(ring); if (r) { @@ -185,11 +180,9 @@ static int vcn_v1_0_hw_fini(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; struct amdgpu_ring *ring = &adev->vcn.ring_dec; - int r; - r = vcn_v1_0_stop(adev); - if (r) - return r; + if (RREG32_SOC15(VCN, 0, mmUVD_STATUS)) + vcn_v1_0_stop(adev); ring->ready = false; @@ -288,14 +281,14 @@ static void vcn_v1_0_mc_resume(struct amdgpu_device *adev) * * Disable clock gating for VCN block */ -static void vcn_v1_0_disable_clock_gating(struct amdgpu_device *adev, bool sw) +static void vcn_v1_0_disable_clock_gating(struct amdgpu_device *adev) { uint32_t data; /* JPEG disable CGC */ data = RREG32_SOC15(VCN, 0, mmJPEG_CGC_CTRL); - if (sw) + if (adev->cg_flags & AMD_CG_SUPPORT_VCN_MGCG) data |= 1 << JPEG_CGC_CTRL__DYN_CLOCK_MODE__SHIFT; else data &= ~JPEG_CGC_CTRL__DYN_CLOCK_MODE_MASK; @@ -310,7 +303,7 @@ static void vcn_v1_0_disable_clock_gating(struct amdgpu_device *adev, bool sw) /* UVD disable CGC */ data = RREG32_SOC15(VCN, 0, mmUVD_CGC_CTRL); - if (sw) + if (adev->cg_flags & AMD_CG_SUPPORT_VCN_MGCG) data |= 1 << UVD_CGC_CTRL__DYN_CLOCK_MODE__SHIFT; else data &= ~ UVD_CGC_CTRL__DYN_CLOCK_MODE_MASK; @@ -415,13 +408,13 @@ static void vcn_v1_0_disable_clock_gating(struct amdgpu_device *adev, bool sw) * * Enable clock gating for VCN block */ -static void vcn_v1_0_enable_clock_gating(struct amdgpu_device *adev, bool sw) +static void vcn_v1_0_enable_clock_gating(struct amdgpu_device *adev) { uint32_t data = 0; /* enable JPEG CGC */ data = RREG32_SOC15(VCN, 0, mmJPEG_CGC_CTRL); - if (sw) + if (adev->cg_flags & AMD_CG_SUPPORT_VCN_MGCG) data |= 1 << JPEG_CGC_CTRL__DYN_CLOCK_MODE__SHIFT; else data |= 0 << JPEG_CGC_CTRL__DYN_CLOCK_MODE__SHIFT; @@ -435,7 +428,7 @@ static void vcn_v1_0_enable_clock_gating(struct amdgpu_device *adev, bool sw) /* enable UVD CGC */ data = RREG32_SOC15(VCN, 0, mmUVD_CGC_CTRL); - if (sw) + if (adev->cg_flags & AMD_CG_SUPPORT_VCN_MGCG) data |= 1 << UVD_CGC_CTRL__DYN_CLOCK_MODE__SHIFT; else data |= 0 << UVD_CGC_CTRL__DYN_CLOCK_MODE__SHIFT; @@ -480,6 +473,94 @@ static void vcn_v1_0_enable_clock_gating(struct amdgpu_device *adev, bool sw) WREG32_SOC15(VCN, 0, mmUVD_SUVD_CGC_CTRL, data); } +static void vcn_1_0_disable_static_power_gating(struct amdgpu_device *adev) +{ + uint32_t data = 0; + int ret; + + if (adev->pg_flags & AMD_PG_SUPPORT_VCN) { + data = (1 << UVD_PGFSM_CONFIG__UVDM_PWR_CONFIG__SHIFT + | 1 << UVD_PGFSM_CONFIG__UVDU_PWR_CONFIG__SHIFT + | 2 << UVD_PGFSM_CONFIG__UVDF_PWR_CONFIG__SHIFT + | 2 << UVD_PGFSM_CONFIG__UVDC_PWR_CONFIG__SHIFT + | 2 << UVD_PGFSM_CONFIG__UVDB_PWR_CONFIG__SHIFT + | 2 << UVD_PGFSM_CONFIG__UVDIL_PWR_CONFIG__SHIFT + | 2 << UVD_PGFSM_CONFIG__UVDIR_PWR_CONFIG__SHIFT + | 2 << UVD_PGFSM_CONFIG__UVDTD_PWR_CONFIG__SHIFT + | 2 << UVD_PGFSM_CONFIG__UVDTE_PWR_CONFIG__SHIFT + | 2 << UVD_PGFSM_CONFIG__UVDE_PWR_CONFIG__SHIFT + | 2 << UVD_PGFSM_CONFIG__UVDW_PWR_CONFIG__SHIFT); + + WREG32_SOC15(VCN, 0, mmUVD_PGFSM_CONFIG, data); + SOC15_WAIT_ON_RREG(VCN, 0, mmUVD_PGFSM_STATUS, UVD_PGFSM_STATUS__UVDM_UVDU_PWR_ON, 0xFFFFFF, ret); + } else { + data = (1 << UVD_PGFSM_CONFIG__UVDM_PWR_CONFIG__SHIFT + | 1 << UVD_PGFSM_CONFIG__UVDU_PWR_CONFIG__SHIFT + | 1 << UVD_PGFSM_CONFIG__UVDF_PWR_CONFIG__SHIFT + | 1 << UVD_PGFSM_CONFIG__UVDC_PWR_CONFIG__SHIFT + | 1 << UVD_PGFSM_CONFIG__UVDB_PWR_CONFIG__SHIFT + | 1 << UVD_PGFSM_CONFIG__UVDIL_PWR_CONFIG__SHIFT + | 1 << UVD_PGFSM_CONFIG__UVDIR_PWR_CONFIG__SHIFT + | 1 << UVD_PGFSM_CONFIG__UVDTD_PWR_CONFIG__SHIFT + | 1 << UVD_PGFSM_CONFIG__UVDTE_PWR_CONFIG__SHIFT + | 1 << UVD_PGFSM_CONFIG__UVDE_PWR_CONFIG__SHIFT + | 1 << UVD_PGFSM_CONFIG__UVDW_PWR_CONFIG__SHIFT); + WREG32_SOC15(VCN, 0, mmUVD_PGFSM_CONFIG, data); + SOC15_WAIT_ON_RREG(VCN, 0, mmUVD_PGFSM_STATUS, 0, 0xFFFFFFFF, ret); + } + + /* polling UVD_PGFSM_STATUS to confirm UVDM_PWR_STATUS , UVDU_PWR_STATUS are 0 (power on) */ + + data = RREG32_SOC15(VCN, 0, mmUVD_POWER_STATUS); + data &= ~0x103; + if (adev->pg_flags & AMD_PG_SUPPORT_VCN) + data |= UVD_PGFSM_CONFIG__UVDM_UVDU_PWR_ON | UVD_POWER_STATUS__UVD_PG_EN_MASK; + + WREG32_SOC15(VCN, 0, mmUVD_POWER_STATUS, data); +} + +static void vcn_1_0_enable_static_power_gating(struct amdgpu_device *adev) +{ + uint32_t data = 0; + int ret; + + if (adev->pg_flags & AMD_PG_SUPPORT_VCN) { + /* Before power off, this indicator has to be turned on */ + data = RREG32_SOC15(VCN, 0, mmUVD_POWER_STATUS); + data &= ~UVD_POWER_STATUS__UVD_POWER_STATUS_MASK; + data |= UVD_POWER_STATUS__UVD_POWER_STATUS_TILES_OFF; + WREG32_SOC15(VCN, 0, mmUVD_POWER_STATUS, data); + + + data = (2 << UVD_PGFSM_CONFIG__UVDM_PWR_CONFIG__SHIFT + | 2 << UVD_PGFSM_CONFIG__UVDU_PWR_CONFIG__SHIFT + | 2 << UVD_PGFSM_CONFIG__UVDF_PWR_CONFIG__SHIFT + | 2 << UVD_PGFSM_CONFIG__UVDC_PWR_CONFIG__SHIFT + | 2 << UVD_PGFSM_CONFIG__UVDB_PWR_CONFIG__SHIFT + | 2 << UVD_PGFSM_CONFIG__UVDIL_PWR_CONFIG__SHIFT + | 2 << UVD_PGFSM_CONFIG__UVDIR_PWR_CONFIG__SHIFT + | 2 << UVD_PGFSM_CONFIG__UVDTD_PWR_CONFIG__SHIFT + | 2 << UVD_PGFSM_CONFIG__UVDTE_PWR_CONFIG__SHIFT + | 2 << UVD_PGFSM_CONFIG__UVDE_PWR_CONFIG__SHIFT + | 2 << UVD_PGFSM_CONFIG__UVDW_PWR_CONFIG__SHIFT); + + WREG32_SOC15(VCN, 0, mmUVD_PGFSM_CONFIG, data); + + data = (2 << UVD_PGFSM_STATUS__UVDM_PWR_STATUS__SHIFT + | 2 << UVD_PGFSM_STATUS__UVDU_PWR_STATUS__SHIFT + | 2 << UVD_PGFSM_STATUS__UVDF_PWR_STATUS__SHIFT + | 2 << UVD_PGFSM_STATUS__UVDC_PWR_STATUS__SHIFT + | 2 << UVD_PGFSM_STATUS__UVDB_PWR_STATUS__SHIFT + | 2 << UVD_PGFSM_STATUS__UVDIL_PWR_STATUS__SHIFT + | 2 << UVD_PGFSM_STATUS__UVDIR_PWR_STATUS__SHIFT + | 2 << UVD_PGFSM_STATUS__UVDTD_PWR_STATUS__SHIFT + | 2 << UVD_PGFSM_STATUS__UVDTE_PWR_STATUS__SHIFT + | 2 << UVD_PGFSM_STATUS__UVDE_PWR_STATUS__SHIFT + | 2 << UVD_PGFSM_STATUS__UVDW_PWR_STATUS__SHIFT); + SOC15_WAIT_ON_RREG(VCN, 0, mmUVD_PGFSM_STATUS, data, 0xFFFFFFFF, ret); + } +} + /** * vcn_v1_0_start - start VCN block * @@ -499,8 +580,9 @@ static int vcn_v1_0_start(struct amdgpu_device *adev) vcn_v1_0_mc_resume(adev); + vcn_1_0_disable_static_power_gating(adev); /* disable clock gating */ - vcn_v1_0_disable_clock_gating(adev, true); + vcn_v1_0_disable_clock_gating(adev); /* disable interupt */ WREG32_P(SOC15_REG_OFFSET(UVD, 0, mmUVD_MASTINT_EN), 0, @@ -680,16 +762,45 @@ static int vcn_v1_0_stop(struct amdgpu_device *adev) WREG32_P(SOC15_REG_OFFSET(UVD, 0, mmUVD_LMI_CTRL2), 0, ~UVD_LMI_CTRL2__STALL_ARB_UMC_MASK); - /* enable clock gating */ - vcn_v1_0_enable_clock_gating(adev, true); + WREG32_SOC15(VCN, 0, mmUVD_STATUS, 0); + vcn_v1_0_enable_clock_gating(adev); + vcn_1_0_enable_static_power_gating(adev); return 0; } +bool vcn_v1_0_is_idle(void *handle) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + + return (RREG32_SOC15(VCN, 0, mmUVD_STATUS) == 0x2); +} + +int vcn_v1_0_wait_for_idle(void *handle) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + int ret = 0; + + SOC15_WAIT_ON_RREG(VCN, 0, mmUVD_STATUS, 0x2, 0x2, ret); + + return ret; +} + static int vcn_v1_0_set_clockgating_state(void *handle, enum amd_clockgating_state state) { - /* needed for driver unload*/ + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + bool enable = (state == AMD_CG_STATE_GATE) ? true : false; + + if (enable) { + /* wait for STATUS to clear */ + if (vcn_v1_0_is_idle(handle)) + return -EBUSY; + vcn_v1_0_enable_clock_gating(adev); + } else { + /* disable HW gating and enable Sw gating */ + vcn_v1_0_disable_clock_gating(adev); + } return 0; } @@ -1048,16 +1159,36 @@ static int vcn_v1_0_process_interrupt(struct amdgpu_device *adev, return 0; } -static void vcn_v1_0_ring_insert_nop(struct amdgpu_ring *ring, uint32_t count) +static void vcn_v1_0_dec_ring_insert_nop(struct amdgpu_ring *ring, uint32_t count) { - int i; struct amdgpu_device *adev = ring->adev; + int i; - for (i = 0; i < count; i++) - amdgpu_ring_write(ring, PACKET0(SOC15_REG_OFFSET(UVD, 0, mmUVD_NO_OP), 0)); + WARN_ON(ring->wptr % 2 || count % 2); + for (i = 0; i < count / 2; i++) { + amdgpu_ring_write(ring, PACKET0(SOC15_REG_OFFSET(UVD, 0, mmUVD_NO_OP), 0)); + amdgpu_ring_write(ring, 0); + } } +static int vcn_v1_0_set_powergating_state(void *handle, + enum amd_powergating_state state) +{ + /* This doesn't actually powergate the VCN block. + * That's done in the dpm code via the SMC. This + * just re-inits the block as necessary. The actual + * gating still happens in the dpm code. We should + * revisit this when there is a cleaner line between + * the smc and the hw blocks + */ + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + + if (state == AMD_PG_STATE_GATE) + return vcn_v1_0_stop(adev); + else + return vcn_v1_0_start(adev); +} static const struct amd_ip_funcs vcn_v1_0_ip_funcs = { .name = "vcn_v1_0", @@ -1069,20 +1200,19 @@ static const struct amd_ip_funcs vcn_v1_0_ip_funcs = { .hw_fini = vcn_v1_0_hw_fini, .suspend = vcn_v1_0_suspend, .resume = vcn_v1_0_resume, - .is_idle = NULL /* vcn_v1_0_is_idle */, - .wait_for_idle = NULL /* vcn_v1_0_wait_for_idle */, + .is_idle = vcn_v1_0_is_idle, + .wait_for_idle = vcn_v1_0_wait_for_idle, .check_soft_reset = NULL /* vcn_v1_0_check_soft_reset */, .pre_soft_reset = NULL /* vcn_v1_0_pre_soft_reset */, .soft_reset = NULL /* vcn_v1_0_soft_reset */, .post_soft_reset = NULL /* vcn_v1_0_post_soft_reset */, .set_clockgating_state = vcn_v1_0_set_clockgating_state, - .set_powergating_state = NULL /* vcn_v1_0_set_powergating_state */, + .set_powergating_state = vcn_v1_0_set_powergating_state, }; static const struct amdgpu_ring_funcs vcn_v1_0_dec_ring_vm_funcs = { .type = AMDGPU_RING_TYPE_VCN_DEC, .align_mask = 0xf, - .nop = PACKET0(0x81ff, 0), .support_64bit_ptrs = false, .vmhub = AMDGPU_MMHUB, .get_rptr = vcn_v1_0_dec_ring_get_rptr, @@ -1101,7 +1231,7 @@ static const struct amdgpu_ring_funcs vcn_v1_0_dec_ring_vm_funcs = { .emit_vm_flush = vcn_v1_0_dec_ring_emit_vm_flush, .test_ring = amdgpu_vcn_dec_ring_test_ring, .test_ib = amdgpu_vcn_dec_ring_test_ib, - .insert_nop = vcn_v1_0_ring_insert_nop, + .insert_nop = vcn_v1_0_dec_ring_insert_nop, .insert_start = vcn_v1_0_dec_ring_insert_start, .insert_end = vcn_v1_0_dec_ring_insert_end, .pad_ib = amdgpu_ring_generic_pad_ib, diff --git a/drivers/gpu/drm/amd/amdgpu/vega20_reg_init.c b/drivers/gpu/drm/amd/amdgpu/vega20_reg_init.c new file mode 100644 index 000000000000..52778de93ab0 --- /dev/null +++ b/drivers/gpu/drm/amd/amdgpu/vega20_reg_init.c @@ -0,0 +1,53 @@ +/* + * Copyright 2018 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + */ +#include "amdgpu.h" +#include "soc15.h" + +#include "soc15_common.h" +#include "soc15_hw_ip.h" +#include "vega20_ip_offset.h" + +int vega20_reg_base_init(struct amdgpu_device *adev) +{ + /* HW has more IP blocks, only initialized the blocke beend by our driver */ + uint32_t i; + for (i = 0 ; i < MAX_INSTANCE ; ++i) { + adev->reg_offset[GC_HWIP][i] = (uint32_t *)(&(GC_BASE.instance[i])); + adev->reg_offset[HDP_HWIP][i] = (uint32_t *)(&(HDP_BASE.instance[i])); + adev->reg_offset[MMHUB_HWIP][i] = (uint32_t *)(&(MMHUB_BASE.instance[i])); + adev->reg_offset[ATHUB_HWIP][i] = (uint32_t *)(&(ATHUB_BASE.instance[i])); + adev->reg_offset[NBIO_HWIP][i] = (uint32_t *)(&(NBIO_BASE.instance[i])); + adev->reg_offset[MP0_HWIP][i] = (uint32_t *)(&(MP0_BASE.instance[i])); + adev->reg_offset[UVD_HWIP][i] = (uint32_t *)(&(UVD_BASE.instance[i])); + adev->reg_offset[VCE_HWIP][i] = (uint32_t *)(&(VCE_BASE.instance[i])); + adev->reg_offset[DF_HWIP][i] = (uint32_t *)(&(DF_BASE.instance[i])); + adev->reg_offset[DCE_HWIP][i] = (uint32_t *)(&(DCE_BASE.instance[i])); + adev->reg_offset[OSSSYS_HWIP][i] = (uint32_t *)(&(OSSSYS_BASE.instance[i])); + adev->reg_offset[SDMA0_HWIP][i] = (uint32_t *)(&(SDMA0_BASE.instance[i])); + adev->reg_offset[SDMA1_HWIP][i] = (uint32_t *)(&(SDMA1_BASE.instance[i])); + adev->reg_offset[SMUIO_HWIP][i] = (uint32_t *)(&(SMUIO_BASE.instance[i])); + } + return 0; +} + + diff --git a/drivers/gpu/drm/amd/display/Kconfig b/drivers/gpu/drm/amd/display/Kconfig index e6ca72c0d347..d5d4586e6176 100644 --- a/drivers/gpu/drm/amd/display/Kconfig +++ b/drivers/gpu/drm/amd/display/Kconfig @@ -34,10 +34,4 @@ config DEBUG_KERNEL_DC if you want to hit kdgb_break in assert. -config DRM_AMD_DC_VEGAM - bool "VEGAM support" - depends on DRM_AMD_DC - help - Choose this option if you want to have - VEGAM support for display engine endmenu diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c index f2f54a9df56f..1ce10bc2d37b 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c @@ -911,6 +911,7 @@ amdgpu_dm_update_connector_after_detect(struct amdgpu_dm_connector *aconnector) drm_mode_connector_update_edid_property(connector, NULL); aconnector->num_modes = 0; aconnector->dc_sink = NULL; + aconnector->edid = NULL; } mutex_unlock(&dev->mode_config.mutex); @@ -1115,6 +1116,7 @@ static int dce110_register_irq_handlers(struct amdgpu_device *adev) if (adev->asic_type == CHIP_VEGA10 || adev->asic_type == CHIP_VEGA12 || + adev->asic_type == CHIP_VEGA20 || adev->asic_type == CHIP_RAVEN) client_id = SOC15_IH_CLIENTID_DCE; @@ -1513,11 +1515,10 @@ static int amdgpu_dm_initialize_drm_device(struct amdgpu_device *adev) case CHIP_POLARIS11: case CHIP_POLARIS10: case CHIP_POLARIS12: -#if defined(CONFIG_DRM_AMD_DC_VEGAM) case CHIP_VEGAM: -#endif case CHIP_VEGA10: case CHIP_VEGA12: + case CHIP_VEGA20: if (dce110_register_irq_handlers(dm->adev)) { DRM_ERROR("DM: Failed to initialize IRQ\n"); goto fail; @@ -1708,9 +1709,7 @@ static int dm_early_init(void *handle) adev->mode_info.plane_type = dm_plane_type_default; break; case CHIP_POLARIS10: -#if defined(CONFIG_DRM_AMD_DC_VEGAM) case CHIP_VEGAM: -#endif adev->mode_info.num_crtc = 6; adev->mode_info.num_hpd = 6; adev->mode_info.num_dig = 6; @@ -1718,6 +1717,7 @@ static int dm_early_init(void *handle) break; case CHIP_VEGA10: case CHIP_VEGA12: + case CHIP_VEGA20: adev->mode_info.num_crtc = 6; adev->mode_info.num_hpd = 6; adev->mode_info.num_dig = 6; @@ -1966,6 +1966,7 @@ static int fill_plane_attributes_from_fb(struct amdgpu_device *adev, if (adev->asic_type == CHIP_VEGA10 || adev->asic_type == CHIP_VEGA12 || + adev->asic_type == CHIP_VEGA20 || adev->asic_type == CHIP_RAVEN) { /* Fill GFX9 params */ plane_state->tiling_info.gfx9.num_pipes = diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c index e3d90e918d1b..b329393307e5 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c @@ -88,9 +88,9 @@ static void __drm_lut_to_dc_gamma(struct drm_color_lut *lut, g = drm_color_lut_extract(lut[i].green, 16); b = drm_color_lut_extract(lut[i].blue, 16); - gamma->entries.red[i] = dal_fixed31_32_from_int(r); - gamma->entries.green[i] = dal_fixed31_32_from_int(g); - gamma->entries.blue[i] = dal_fixed31_32_from_int(b); + gamma->entries.red[i] = dc_fixpt_from_int(r); + gamma->entries.green[i] = dc_fixpt_from_int(g); + gamma->entries.blue[i] = dc_fixpt_from_int(b); } return; } @@ -101,9 +101,9 @@ static void __drm_lut_to_dc_gamma(struct drm_color_lut *lut, g = drm_color_lut_extract(lut[i].green, 16); b = drm_color_lut_extract(lut[i].blue, 16); - gamma->entries.red[i] = dal_fixed31_32_from_fraction(r, MAX_DRM_LUT_VALUE); - gamma->entries.green[i] = dal_fixed31_32_from_fraction(g, MAX_DRM_LUT_VALUE); - gamma->entries.blue[i] = dal_fixed31_32_from_fraction(b, MAX_DRM_LUT_VALUE); + gamma->entries.red[i] = dc_fixpt_from_fraction(r, MAX_DRM_LUT_VALUE); + gamma->entries.green[i] = dc_fixpt_from_fraction(g, MAX_DRM_LUT_VALUE); + gamma->entries.blue[i] = dc_fixpt_from_fraction(b, MAX_DRM_LUT_VALUE); } } @@ -208,7 +208,7 @@ void amdgpu_dm_set_ctm(struct dm_crtc_state *crtc) for (i = 0; i < 12; i++) { /* Skip 4th element */ if (i % 4 == 3) { - stream->gamut_remap_matrix.matrix[i] = dal_fixed31_32_zero; + stream->gamut_remap_matrix.matrix[i] = dc_fixpt_zero; continue; } diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c index ca0b08bfa2cf..bd449351803f 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c @@ -330,11 +330,6 @@ bool dm_helpers_dp_mst_send_payload_allocation( return true; } -bool dm_helpers_dc_conn_log(struct dc_context *ctx, struct log_entry *entry, enum dc_log_type event) -{ - return true; -} - void dm_dtn_log_begin(struct dc_context *ctx) {} diff --git a/drivers/gpu/drm/amd/display/dc/basics/Makefile b/drivers/gpu/drm/amd/display/dc/basics/Makefile index bca33bd9a0d2..b49ea96b5dae 100644 --- a/drivers/gpu/drm/amd/display/dc/basics/Makefile +++ b/drivers/gpu/drm/amd/display/dc/basics/Makefile @@ -24,7 +24,7 @@ # It provides the general basic services required by other DAL # subcomponents. -BASICS = conversion.o fixpt31_32.o fixpt32_32.o \ +BASICS = conversion.o fixpt31_32.o \ logger.o log_helpers.o vector.o AMD_DAL_BASICS = $(addprefix $(AMDDALPATH)/dc/basics/,$(BASICS)) diff --git a/drivers/gpu/drm/amd/display/dc/basics/conversion.c b/drivers/gpu/drm/amd/display/dc/basics/conversion.c index 310964915a83..50b47f11875c 100644 --- a/drivers/gpu/drm/amd/display/dc/basics/conversion.c +++ b/drivers/gpu/drm/amd/display/dc/basics/conversion.c @@ -41,22 +41,22 @@ uint16_t fixed_point_to_int_frac( uint16_t result; - uint16_t d = (uint16_t)dal_fixed31_32_floor( - dal_fixed31_32_abs( + uint16_t d = (uint16_t)dc_fixpt_floor( + dc_fixpt_abs( arg)); if (d <= (uint16_t)(1 << integer_bits) - (1 / (uint16_t)divisor)) - numerator = (uint16_t)dal_fixed31_32_round( - dal_fixed31_32_mul_int( + numerator = (uint16_t)dc_fixpt_round( + dc_fixpt_mul_int( arg, divisor)); else { - numerator = dal_fixed31_32_floor( - dal_fixed31_32_sub( - dal_fixed31_32_from_int( + numerator = dc_fixpt_floor( + dc_fixpt_sub( + dc_fixpt_from_int( 1LL << integer_bits), - dal_fixed31_32_recip( - dal_fixed31_32_from_int( + dc_fixpt_recip( + dc_fixpt_from_int( divisor)))); } @@ -66,8 +66,8 @@ uint16_t fixed_point_to_int_frac( result = (uint16_t)( (1 << (integer_bits + fractional_bits + 1)) + numerator); - if ((result != 0) && dal_fixed31_32_lt( - arg, dal_fixed31_32_zero)) + if ((result != 0) && dc_fixpt_lt( + arg, dc_fixpt_zero)) result |= 1 << (integer_bits + fractional_bits); return result; @@ -84,15 +84,15 @@ void convert_float_matrix( uint32_t buffer_size) { const struct fixed31_32 min_2_13 = - dal_fixed31_32_from_fraction(S2D13_MIN, DIVIDER); + dc_fixpt_from_fraction(S2D13_MIN, DIVIDER); const struct fixed31_32 max_2_13 = - dal_fixed31_32_from_fraction(S2D13_MAX, DIVIDER); + dc_fixpt_from_fraction(S2D13_MAX, DIVIDER); uint32_t i; for (i = 0; i < buffer_size; ++i) { uint32_t reg_value = fixed_point_to_int_frac( - dal_fixed31_32_clamp( + dc_fixpt_clamp( flt[i], min_2_13, max_2_13), diff --git a/drivers/gpu/drm/amd/display/dc/basics/fixpt31_32.c b/drivers/gpu/drm/amd/display/dc/basics/fixpt31_32.c index 7191c3213743..e61dd97d0928 100644 --- a/drivers/gpu/drm/amd/display/dc/basics/fixpt31_32.c +++ b/drivers/gpu/drm/amd/display/dc/basics/fixpt31_32.c @@ -64,9 +64,7 @@ static inline unsigned long long complete_integer_division_u64( #define GET_FRACTIONAL_PART(x) \ (FRACTIONAL_PART_MASK & (x)) -struct fixed31_32 dal_fixed31_32_from_fraction( - long long numerator, - long long denominator) +struct fixed31_32 dc_fixpt_from_fraction(long long numerator, long long denominator) { struct fixed31_32 res; @@ -118,63 +116,7 @@ struct fixed31_32 dal_fixed31_32_from_fraction( return res; } -struct fixed31_32 dal_fixed31_32_from_int_nonconst( - long long arg) -{ - struct fixed31_32 res; - - ASSERT((LONG_MIN <= arg) && (arg <= LONG_MAX)); - - res.value = arg << FIXED31_32_BITS_PER_FRACTIONAL_PART; - - return res; -} - -struct fixed31_32 dal_fixed31_32_shl( - struct fixed31_32 arg, - unsigned char shift) -{ - struct fixed31_32 res; - - ASSERT(((arg.value >= 0) && (arg.value <= LLONG_MAX >> shift)) || - ((arg.value < 0) && (arg.value >= LLONG_MIN >> shift))); - - res.value = arg.value << shift; - - return res; -} - -struct fixed31_32 dal_fixed31_32_add( - struct fixed31_32 arg1, - struct fixed31_32 arg2) -{ - struct fixed31_32 res; - - ASSERT(((arg1.value >= 0) && (LLONG_MAX - arg1.value >= arg2.value)) || - ((arg1.value < 0) && (LLONG_MIN - arg1.value <= arg2.value))); - - res.value = arg1.value + arg2.value; - - return res; -} - -struct fixed31_32 dal_fixed31_32_sub( - struct fixed31_32 arg1, - struct fixed31_32 arg2) -{ - struct fixed31_32 res; - - ASSERT(((arg2.value >= 0) && (LLONG_MIN + arg2.value <= arg1.value)) || - ((arg2.value < 0) && (LLONG_MAX + arg2.value >= arg1.value))); - - res.value = arg1.value - arg2.value; - - return res; -} - -struct fixed31_32 dal_fixed31_32_mul( - struct fixed31_32 arg1, - struct fixed31_32 arg2) +struct fixed31_32 dc_fixpt_mul(struct fixed31_32 arg1, struct fixed31_32 arg2) { struct fixed31_32 res; @@ -213,7 +155,7 @@ struct fixed31_32 dal_fixed31_32_mul( tmp = arg1_fra * arg2_fra; tmp = (tmp >> FIXED31_32_BITS_PER_FRACTIONAL_PART) + - (tmp >= (unsigned long long)dal_fixed31_32_half.value); + (tmp >= (unsigned long long)dc_fixpt_half.value); ASSERT(tmp <= (unsigned long long)(LLONG_MAX - res.value)); @@ -225,8 +167,7 @@ struct fixed31_32 dal_fixed31_32_mul( return res; } -struct fixed31_32 dal_fixed31_32_sqr( - struct fixed31_32 arg) +struct fixed31_32 dc_fixpt_sqr(struct fixed31_32 arg) { struct fixed31_32 res; @@ -257,7 +198,7 @@ struct fixed31_32 dal_fixed31_32_sqr( tmp = arg_fra * arg_fra; tmp = (tmp >> FIXED31_32_BITS_PER_FRACTIONAL_PART) + - (tmp >= (unsigned long long)dal_fixed31_32_half.value); + (tmp >= (unsigned long long)dc_fixpt_half.value); ASSERT(tmp <= (unsigned long long)(LLONG_MAX - res.value)); @@ -266,8 +207,7 @@ struct fixed31_32 dal_fixed31_32_sqr( return res; } -struct fixed31_32 dal_fixed31_32_recip( - struct fixed31_32 arg) +struct fixed31_32 dc_fixpt_recip(struct fixed31_32 arg) { /* * @note @@ -276,41 +216,40 @@ struct fixed31_32 dal_fixed31_32_recip( ASSERT(arg.value); - return dal_fixed31_32_from_fraction( - dal_fixed31_32_one.value, + return dc_fixpt_from_fraction( + dc_fixpt_one.value, arg.value); } -struct fixed31_32 dal_fixed31_32_sinc( - struct fixed31_32 arg) +struct fixed31_32 dc_fixpt_sinc(struct fixed31_32 arg) { struct fixed31_32 square; - struct fixed31_32 res = dal_fixed31_32_one; + struct fixed31_32 res = dc_fixpt_one; int n = 27; struct fixed31_32 arg_norm = arg; - if (dal_fixed31_32_le( - dal_fixed31_32_two_pi, - dal_fixed31_32_abs(arg))) { - arg_norm = dal_fixed31_32_sub( + if (dc_fixpt_le( + dc_fixpt_two_pi, + dc_fixpt_abs(arg))) { + arg_norm = dc_fixpt_sub( arg_norm, - dal_fixed31_32_mul_int( - dal_fixed31_32_two_pi, + dc_fixpt_mul_int( + dc_fixpt_two_pi, (int)div64_s64( arg_norm.value, - dal_fixed31_32_two_pi.value))); + dc_fixpt_two_pi.value))); } - square = dal_fixed31_32_sqr(arg_norm); + square = dc_fixpt_sqr(arg_norm); do { - res = dal_fixed31_32_sub( - dal_fixed31_32_one, - dal_fixed31_32_div_int( - dal_fixed31_32_mul( + res = dc_fixpt_sub( + dc_fixpt_one, + dc_fixpt_div_int( + dc_fixpt_mul( square, res), n * (n - 1))); @@ -319,37 +258,35 @@ struct fixed31_32 dal_fixed31_32_sinc( } while (n > 2); if (arg.value != arg_norm.value) - res = dal_fixed31_32_div( - dal_fixed31_32_mul(res, arg_norm), + res = dc_fixpt_div( + dc_fixpt_mul(res, arg_norm), arg); return res; } -struct fixed31_32 dal_fixed31_32_sin( - struct fixed31_32 arg) +struct fixed31_32 dc_fixpt_sin(struct fixed31_32 arg) { - return dal_fixed31_32_mul( + return dc_fixpt_mul( arg, - dal_fixed31_32_sinc(arg)); + dc_fixpt_sinc(arg)); } -struct fixed31_32 dal_fixed31_32_cos( - struct fixed31_32 arg) +struct fixed31_32 dc_fixpt_cos(struct fixed31_32 arg) { /* TODO implement argument normalization */ - const struct fixed31_32 square = dal_fixed31_32_sqr(arg); + const struct fixed31_32 square = dc_fixpt_sqr(arg); - struct fixed31_32 res = dal_fixed31_32_one; + struct fixed31_32 res = dc_fixpt_one; int n = 26; do { - res = dal_fixed31_32_sub( - dal_fixed31_32_one, - dal_fixed31_32_div_int( - dal_fixed31_32_mul( + res = dc_fixpt_sub( + dc_fixpt_one, + dc_fixpt_div_int( + dc_fixpt_mul( square, res), n * (n - 1))); @@ -367,37 +304,35 @@ struct fixed31_32 dal_fixed31_32_cos( * * Calculated as Taylor series. */ -static struct fixed31_32 fixed31_32_exp_from_taylor_series( - struct fixed31_32 arg) +static struct fixed31_32 fixed31_32_exp_from_taylor_series(struct fixed31_32 arg) { unsigned int n = 9; - struct fixed31_32 res = dal_fixed31_32_from_fraction( + struct fixed31_32 res = dc_fixpt_from_fraction( n + 2, n + 1); /* TODO find correct res */ - ASSERT(dal_fixed31_32_lt(arg, dal_fixed31_32_one)); + ASSERT(dc_fixpt_lt(arg, dc_fixpt_one)); do - res = dal_fixed31_32_add( - dal_fixed31_32_one, - dal_fixed31_32_div_int( - dal_fixed31_32_mul( + res = dc_fixpt_add( + dc_fixpt_one, + dc_fixpt_div_int( + dc_fixpt_mul( arg, res), n)); while (--n != 1); - return dal_fixed31_32_add( - dal_fixed31_32_one, - dal_fixed31_32_mul( + return dc_fixpt_add( + dc_fixpt_one, + dc_fixpt_mul( arg, res)); } -struct fixed31_32 dal_fixed31_32_exp( - struct fixed31_32 arg) +struct fixed31_32 dc_fixpt_exp(struct fixed31_32 arg) { /* * @brief @@ -406,44 +341,43 @@ struct fixed31_32 dal_fixed31_32_exp( * where m = round(x / ln(2)), r = x - m * ln(2) */ - if (dal_fixed31_32_le( - dal_fixed31_32_ln2_div_2, - dal_fixed31_32_abs(arg))) { - int m = dal_fixed31_32_round( - dal_fixed31_32_div( + if (dc_fixpt_le( + dc_fixpt_ln2_div_2, + dc_fixpt_abs(arg))) { + int m = dc_fixpt_round( + dc_fixpt_div( arg, - dal_fixed31_32_ln2)); + dc_fixpt_ln2)); - struct fixed31_32 r = dal_fixed31_32_sub( + struct fixed31_32 r = dc_fixpt_sub( arg, - dal_fixed31_32_mul_int( - dal_fixed31_32_ln2, + dc_fixpt_mul_int( + dc_fixpt_ln2, m)); ASSERT(m != 0); - ASSERT(dal_fixed31_32_lt( - dal_fixed31_32_abs(r), - dal_fixed31_32_one)); + ASSERT(dc_fixpt_lt( + dc_fixpt_abs(r), + dc_fixpt_one)); if (m > 0) - return dal_fixed31_32_shl( + return dc_fixpt_shl( fixed31_32_exp_from_taylor_series(r), (unsigned char)m); else - return dal_fixed31_32_div_int( + return dc_fixpt_div_int( fixed31_32_exp_from_taylor_series(r), 1LL << -m); } else if (arg.value != 0) return fixed31_32_exp_from_taylor_series(arg); else - return dal_fixed31_32_one; + return dc_fixpt_one; } -struct fixed31_32 dal_fixed31_32_log( - struct fixed31_32 arg) +struct fixed31_32 dc_fixpt_log(struct fixed31_32 arg) { - struct fixed31_32 res = dal_fixed31_32_neg(dal_fixed31_32_one); + struct fixed31_32 res = dc_fixpt_neg(dc_fixpt_one); /* TODO improve 1st estimation */ struct fixed31_32 error; @@ -453,15 +387,15 @@ struct fixed31_32 dal_fixed31_32_log( /* TODO if arg is zero, return -INF */ do { - struct fixed31_32 res1 = dal_fixed31_32_add( - dal_fixed31_32_sub( + struct fixed31_32 res1 = dc_fixpt_add( + dc_fixpt_sub( res, - dal_fixed31_32_one), - dal_fixed31_32_div( + dc_fixpt_one), + dc_fixpt_div( arg, - dal_fixed31_32_exp(res))); + dc_fixpt_exp(res))); - error = dal_fixed31_32_sub( + error = dc_fixpt_sub( res, res1); @@ -472,66 +406,11 @@ struct fixed31_32 dal_fixed31_32_log( return res; } -struct fixed31_32 dal_fixed31_32_pow( - struct fixed31_32 arg1, - struct fixed31_32 arg2) -{ - return dal_fixed31_32_exp( - dal_fixed31_32_mul( - dal_fixed31_32_log(arg1), - arg2)); -} - -int dal_fixed31_32_floor( - struct fixed31_32 arg) -{ - unsigned long long arg_value = abs_i64(arg.value); - - if (arg.value >= 0) - return (int)GET_INTEGER_PART(arg_value); - else - return -(int)GET_INTEGER_PART(arg_value); -} - -int dal_fixed31_32_round( - struct fixed31_32 arg) -{ - unsigned long long arg_value = abs_i64(arg.value); - - const long long summand = dal_fixed31_32_half.value; - - ASSERT(LLONG_MAX - (long long)arg_value >= summand); - - arg_value += summand; - - if (arg.value >= 0) - return (int)GET_INTEGER_PART(arg_value); - else - return -(int)GET_INTEGER_PART(arg_value); -} - -int dal_fixed31_32_ceil( - struct fixed31_32 arg) -{ - unsigned long long arg_value = abs_i64(arg.value); - - const long long summand = dal_fixed31_32_one.value - - dal_fixed31_32_epsilon.value; - - ASSERT(LLONG_MAX - (long long)arg_value >= summand); - - arg_value += summand; - - if (arg.value >= 0) - return (int)GET_INTEGER_PART(arg_value); - else - return -(int)GET_INTEGER_PART(arg_value); -} /* this function is a generic helper to translate fixed point value to * specified integer format that will consist of integer_bits integer part and * fractional_bits fractional part. For example it is used in - * dal_fixed31_32_u2d19 to receive 2 bits integer part and 19 bits fractional + * dc_fixpt_u2d19 to receive 2 bits integer part and 19 bits fractional * part in 32 bits. It is used in hw programming (scaler) */ @@ -570,35 +449,30 @@ static inline unsigned int clamp_ux_dy( return min_clamp; } -unsigned int dal_fixed31_32_u2d19( - struct fixed31_32 arg) +unsigned int dc_fixpt_u2d19(struct fixed31_32 arg) { return ux_dy(arg.value, 2, 19); } -unsigned int dal_fixed31_32_u0d19( - struct fixed31_32 arg) +unsigned int dc_fixpt_u0d19(struct fixed31_32 arg) { return ux_dy(arg.value, 0, 19); } -unsigned int dal_fixed31_32_clamp_u0d14( - struct fixed31_32 arg) +unsigned int dc_fixpt_clamp_u0d14(struct fixed31_32 arg) { return clamp_ux_dy(arg.value, 0, 14, 1); } -unsigned int dal_fixed31_32_clamp_u0d10( - struct fixed31_32 arg) +unsigned int dc_fixpt_clamp_u0d10(struct fixed31_32 arg) { return clamp_ux_dy(arg.value, 0, 10, 1); } -int dal_fixed31_32_s4d19( - struct fixed31_32 arg) +int dc_fixpt_s4d19(struct fixed31_32 arg) { if (arg.value < 0) - return -(int)ux_dy(dal_fixed31_32_abs(arg).value, 4, 19); + return -(int)ux_dy(dc_fixpt_abs(arg).value, 4, 19); else return ux_dy(arg.value, 4, 19); } diff --git a/drivers/gpu/drm/amd/display/dc/basics/fixpt32_32.c b/drivers/gpu/drm/amd/display/dc/basics/fixpt32_32.c deleted file mode 100644 index 4d3aaa82a07b..000000000000 --- a/drivers/gpu/drm/amd/display/dc/basics/fixpt32_32.c +++ /dev/null @@ -1,161 +0,0 @@ -/* - * Copyright 2012-15 Advanced Micro Devices, Inc. - * - * Permission is hereby granted, free of charge, to any person obtaining a - * copy of this software and associated documentation files (the "Software"), - * to deal in the Software without restriction, including without limitation - * the rights to use, copy, modify, merge, publish, distribute, sublicense, - * and/or sell copies of the Software, and to permit persons to whom the - * Software is furnished to do so, subject to the following conditions: - * - * The above copyright notice and this permission notice shall be included in - * all copies or substantial portions of the Software. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL - * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR - * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, - * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR - * OTHER DEALINGS IN THE SOFTWARE. - * - * Authors: AMD - * - */ - -#include "dm_services.h" -#include "include/fixed32_32.h" - -static uint64_t u64_div(uint64_t n, uint64_t d) -{ - uint32_t i = 0; - uint64_t r; - uint64_t q = div64_u64_rem(n, d, &r); - - for (i = 0; i < 32; ++i) { - uint64_t sbit = q & (1ULL<<63); - - r <<= 1; - r |= sbit ? 1 : 0; - q <<= 1; - if (r >= d) { - r -= d; - q |= 1; - } - } - - if (2*r >= d) - q += 1; - return q; -} - -struct fixed32_32 dal_fixed32_32_from_fraction(uint32_t n, uint32_t d) -{ - struct fixed32_32 fx; - - fx.value = u64_div((uint64_t)n << 32, (uint64_t)d << 32); - return fx; -} - -struct fixed32_32 dal_fixed32_32_add( - struct fixed32_32 lhs, - struct fixed32_32 rhs) -{ - struct fixed32_32 fx = {lhs.value + rhs.value}; - - ASSERT(fx.value >= rhs.value); - return fx; -} - -struct fixed32_32 dal_fixed32_32_add_int(struct fixed32_32 lhs, uint32_t rhs) -{ - struct fixed32_32 fx = {lhs.value + ((uint64_t)rhs << 32)}; - - ASSERT(fx.value >= (uint64_t)rhs << 32); - return fx; - -} -struct fixed32_32 dal_fixed32_32_sub( - struct fixed32_32 lhs, - struct fixed32_32 rhs) -{ - struct fixed32_32 fx; - - ASSERT(lhs.value >= rhs.value); - fx.value = lhs.value - rhs.value; - return fx; -} - -struct fixed32_32 dal_fixed32_32_sub_int(struct fixed32_32 lhs, uint32_t rhs) -{ - struct fixed32_32 fx; - - ASSERT(lhs.value >= ((uint64_t)rhs<<32)); - fx.value = lhs.value - ((uint64_t)rhs<<32); - return fx; -} - -struct fixed32_32 dal_fixed32_32_mul( - struct fixed32_32 lhs, - struct fixed32_32 rhs) -{ - struct fixed32_32 fx; - uint64_t lhs_int = lhs.value>>32; - uint64_t lhs_frac = (uint32_t)lhs.value; - uint64_t rhs_int = rhs.value>>32; - uint64_t rhs_frac = (uint32_t)rhs.value; - uint64_t ahbh = lhs_int * rhs_int; - uint64_t ahbl = lhs_int * rhs_frac; - uint64_t albh = lhs_frac * rhs_int; - uint64_t albl = lhs_frac * rhs_frac; - - ASSERT((ahbh>>32) == 0); - - fx.value = (ahbh<<32) + ahbl + albh + (albl>>32); - return fx; - -} - -struct fixed32_32 dal_fixed32_32_mul_int(struct fixed32_32 lhs, uint32_t rhs) -{ - struct fixed32_32 fx; - uint64_t lhsi = (lhs.value>>32) * (uint64_t)rhs; - uint64_t lhsf; - - ASSERT((lhsi>>32) == 0); - lhsf = ((uint32_t)lhs.value) * (uint64_t)rhs; - ASSERT((lhsi<<32) + lhsf >= lhsf); - fx.value = (lhsi<<32) + lhsf; - return fx; -} - -struct fixed32_32 dal_fixed32_32_div( - struct fixed32_32 lhs, - struct fixed32_32 rhs) -{ - struct fixed32_32 fx; - - fx.value = u64_div(lhs.value, rhs.value); - return fx; -} - -struct fixed32_32 dal_fixed32_32_div_int(struct fixed32_32 lhs, uint32_t rhs) -{ - struct fixed32_32 fx; - - fx.value = u64_div(lhs.value, (uint64_t)rhs << 32); - return fx; -} - -uint32_t dal_fixed32_32_ceil(struct fixed32_32 v) -{ - ASSERT((uint32_t)v.value ? (v.value >> 32) + 1 >= 1 : true); - return (v.value>>32) + ((uint32_t)v.value ? 1 : 0); -} - -uint32_t dal_fixed32_32_round(struct fixed32_32 v) -{ - ASSERT(v.value + (1ULL<<31) >= (1ULL<<31)); - return (v.value + (1ULL<<31))>>32; -} - diff --git a/drivers/gpu/drm/amd/display/dc/basics/log_helpers.c b/drivers/gpu/drm/amd/display/dc/basics/log_helpers.c index 854678a0c54b..021451549ff7 100644 --- a/drivers/gpu/drm/amd/display/dc/basics/log_helpers.c +++ b/drivers/gpu/drm/amd/display/dc/basics/log_helpers.c @@ -94,7 +94,6 @@ void dc_conn_log(struct dc_context *ctx, dm_logger_append(&entry, "%2.2X ", hex_data[i]); dm_logger_append(&entry, "^\n"); - dm_helpers_dc_conn_log(ctx, &entry, event); fail: dm_logger_close(&entry); diff --git a/drivers/gpu/drm/amd/display/dc/basics/logger.c b/drivers/gpu/drm/amd/display/dc/basics/logger.c index 31bee054f43a..738a818d58d1 100644 --- a/drivers/gpu/drm/amd/display/dc/basics/logger.c +++ b/drivers/gpu/drm/amd/display/dc/basics/logger.c @@ -61,7 +61,7 @@ static const struct dc_log_type_info log_type_info_tbl[] = { {LOG_EVENT_UNDERFLOW, "Underflow"}, {LOG_IF_TRACE, "InterfaceTrace"}, {LOG_DTN, "DTN"}, - {LOG_PROFILING, "Profiling"} + {LOG_DISPLAYSTATS, "DisplayStats"} }; @@ -402,3 +402,4 @@ cleanup: entry->max_buf_bytes = 0; } } + diff --git a/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c b/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c index 10a5807a7e8b..b8cef7af3c4a 100644 --- a/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c +++ b/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c @@ -1330,6 +1330,9 @@ static enum bp_result bios_parser_get_firmware_info( case 2: result = get_firmware_info_v3_2(bp, info); break; + case 3: + result = get_firmware_info_v3_2(bp, info); + break; default: break; } diff --git a/drivers/gpu/drm/amd/display/dc/bios/command_table_helper.c b/drivers/gpu/drm/amd/display/dc/bios/command_table_helper.c index be066c49b984..253bbb1eea60 100644 --- a/drivers/gpu/drm/amd/display/dc/bios/command_table_helper.c +++ b/drivers/gpu/drm/amd/display/dc/bios/command_table_helper.c @@ -51,9 +51,7 @@ bool dal_bios_parser_init_cmd_tbl_helper( return true; case DCE_VERSION_11_2: -#if defined(CONFIG_DRM_AMD_DC_VEGAM) case DCE_VERSION_11_22: -#endif *h = dal_cmd_tbl_helper_dce112_get_table(); return true; diff --git a/drivers/gpu/drm/amd/display/dc/bios/command_table_helper2.c b/drivers/gpu/drm/amd/display/dc/bios/command_table_helper2.c index 9b9e06995805..bbbcef566c55 100644 --- a/drivers/gpu/drm/amd/display/dc/bios/command_table_helper2.c +++ b/drivers/gpu/drm/amd/display/dc/bios/command_table_helper2.c @@ -52,9 +52,7 @@ bool dal_bios_parser_init_cmd_tbl_helper2( return true; case DCE_VERSION_11_2: -#if defined(CONFIG_DRM_AMD_DC_VEGAM) case DCE_VERSION_11_22: -#endif *h = dal_cmd_tbl_helper_dce112_get_table2(); return true; #if defined(CONFIG_DRM_AMD_DC_DCN1_0) diff --git a/drivers/gpu/drm/amd/display/dc/calcs/custom_float.c b/drivers/gpu/drm/amd/display/dc/calcs/custom_float.c index 7243c37f569e..31d167bc548f 100644 --- a/drivers/gpu/drm/amd/display/dc/calcs/custom_float.c +++ b/drivers/gpu/drm/amd/display/dc/calcs/custom_float.c @@ -36,41 +36,41 @@ static bool build_custom_float( uint32_t exp_offset = (1 << (format->exponenta_bits - 1)) - 1; const struct fixed31_32 mantissa_constant_plus_max_fraction = - dal_fixed31_32_from_fraction( + dc_fixpt_from_fraction( (1LL << (format->mantissa_bits + 1)) - 1, 1LL << format->mantissa_bits); struct fixed31_32 mantiss; - if (dal_fixed31_32_eq( + if (dc_fixpt_eq( value, - dal_fixed31_32_zero)) { + dc_fixpt_zero)) { *negative = false; *mantissa = 0; *exponenta = 0; return true; } - if (dal_fixed31_32_lt( + if (dc_fixpt_lt( value, - dal_fixed31_32_zero)) { + dc_fixpt_zero)) { *negative = format->sign; - value = dal_fixed31_32_neg(value); + value = dc_fixpt_neg(value); } else { *negative = false; } - if (dal_fixed31_32_lt( + if (dc_fixpt_lt( value, - dal_fixed31_32_one)) { + dc_fixpt_one)) { uint32_t i = 1; do { - value = dal_fixed31_32_shl(value, 1); + value = dc_fixpt_shl(value, 1); ++i; - } while (dal_fixed31_32_lt( + } while (dc_fixpt_lt( value, - dal_fixed31_32_one)); + dc_fixpt_one)); --i; @@ -81,15 +81,15 @@ static bool build_custom_float( } *exponenta = exp_offset - i; - } else if (dal_fixed31_32_le( + } else if (dc_fixpt_le( mantissa_constant_plus_max_fraction, value)) { uint32_t i = 1; do { - value = dal_fixed31_32_shr(value, 1); + value = dc_fixpt_shr(value, 1); ++i; - } while (dal_fixed31_32_lt( + } while (dc_fixpt_lt( mantissa_constant_plus_max_fraction, value)); @@ -98,23 +98,23 @@ static bool build_custom_float( *exponenta = exp_offset; } - mantiss = dal_fixed31_32_sub( + mantiss = dc_fixpt_sub( value, - dal_fixed31_32_one); + dc_fixpt_one); - if (dal_fixed31_32_lt( + if (dc_fixpt_lt( mantiss, - dal_fixed31_32_zero) || - dal_fixed31_32_lt( - dal_fixed31_32_one, + dc_fixpt_zero) || + dc_fixpt_lt( + dc_fixpt_one, mantiss)) - mantiss = dal_fixed31_32_zero; + mantiss = dc_fixpt_zero; else - mantiss = dal_fixed31_32_shl( + mantiss = dc_fixpt_shl( mantiss, format->mantissa_bits); - *mantissa = dal_fixed31_32_floor(mantiss); + *mantissa = dc_fixpt_floor(mantiss); return true; } diff --git a/drivers/gpu/drm/amd/display/dc/calcs/dce_calcs.c b/drivers/gpu/drm/amd/display/dc/calcs/dce_calcs.c index 4ee3c26f7c13..2c4e8f0cb2dc 100644 --- a/drivers/gpu/drm/amd/display/dc/calcs/dce_calcs.c +++ b/drivers/gpu/drm/amd/display/dc/calcs/dce_calcs.c @@ -59,10 +59,8 @@ static enum bw_calcs_version bw_calcs_version_from_asic_id(struct hw_asic_id asi return BW_CALCS_VERSION_POLARIS10; if (ASIC_REV_IS_POLARIS11_M(asic_id.hw_internal_rev)) return BW_CALCS_VERSION_POLARIS11; -#if defined(CONFIG_DRM_AMD_DC_VEGAM) if (ASIC_REV_IS_VEGAM(asic_id.hw_internal_rev)) return BW_CALCS_VERSION_VEGAM; -#endif return BW_CALCS_VERSION_INVALID; case FAMILY_AI: @@ -2151,11 +2149,9 @@ void bw_calcs_init(struct bw_calcs_dceip *bw_dceip, dceip.mcifwr_all_surfaces_burst_time = bw_int_to_fixed(0); /* todo: this is a bug*/ break; case BW_CALCS_VERSION_POLARIS10: -#if defined(CONFIG_DRM_AMD_DC_VEGAM) /* TODO: Treat VEGAM the same as P10 for now * Need to tune the para for VEGAM if needed */ case BW_CALCS_VERSION_VEGAM: -#endif vbios.memory_type = bw_def_gddr5; vbios.dram_channel_width_in_bits = 32; vbios.number_of_dram_channels = asic_id.vram_width / vbios.dram_channel_width_in_bits; diff --git a/drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c b/drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c index a102c192328d..49a4ea45466d 100644 --- a/drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c +++ b/drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c @@ -873,14 +873,14 @@ bool dcn_validate_bandwidth( } if (pipe->plane_state->rotation % 2 == 0) { - ASSERT(pipe->plane_res.scl_data.ratios.horz.value != dal_fixed31_32_one.value + ASSERT(pipe->plane_res.scl_data.ratios.horz.value != dc_fixpt_one.value || v->scaler_rec_out_width[input_idx] == v->viewport_width[input_idx]); - ASSERT(pipe->plane_res.scl_data.ratios.vert.value != dal_fixed31_32_one.value + ASSERT(pipe->plane_res.scl_data.ratios.vert.value != dc_fixpt_one.value || v->scaler_recout_height[input_idx] == v->viewport_height[input_idx]); } else { - ASSERT(pipe->plane_res.scl_data.ratios.horz.value != dal_fixed31_32_one.value + ASSERT(pipe->plane_res.scl_data.ratios.horz.value != dc_fixpt_one.value || v->scaler_recout_height[input_idx] == v->viewport_width[input_idx]); - ASSERT(pipe->plane_res.scl_data.ratios.vert.value != dal_fixed31_32_one.value + ASSERT(pipe->plane_res.scl_data.ratios.vert.value != dc_fixpt_one.value || v->scaler_rec_out_width[input_idx] == v->viewport_height[input_idx]); } v->dcc_enable[input_idx] = pipe->plane_state->dcc.enable ? dcn_bw_yes : dcn_bw_no; diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link.c b/drivers/gpu/drm/amd/display/dc/core/dc_link.c index b44cf52090a5..2fa521812d23 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_link.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_link.c @@ -469,6 +469,13 @@ static void link_disconnect_sink(struct dc_link *link) link->dpcd_sink_count = 0; } +static void link_disconnect_remap(struct dc_sink *prev_sink, struct dc_link *link) +{ + dc_sink_release(link->local_sink); + link->local_sink = prev_sink; +} + + static bool detect_dp( struct dc_link *link, struct display_sink_capability *sink_caps, @@ -551,6 +558,17 @@ static bool detect_dp( return true; } +static bool is_same_edid(struct dc_edid *old_edid, struct dc_edid *new_edid) +{ + if (old_edid->length != new_edid->length) + return false; + + if (new_edid->length == 0) + return false; + + return (memcmp(old_edid->raw_edid, new_edid->raw_edid, new_edid->length) == 0); +} + bool dc_link_detect(struct dc_link *link, enum dc_detect_reason reason) { struct dc_sink_init_data sink_init_data = { 0 }; @@ -558,9 +576,13 @@ bool dc_link_detect(struct dc_link *link, enum dc_detect_reason reason) uint8_t i; bool converter_disable_audio = false; struct audio_support *aud_support = &link->dc->res_pool->audio_support; + bool same_edid = false; enum dc_edid_status edid_status; struct dc_context *dc_ctx = link->ctx; struct dc_sink *sink = NULL; + struct dc_sink *prev_sink = NULL; + struct dpcd_caps prev_dpcd_caps; + bool same_dpcd = true; enum dc_connection_type new_connection_type = dc_connection_none; DC_LOGGER_INIT(link->ctx->logger); if (link->connector_signal == SIGNAL_TYPE_VIRTUAL) @@ -575,6 +597,11 @@ bool dc_link_detect(struct dc_link *link, enum dc_detect_reason reason) link->local_sink) return true; + prev_sink = link->local_sink; + if (prev_sink != NULL) { + dc_sink_retain(prev_sink); + memcpy(&prev_dpcd_caps, &link->dpcd_caps, sizeof(struct dpcd_caps)); + } link_disconnect_sink(link); if (new_connection_type != dc_connection_none) { @@ -616,14 +643,25 @@ bool dc_link_detect(struct dc_link *link, enum dc_detect_reason reason) link, &sink_caps, &converter_disable_audio, - aud_support, reason)) + aud_support, reason)) { + if (prev_sink != NULL) + dc_sink_release(prev_sink); return false; + } + // Check if dpcp block is the same + if (prev_sink != NULL) { + if (memcmp(&link->dpcd_caps, &prev_dpcd_caps, sizeof(struct dpcd_caps))) + same_dpcd = false; + } /* Active dongle downstream unplug */ if (link->type == dc_connection_active_dongle && link->dpcd_caps.sink_count. - bits.SINK_COUNT == 0) + bits.SINK_COUNT == 0) { + if (prev_sink != NULL) + dc_sink_release(prev_sink); return true; + } if (link->type == dc_connection_mst_branch) { LINK_INFO("link=%d, mst branch is now Connected\n", @@ -631,9 +669,11 @@ bool dc_link_detect(struct dc_link *link, enum dc_detect_reason reason) /* Need to setup mst link_cap struct here * otherwise dc_link_detect() will leave mst link_cap * empty which leads to allocate_mst_payload() has "0" - * pbn_per_slot value leading to exception on dal_fixed31_32_div() + * pbn_per_slot value leading to exception on dc_fixpt_div() */ link->verified_link_cap = link->reported_link_cap; + if (prev_sink != NULL) + dc_sink_release(prev_sink); return false; } @@ -643,6 +683,8 @@ bool dc_link_detect(struct dc_link *link, enum dc_detect_reason reason) default: DC_ERROR("Invalid connector type! signal:%d\n", link->connector_signal); + if (prev_sink != NULL) + dc_sink_release(prev_sink); return false; } /* switch() */ @@ -665,6 +707,8 @@ bool dc_link_detect(struct dc_link *link, enum dc_detect_reason reason) sink = dc_sink_create(&sink_init_data); if (!sink) { DC_ERROR("Failed to create sink!\n"); + if (prev_sink != NULL) + dc_sink_release(prev_sink); return false; } @@ -688,22 +732,33 @@ bool dc_link_detect(struct dc_link *link, enum dc_detect_reason reason) break; } - if (link->connector_signal == SIGNAL_TYPE_DISPLAY_PORT && - sink_caps.transaction_type == - DDC_TRANSACTION_TYPE_I2C_OVER_AUX) { - /* - * TODO debug why Dell 2413 doesn't like - * two link trainings - */ + // Check if edid is the same + if ((prev_sink != NULL) && ((edid_status == EDID_THE_SAME) || (edid_status == EDID_OK))) + same_edid = is_same_edid(&prev_sink->dc_edid, &sink->dc_edid); - /* deal with non-mst cases */ - dp_hbr_verify_link_cap(link, &link->reported_link_cap); - } + // If both edid and dpcd are the same, then discard new sink and revert back to original sink + if ((same_edid) && (same_dpcd)) { + link_disconnect_remap(prev_sink, link); + sink = prev_sink; + prev_sink = NULL; + } else { + if (link->connector_signal == SIGNAL_TYPE_DISPLAY_PORT && + sink_caps.transaction_type == + DDC_TRANSACTION_TYPE_I2C_OVER_AUX) { + /* + * TODO debug why Dell 2413 doesn't like + * two link trainings + */ + + /* deal with non-mst cases */ + dp_hbr_verify_link_cap(link, &link->reported_link_cap); + } - /* HDMI-DVI Dongle */ - if (sink->sink_signal == SIGNAL_TYPE_HDMI_TYPE_A && - !sink->edid_caps.edid_hdmi) - sink->sink_signal = SIGNAL_TYPE_DVI_SINGLE_LINK; + /* HDMI-DVI Dongle */ + if (sink->sink_signal == SIGNAL_TYPE_HDMI_TYPE_A && + !sink->edid_caps.edid_hdmi) + sink->sink_signal = SIGNAL_TYPE_DVI_SINGLE_LINK; + } /* Connectivity log: detection */ for (i = 0; i < sink->dc_edid.length / EDID_BLOCK_SIZE; i++) { @@ -762,10 +817,14 @@ bool dc_link_detect(struct dc_link *link, enum dc_detect_reason reason) sink_caps.signal = SIGNAL_TYPE_NONE; } - LINK_INFO("link=%d, dc_sink_in=%p is now %s\n", + LINK_INFO("link=%d, dc_sink_in=%p is now %s prev_sink=%p dpcd same=%d edid same=%d\n", link->link_index, sink, (sink_caps.signal == SIGNAL_TYPE_NONE ? - "Disconnected":"Connected")); + "Disconnected":"Connected"), prev_sink, + same_dpcd, same_edid); + + if (prev_sink != NULL) + dc_sink_release(prev_sink); return true; } @@ -2059,10 +2118,10 @@ static struct fixed31_32 get_pbn_per_slot(struct dc_stream_state *stream) &stream->sink->link->cur_link_settings; uint32_t link_rate_in_mbps = link_settings->link_rate * LINK_RATE_REF_FREQ_IN_MHZ; - struct fixed31_32 mbps = dal_fixed31_32_from_int( + struct fixed31_32 mbps = dc_fixpt_from_int( link_rate_in_mbps * link_settings->lane_count); - return dal_fixed31_32_div_int(mbps, 54); + return dc_fixpt_div_int(mbps, 54); } static int get_color_depth(enum dc_color_depth color_depth) @@ -2103,7 +2162,7 @@ static struct fixed31_32 get_pbn_from_timing(struct pipe_ctx *pipe_ctx) numerator = 64 * PEAK_FACTOR_X1000; denominator = 54 * 8 * 1000 * 1000; kbps *= numerator; - peak_kbps = dal_fixed31_32_from_fraction(kbps, denominator); + peak_kbps = dc_fixpt_from_fraction(kbps, denominator); return peak_kbps; } @@ -2230,7 +2289,7 @@ static enum dc_status allocate_mst_payload(struct pipe_ctx *pipe_ctx) /* slot X.Y for only current stream */ pbn_per_slot = get_pbn_per_slot(stream); pbn = get_pbn_from_timing(pipe_ctx); - avg_time_slots_per_mtp = dal_fixed31_32_div(pbn, pbn_per_slot); + avg_time_slots_per_mtp = dc_fixpt_div(pbn, pbn_per_slot); stream_encoder->funcs->set_mst_bandwidth( stream_encoder, @@ -2247,7 +2306,7 @@ static enum dc_status deallocate_mst_payload(struct pipe_ctx *pipe_ctx) struct link_encoder *link_encoder = link->link_enc; struct stream_encoder *stream_encoder = pipe_ctx->stream_res.stream_enc; struct dp_mst_stream_allocation_table proposed_table = {0}; - struct fixed31_32 avg_time_slots_per_mtp = dal_fixed31_32_from_int(0); + struct fixed31_32 avg_time_slots_per_mtp = dc_fixpt_from_int(0); uint8_t i; bool mst_mode = (link->type == dc_connection_mst_branch); DC_LOGGER_INIT(link->ctx->logger); diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link_hwss.c b/drivers/gpu/drm/amd/display/dc/core/dc_link_hwss.c index 7c866a7d5e77..82cd1d6e6e59 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_link_hwss.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_link_hwss.c @@ -11,8 +11,6 @@ #include "dc_link_dp.h" #include "dc_link_ddc.h" #include "dm_helpers.h" -#include "dce/dce_link_encoder.h" -#include "dce/dce_stream_encoder.h" #include "dpcd_defs.h" enum dc_status core_link_read_dpcd( diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c index 9eb731fb5251..751f3ac9d921 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c @@ -79,10 +79,8 @@ enum dce_version resource_parse_asic_id(struct hw_asic_id asic_id) ASIC_REV_IS_POLARIS12_V(asic_id.hw_internal_rev)) { dc_version = DCE_VERSION_11_2; } -#if defined(CONFIG_DRM_AMD_DC_VEGAM) if (ASIC_REV_IS_VEGAM(asic_id.hw_internal_rev)) dc_version = DCE_VERSION_11_22; -#endif break; case FAMILY_AI: dc_version = DCE_VERSION_12_0; @@ -129,9 +127,7 @@ struct resource_pool *dc_create_resource_pool( num_virtual_links, dc, asic_id); break; case DCE_VERSION_11_2: -#if defined(CONFIG_DRM_AMD_DC_VEGAM) case DCE_VERSION_11_22: -#endif res_pool = dce112_create_resource_pool( num_virtual_links, dc); break; @@ -500,9 +496,9 @@ static void calculate_viewport(struct pipe_ctx *pipe_ctx) data->viewport_c.x = data->viewport.x / vpc_div; data->viewport_c.y = data->viewport.y / vpc_div; data->inits.h_c = (data->viewport.x % vpc_div) != 0 ? - dal_fixed31_32_half : dal_fixed31_32_zero; + dc_fixpt_half : dc_fixpt_zero; data->inits.v_c = (data->viewport.y % vpc_div) != 0 ? - dal_fixed31_32_half : dal_fixed31_32_zero; + dc_fixpt_half : dc_fixpt_zero; /* Round up, assume original video size always even dimensions */ data->viewport_c.width = (data->viewport.width + vpc_div - 1) / vpc_div; data->viewport_c.height = (data->viewport.height + vpc_div - 1) / vpc_div; @@ -631,10 +627,10 @@ static void calculate_scaling_ratios(struct pipe_ctx *pipe_ctx) pipe_ctx->plane_state->rotation == ROTATION_ANGLE_270) rect_swap_helper(&surf_src); - pipe_ctx->plane_res.scl_data.ratios.horz = dal_fixed31_32_from_fraction( + pipe_ctx->plane_res.scl_data.ratios.horz = dc_fixpt_from_fraction( surf_src.width, plane_state->dst_rect.width); - pipe_ctx->plane_res.scl_data.ratios.vert = dal_fixed31_32_from_fraction( + pipe_ctx->plane_res.scl_data.ratios.vert = dc_fixpt_from_fraction( surf_src.height, plane_state->dst_rect.height); @@ -656,6 +652,14 @@ static void calculate_scaling_ratios(struct pipe_ctx *pipe_ctx) pipe_ctx->plane_res.scl_data.ratios.horz_c.value /= 2; pipe_ctx->plane_res.scl_data.ratios.vert_c.value /= 2; } + pipe_ctx->plane_res.scl_data.ratios.horz = dc_fixpt_truncate( + pipe_ctx->plane_res.scl_data.ratios.horz, 19); + pipe_ctx->plane_res.scl_data.ratios.vert = dc_fixpt_truncate( + pipe_ctx->plane_res.scl_data.ratios.vert, 19); + pipe_ctx->plane_res.scl_data.ratios.horz_c = dc_fixpt_truncate( + pipe_ctx->plane_res.scl_data.ratios.horz_c, 19); + pipe_ctx->plane_res.scl_data.ratios.vert_c = dc_fixpt_truncate( + pipe_ctx->plane_res.scl_data.ratios.vert_c, 19); } static void calculate_inits_and_adj_vp(struct pipe_ctx *pipe_ctx, struct view *recout_skip) @@ -692,32 +696,33 @@ static void calculate_inits_and_adj_vp(struct pipe_ctx *pipe_ctx, struct view *r * init_bot = init + scaling_ratio * init_c = init + truncated_vp_c_offset(from calculate viewport) */ - data->inits.h = dal_fixed31_32_div_int( - dal_fixed31_32_add_int(data->ratios.horz, data->taps.h_taps + 1), 2); + data->inits.h = dc_fixpt_truncate(dc_fixpt_div_int( + dc_fixpt_add_int(data->ratios.horz, data->taps.h_taps + 1), 2), 19); + + data->inits.h_c = dc_fixpt_truncate(dc_fixpt_add(data->inits.h_c, dc_fixpt_div_int( + dc_fixpt_add_int(data->ratios.horz_c, data->taps.h_taps_c + 1), 2)), 19); - data->inits.h_c = dal_fixed31_32_add(data->inits.h_c, dal_fixed31_32_div_int( - dal_fixed31_32_add_int(data->ratios.horz_c, data->taps.h_taps_c + 1), 2)); + data->inits.v = dc_fixpt_truncate(dc_fixpt_div_int( + dc_fixpt_add_int(data->ratios.vert, data->taps.v_taps + 1), 2), 19); - data->inits.v = dal_fixed31_32_div_int( - dal_fixed31_32_add_int(data->ratios.vert, data->taps.v_taps + 1), 2); + data->inits.v_c = dc_fixpt_truncate(dc_fixpt_add(data->inits.v_c, dc_fixpt_div_int( + dc_fixpt_add_int(data->ratios.vert_c, data->taps.v_taps_c + 1), 2)), 19); - data->inits.v_c = dal_fixed31_32_add(data->inits.v_c, dal_fixed31_32_div_int( - dal_fixed31_32_add_int(data->ratios.vert_c, data->taps.v_taps_c + 1), 2)); /* Adjust for viewport end clip-off */ if ((data->viewport.x + data->viewport.width) < (src.x + src.width) && !flip_horz_scan_dir) { int vp_clip = src.x + src.width - data->viewport.width - data->viewport.x; - int int_part = dal_fixed31_32_floor( - dal_fixed31_32_sub(data->inits.h, data->ratios.horz)); + int int_part = dc_fixpt_floor( + dc_fixpt_sub(data->inits.h, data->ratios.horz)); int_part = int_part > 0 ? int_part : 0; data->viewport.width += int_part < vp_clip ? int_part : vp_clip; } if ((data->viewport.y + data->viewport.height) < (src.y + src.height) && !flip_vert_scan_dir) { int vp_clip = src.y + src.height - data->viewport.height - data->viewport.y; - int int_part = dal_fixed31_32_floor( - dal_fixed31_32_sub(data->inits.v, data->ratios.vert)); + int int_part = dc_fixpt_floor( + dc_fixpt_sub(data->inits.v, data->ratios.vert)); int_part = int_part > 0 ? int_part : 0; data->viewport.height += int_part < vp_clip ? int_part : vp_clip; @@ -725,8 +730,8 @@ static void calculate_inits_and_adj_vp(struct pipe_ctx *pipe_ctx, struct view *r if ((data->viewport_c.x + data->viewport_c.width) < (src.x + src.width) / vpc_div && !flip_horz_scan_dir) { int vp_clip = (src.x + src.width) / vpc_div - data->viewport_c.width - data->viewport_c.x; - int int_part = dal_fixed31_32_floor( - dal_fixed31_32_sub(data->inits.h_c, data->ratios.horz_c)); + int int_part = dc_fixpt_floor( + dc_fixpt_sub(data->inits.h_c, data->ratios.horz_c)); int_part = int_part > 0 ? int_part : 0; data->viewport_c.width += int_part < vp_clip ? int_part : vp_clip; @@ -734,8 +739,8 @@ static void calculate_inits_and_adj_vp(struct pipe_ctx *pipe_ctx, struct view *r if ((data->viewport_c.y + data->viewport_c.height) < (src.y + src.height) / vpc_div && !flip_vert_scan_dir) { int vp_clip = (src.y + src.height) / vpc_div - data->viewport_c.height - data->viewport_c.y; - int int_part = dal_fixed31_32_floor( - dal_fixed31_32_sub(data->inits.v_c, data->ratios.vert_c)); + int int_part = dc_fixpt_floor( + dc_fixpt_sub(data->inits.v_c, data->ratios.vert_c)); int_part = int_part > 0 ? int_part : 0; data->viewport_c.height += int_part < vp_clip ? int_part : vp_clip; @@ -745,9 +750,9 @@ static void calculate_inits_and_adj_vp(struct pipe_ctx *pipe_ctx, struct view *r if (data->viewport.x && !flip_horz_scan_dir) { int int_part; - data->inits.h = dal_fixed31_32_add(data->inits.h, dal_fixed31_32_mul_int( + data->inits.h = dc_fixpt_add(data->inits.h, dc_fixpt_mul_int( data->ratios.horz, recout_skip->width)); - int_part = dal_fixed31_32_floor(data->inits.h) - data->viewport.x; + int_part = dc_fixpt_floor(data->inits.h) - data->viewport.x; if (int_part < data->taps.h_taps) { int int_adj = data->viewport.x >= (data->taps.h_taps - int_part) ? (data->taps.h_taps - int_part) : data->viewport.x; @@ -760,15 +765,15 @@ static void calculate_inits_and_adj_vp(struct pipe_ctx *pipe_ctx, struct view *r int_part = data->taps.h_taps; } data->inits.h.value &= 0xffffffff; - data->inits.h = dal_fixed31_32_add_int(data->inits.h, int_part); + data->inits.h = dc_fixpt_add_int(data->inits.h, int_part); } if (data->viewport_c.x && !flip_horz_scan_dir) { int int_part; - data->inits.h_c = dal_fixed31_32_add(data->inits.h_c, dal_fixed31_32_mul_int( + data->inits.h_c = dc_fixpt_add(data->inits.h_c, dc_fixpt_mul_int( data->ratios.horz_c, recout_skip->width)); - int_part = dal_fixed31_32_floor(data->inits.h_c) - data->viewport_c.x; + int_part = dc_fixpt_floor(data->inits.h_c) - data->viewport_c.x; if (int_part < data->taps.h_taps_c) { int int_adj = data->viewport_c.x >= (data->taps.h_taps_c - int_part) ? (data->taps.h_taps_c - int_part) : data->viewport_c.x; @@ -781,15 +786,15 @@ static void calculate_inits_and_adj_vp(struct pipe_ctx *pipe_ctx, struct view *r int_part = data->taps.h_taps_c; } data->inits.h_c.value &= 0xffffffff; - data->inits.h_c = dal_fixed31_32_add_int(data->inits.h_c, int_part); + data->inits.h_c = dc_fixpt_add_int(data->inits.h_c, int_part); } if (data->viewport.y && !flip_vert_scan_dir) { int int_part; - data->inits.v = dal_fixed31_32_add(data->inits.v, dal_fixed31_32_mul_int( + data->inits.v = dc_fixpt_add(data->inits.v, dc_fixpt_mul_int( data->ratios.vert, recout_skip->height)); - int_part = dal_fixed31_32_floor(data->inits.v) - data->viewport.y; + int_part = dc_fixpt_floor(data->inits.v) - data->viewport.y; if (int_part < data->taps.v_taps) { int int_adj = data->viewport.y >= (data->taps.v_taps - int_part) ? (data->taps.v_taps - int_part) : data->viewport.y; @@ -802,15 +807,15 @@ static void calculate_inits_and_adj_vp(struct pipe_ctx *pipe_ctx, struct view *r int_part = data->taps.v_taps; } data->inits.v.value &= 0xffffffff; - data->inits.v = dal_fixed31_32_add_int(data->inits.v, int_part); + data->inits.v = dc_fixpt_add_int(data->inits.v, int_part); } if (data->viewport_c.y && !flip_vert_scan_dir) { int int_part; - data->inits.v_c = dal_fixed31_32_add(data->inits.v_c, dal_fixed31_32_mul_int( + data->inits.v_c = dc_fixpt_add(data->inits.v_c, dc_fixpt_mul_int( data->ratios.vert_c, recout_skip->height)); - int_part = dal_fixed31_32_floor(data->inits.v_c) - data->viewport_c.y; + int_part = dc_fixpt_floor(data->inits.v_c) - data->viewport_c.y; if (int_part < data->taps.v_taps_c) { int int_adj = data->viewport_c.y >= (data->taps.v_taps_c - int_part) ? (data->taps.v_taps_c - int_part) : data->viewport_c.y; @@ -823,12 +828,12 @@ static void calculate_inits_and_adj_vp(struct pipe_ctx *pipe_ctx, struct view *r int_part = data->taps.v_taps_c; } data->inits.v_c.value &= 0xffffffff; - data->inits.v_c = dal_fixed31_32_add_int(data->inits.v_c, int_part); + data->inits.v_c = dc_fixpt_add_int(data->inits.v_c, int_part); } /* Interlaced inits based on final vert inits */ - data->inits.v_bot = dal_fixed31_32_add(data->inits.v, data->ratios.vert); - data->inits.v_c_bot = dal_fixed31_32_add(data->inits.v_c, data->ratios.vert_c); + data->inits.v_bot = dc_fixpt_add(data->inits.v, data->ratios.vert); + data->inits.v_c_bot = dc_fixpt_add(data->inits.v_c, data->ratios.vert_c); if (pipe_ctx->plane_state->rotation == ROTATION_ANGLE_90 || pipe_ctx->plane_state->rotation == ROTATION_ANGLE_270) { diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h index cd4f4341cb53..9cfde0ccf4e9 100644 --- a/drivers/gpu/drm/amd/display/dc/dc.h +++ b/drivers/gpu/drm/amd/display/dc/dc.h @@ -75,6 +75,7 @@ struct dc_caps { bool dynamic_audio; bool is_apu; bool dual_link_dvi; + bool post_blend_color_processing; }; struct dc_dcc_surface_param { @@ -203,6 +204,7 @@ struct dc_debug { bool clock_trace; bool validation_trace; bool bandwidth_calcs_trace; + int max_downscale_src_width; /* stutter efficiency related */ bool disable_stutter; @@ -239,6 +241,8 @@ struct dc_debug { bool az_endpoint_mute_only; bool always_use_regamma; bool p010_mpo_support; + bool recovery_enabled; + }; struct dc_state; struct resource_pool; @@ -499,18 +503,18 @@ struct dc_surface_update { struct dc_plane_state *surface; /* isr safe update parameters. null means no updates */ - struct dc_flip_addrs *flip_addr; - struct dc_plane_info *plane_info; - struct dc_scaling_info *scaling_info; + const struct dc_flip_addrs *flip_addr; + const struct dc_plane_info *plane_info; + const struct dc_scaling_info *scaling_info; /* following updates require alloc/sleep/spin that is not isr safe, * null means no updates */ - struct dc_gamma *gamma; - struct dc_transfer_func *in_transfer_func; + const struct dc_gamma *gamma; + const struct dc_transfer_func *in_transfer_func; - struct dc_csc_transform *input_csc_color_matrix; - struct fixed31_32 *coeff_reduction_factor; + const struct dc_csc_transform *input_csc_color_matrix; + const struct fixed31_32 *coeff_reduction_factor; }; /* diff --git a/drivers/gpu/drm/amd/display/dc/dc_dp_types.h b/drivers/gpu/drm/amd/display/dc/dc_dp_types.h index 2726b02e006b..90bccd5ccaa2 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_dp_types.h +++ b/drivers/gpu/drm/amd/display/dc/dc_dp_types.h @@ -26,6 +26,8 @@ #ifndef DC_DP_TYPES_H #define DC_DP_TYPES_H +#include "os_types.h" + enum dc_lane_count { LANE_COUNT_UNKNOWN = 0, LANE_COUNT_ONE = 1, diff --git a/drivers/gpu/drm/amd/display/dc/dc_types.h b/drivers/gpu/drm/amd/display/dc/dc_types.h index 9defe3b17617..76df2534c4a4 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_types.h +++ b/drivers/gpu/drm/amd/display/dc/dc_types.h @@ -25,7 +25,7 @@ #ifndef DC_TYPES_H_ #define DC_TYPES_H_ -#include "fixed32_32.h" +#include "os_types.h" #include "fixed31_32.h" #include "irq_types.h" #include "dc_dp_types.h" diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_abm.c b/drivers/gpu/drm/amd/display/dc/dce/dce_abm.c index fe92a1222803..29294db1a96b 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_abm.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_abm.c @@ -26,7 +26,7 @@ #include "dce_abm.h" #include "dm_services.h" #include "reg_helper.h" -#include "fixed32_32.h" +#include "fixed31_32.h" #include "dc.h" #include "atom.h" diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c b/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c index 223db98a568a..599c7ab6befe 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c @@ -590,9 +590,7 @@ static uint32_t dce110_get_pix_clk_dividers( pll_settings, pix_clk_params); break; case DCE_VERSION_11_2: -#if defined(CONFIG_DRM_AMD_DC_VEGAM) case DCE_VERSION_11_22: -#endif case DCE_VERSION_12_0: #if defined(CONFIG_DRM_AMD_DC_DCN1_0) case DCN_VERSION_1_0: @@ -659,12 +657,12 @@ static uint32_t dce110_get_d_to_pixel_rate_in_hz( return 0; } - pix_rate = dal_fixed31_32_from_int(clk_src->ref_freq_khz); - pix_rate = dal_fixed31_32_mul_int(pix_rate, 1000); - pix_rate = dal_fixed31_32_mul_int(pix_rate, phase); - pix_rate = dal_fixed31_32_div_int(pix_rate, modulo); + pix_rate = dc_fixpt_from_int(clk_src->ref_freq_khz); + pix_rate = dc_fixpt_mul_int(pix_rate, 1000); + pix_rate = dc_fixpt_mul_int(pix_rate, phase); + pix_rate = dc_fixpt_div_int(pix_rate, modulo); - return dal_fixed31_32_round(pix_rate); + return dc_fixpt_round(pix_rate); } else { return dce110_get_dp_pixel_rate_from_combo_phy_pll(cs, pix_clk_params, pll_settings); } @@ -713,12 +711,12 @@ static bool calculate_ss( const struct spread_spectrum_data *ss_data, struct delta_sigma_data *ds_data) { - struct fixed32_32 fb_div; - struct fixed32_32 ss_amount; - struct fixed32_32 ss_nslip_amount; - struct fixed32_32 ss_ds_frac_amount; - struct fixed32_32 ss_step_size; - struct fixed32_32 modulation_time; + struct fixed31_32 fb_div; + struct fixed31_32 ss_amount; + struct fixed31_32 ss_nslip_amount; + struct fixed31_32 ss_ds_frac_amount; + struct fixed31_32 ss_step_size; + struct fixed31_32 modulation_time; if (ds_data == NULL) return false; @@ -733,42 +731,42 @@ static bool calculate_ss( /* compute SS_AMOUNT_FBDIV & SS_AMOUNT_NFRAC_SLIP & SS_AMOUNT_DSFRAC*/ /* 6 decimal point support in fractional feedback divider */ - fb_div = dal_fixed32_32_from_fraction( + fb_div = dc_fixpt_from_fraction( pll_settings->fract_feedback_divider, 1000000); - fb_div = dal_fixed32_32_add_int(fb_div, pll_settings->feedback_divider); + fb_div = dc_fixpt_add_int(fb_div, pll_settings->feedback_divider); ds_data->ds_frac_amount = 0; /*spreadSpectrumPercentage is in the unit of .01%, * so have to divided by 100 * 100*/ - ss_amount = dal_fixed32_32_mul( - fb_div, dal_fixed32_32_from_fraction(ss_data->percentage, + ss_amount = dc_fixpt_mul( + fb_div, dc_fixpt_from_fraction(ss_data->percentage, 100 * ss_data->percentage_divider)); - ds_data->feedback_amount = dal_fixed32_32_floor(ss_amount); + ds_data->feedback_amount = dc_fixpt_floor(ss_amount); - ss_nslip_amount = dal_fixed32_32_sub(ss_amount, - dal_fixed32_32_from_int(ds_data->feedback_amount)); - ss_nslip_amount = dal_fixed32_32_mul_int(ss_nslip_amount, 10); - ds_data->nfrac_amount = dal_fixed32_32_floor(ss_nslip_amount); + ss_nslip_amount = dc_fixpt_sub(ss_amount, + dc_fixpt_from_int(ds_data->feedback_amount)); + ss_nslip_amount = dc_fixpt_mul_int(ss_nslip_amount, 10); + ds_data->nfrac_amount = dc_fixpt_floor(ss_nslip_amount); - ss_ds_frac_amount = dal_fixed32_32_sub(ss_nslip_amount, - dal_fixed32_32_from_int(ds_data->nfrac_amount)); - ss_ds_frac_amount = dal_fixed32_32_mul_int(ss_ds_frac_amount, 65536); - ds_data->ds_frac_amount = dal_fixed32_32_floor(ss_ds_frac_amount); + ss_ds_frac_amount = dc_fixpt_sub(ss_nslip_amount, + dc_fixpt_from_int(ds_data->nfrac_amount)); + ss_ds_frac_amount = dc_fixpt_mul_int(ss_ds_frac_amount, 65536); + ds_data->ds_frac_amount = dc_fixpt_floor(ss_ds_frac_amount); /* compute SS_STEP_SIZE_DSFRAC */ - modulation_time = dal_fixed32_32_from_fraction( + modulation_time = dc_fixpt_from_fraction( pll_settings->reference_freq * 1000, pll_settings->reference_divider * ss_data->modulation_freq_hz); if (ss_data->flags.CENTER_SPREAD) - modulation_time = dal_fixed32_32_div_int(modulation_time, 4); + modulation_time = dc_fixpt_div_int(modulation_time, 4); else - modulation_time = dal_fixed32_32_div_int(modulation_time, 2); + modulation_time = dc_fixpt_div_int(modulation_time, 2); - ss_step_size = dal_fixed32_32_div(ss_amount, modulation_time); + ss_step_size = dc_fixpt_div(ss_amount, modulation_time); /* SS_STEP_SIZE_DSFRAC_DEC = Int(SS_STEP_SIZE * 2 ^ 16 * 10)*/ - ss_step_size = dal_fixed32_32_mul_int(ss_step_size, 65536 * 10); - ds_data->ds_frac_size = dal_fixed32_32_floor(ss_step_size); + ss_step_size = dc_fixpt_mul_int(ss_step_size, 65536 * 10); + ds_data->ds_frac_size = dc_fixpt_floor(ss_step_size); return true; } @@ -982,9 +980,7 @@ static bool dce110_program_pix_clk( break; case DCE_VERSION_11_2: -#if defined(CONFIG_DRM_AMD_DC_VEGAM) case DCE_VERSION_11_22: -#endif case DCE_VERSION_12_0: #if defined(CONFIG_DRM_AMD_DC_DCN1_0) case DCN_VERSION_1_0: diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_clocks.c b/drivers/gpu/drm/amd/display/dc/dce/dce_clocks.c index 78e6beb6cf26..8a581c67bf2d 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_clocks.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_clocks.c @@ -26,7 +26,7 @@ #include "dce_clocks.h" #include "dm_services.h" #include "reg_helper.h" -#include "fixed32_32.h" +#include "fixed31_32.h" #include "bios_parser_interface.h" #include "dc.h" #include "dmcu.h" @@ -35,7 +35,7 @@ #endif #include "core_types.h" #include "dc_types.h" - +#include "dal_asic_id.h" #define TO_DCE_CLOCKS(clocks)\ container_of(clocks, struct dce_disp_clk, base) @@ -228,19 +228,19 @@ static int dce_clocks_get_dp_ref_freq(struct display_clock *clk) generated according to average value (case as with previous ASICs) */ if (clk_dce->ss_on_dprefclk && clk_dce->dprefclk_ss_divider != 0) { - struct fixed32_32 ss_percentage = dal_fixed32_32_div_int( - dal_fixed32_32_from_fraction( + struct fixed31_32 ss_percentage = dc_fixpt_div_int( + dc_fixpt_from_fraction( clk_dce->dprefclk_ss_percentage, clk_dce->dprefclk_ss_divider), 200); - struct fixed32_32 adj_dp_ref_clk_khz; + struct fixed31_32 adj_dp_ref_clk_khz; - ss_percentage = dal_fixed32_32_sub(dal_fixed32_32_one, + ss_percentage = dc_fixpt_sub(dc_fixpt_one, ss_percentage); adj_dp_ref_clk_khz = - dal_fixed32_32_mul_int( + dc_fixpt_mul_int( ss_percentage, dp_ref_clk_khz); - dp_ref_clk_khz = dal_fixed32_32_floor(adj_dp_ref_clk_khz); + dp_ref_clk_khz = dc_fixpt_floor(adj_dp_ref_clk_khz); } return dp_ref_clk_khz; @@ -256,19 +256,19 @@ static int dce_clocks_get_dp_ref_freq_wrkaround(struct display_clock *clk) int dp_ref_clk_khz = 600000; if (clk_dce->ss_on_dprefclk && clk_dce->dprefclk_ss_divider != 0) { - struct fixed32_32 ss_percentage = dal_fixed32_32_div_int( - dal_fixed32_32_from_fraction( + struct fixed31_32 ss_percentage = dc_fixpt_div_int( + dc_fixpt_from_fraction( clk_dce->dprefclk_ss_percentage, clk_dce->dprefclk_ss_divider), 200); - struct fixed32_32 adj_dp_ref_clk_khz; + struct fixed31_32 adj_dp_ref_clk_khz; - ss_percentage = dal_fixed32_32_sub(dal_fixed32_32_one, + ss_percentage = dc_fixpt_sub(dc_fixpt_one, ss_percentage); adj_dp_ref_clk_khz = - dal_fixed32_32_mul_int( + dc_fixpt_mul_int( ss_percentage, dp_ref_clk_khz); - dp_ref_clk_khz = dal_fixed32_32_floor(adj_dp_ref_clk_khz); + dp_ref_clk_khz = dc_fixpt_floor(adj_dp_ref_clk_khz); } return dp_ref_clk_khz; @@ -413,9 +413,12 @@ static int dce112_set_clock( /*VBIOS will determine DPREFCLK frequency, so we don't set it*/ dce_clk_params.target_clock_frequency = 0; dce_clk_params.clock_type = DCECLOCK_TYPE_DPREFCLK; - dce_clk_params.flags.USE_GENLOCK_AS_SOURCE_FOR_DPREFCLK = + if (!ASICREV_IS_VEGA20_P(clk->ctx->asic_id.hw_internal_rev)) + dce_clk_params.flags.USE_GENLOCK_AS_SOURCE_FOR_DPREFCLK = (dce_clk_params.pll_id == CLOCK_SOURCE_COMBO_DISPLAY_PLL0); + else + dce_clk_params.flags.USE_GENLOCK_AS_SOURCE_FOR_DPREFCLK = false; bp->funcs->set_dce_clock(bp, &dce_clk_params); diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_dmcu.c b/drivers/gpu/drm/amd/display/dc/dce/dce_dmcu.c index 2ee3d9bf1062..a576b8bbb3cd 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_dmcu.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_dmcu.c @@ -28,7 +28,7 @@ #include "dce_dmcu.h" #include "dm_services.h" #include "reg_helper.h" -#include "fixed32_32.h" +#include "fixed31_32.h" #include "dc.h" #define TO_DCE_DMCU(dmcu)\ diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_ipp.c b/drivers/gpu/drm/amd/display/dc/dce/dce_ipp.c index d737e911971b..5d9506b3d46b 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_ipp.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_ipp.c @@ -195,13 +195,13 @@ static void dce_ipp_program_input_lut( for (i = 0; i < gamma->num_entries; i++) { REG_SET(DC_LUT_SEQ_COLOR, 0, DC_LUT_SEQ_COLOR, - dal_fixed31_32_round( + dc_fixpt_round( gamma->entries.red[i])); REG_SET(DC_LUT_SEQ_COLOR, 0, DC_LUT_SEQ_COLOR, - dal_fixed31_32_round( + dc_fixpt_round( gamma->entries.green[i])); REG_SET(DC_LUT_SEQ_COLOR, 0, DC_LUT_SEQ_COLOR, - dal_fixed31_32_round( + dc_fixpt_round( gamma->entries.blue[i])); } diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_scl_filters.c b/drivers/gpu/drm/amd/display/dc/dce/dce_scl_filters.c index 6243450b41b7..48862bebf29e 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_scl_filters.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_scl_filters.c @@ -1014,11 +1014,11 @@ static const uint16_t filter_8tap_64p_183[264] = { const uint16_t *get_filter_3tap_16p(struct fixed31_32 ratio) { - if (ratio.value < dal_fixed31_32_one.value) + if (ratio.value < dc_fixpt_one.value) return filter_3tap_16p_upscale; - else if (ratio.value < dal_fixed31_32_from_fraction(4, 3).value) + else if (ratio.value < dc_fixpt_from_fraction(4, 3).value) return filter_3tap_16p_117; - else if (ratio.value < dal_fixed31_32_from_fraction(5, 3).value) + else if (ratio.value < dc_fixpt_from_fraction(5, 3).value) return filter_3tap_16p_150; else return filter_3tap_16p_183; @@ -1026,11 +1026,11 @@ const uint16_t *get_filter_3tap_16p(struct fixed31_32 ratio) const uint16_t *get_filter_3tap_64p(struct fixed31_32 ratio) { - if (ratio.value < dal_fixed31_32_one.value) + if (ratio.value < dc_fixpt_one.value) return filter_3tap_64p_upscale; - else if (ratio.value < dal_fixed31_32_from_fraction(4, 3).value) + else if (ratio.value < dc_fixpt_from_fraction(4, 3).value) return filter_3tap_64p_117; - else if (ratio.value < dal_fixed31_32_from_fraction(5, 3).value) + else if (ratio.value < dc_fixpt_from_fraction(5, 3).value) return filter_3tap_64p_150; else return filter_3tap_64p_183; @@ -1038,11 +1038,11 @@ const uint16_t *get_filter_3tap_64p(struct fixed31_32 ratio) const uint16_t *get_filter_4tap_16p(struct fixed31_32 ratio) { - if (ratio.value < dal_fixed31_32_one.value) + if (ratio.value < dc_fixpt_one.value) return filter_4tap_16p_upscale; - else if (ratio.value < dal_fixed31_32_from_fraction(4, 3).value) + else if (ratio.value < dc_fixpt_from_fraction(4, 3).value) return filter_4tap_16p_117; - else if (ratio.value < dal_fixed31_32_from_fraction(5, 3).value) + else if (ratio.value < dc_fixpt_from_fraction(5, 3).value) return filter_4tap_16p_150; else return filter_4tap_16p_183; @@ -1050,11 +1050,11 @@ const uint16_t *get_filter_4tap_16p(struct fixed31_32 ratio) const uint16_t *get_filter_4tap_64p(struct fixed31_32 ratio) { - if (ratio.value < dal_fixed31_32_one.value) + if (ratio.value < dc_fixpt_one.value) return filter_4tap_64p_upscale; - else if (ratio.value < dal_fixed31_32_from_fraction(4, 3).value) + else if (ratio.value < dc_fixpt_from_fraction(4, 3).value) return filter_4tap_64p_117; - else if (ratio.value < dal_fixed31_32_from_fraction(5, 3).value) + else if (ratio.value < dc_fixpt_from_fraction(5, 3).value) return filter_4tap_64p_150; else return filter_4tap_64p_183; @@ -1062,11 +1062,11 @@ const uint16_t *get_filter_4tap_64p(struct fixed31_32 ratio) const uint16_t *get_filter_5tap_64p(struct fixed31_32 ratio) { - if (ratio.value < dal_fixed31_32_one.value) + if (ratio.value < dc_fixpt_one.value) return filter_5tap_64p_upscale; - else if (ratio.value < dal_fixed31_32_from_fraction(4, 3).value) + else if (ratio.value < dc_fixpt_from_fraction(4, 3).value) return filter_5tap_64p_117; - else if (ratio.value < dal_fixed31_32_from_fraction(5, 3).value) + else if (ratio.value < dc_fixpt_from_fraction(5, 3).value) return filter_5tap_64p_150; else return filter_5tap_64p_183; @@ -1074,11 +1074,11 @@ const uint16_t *get_filter_5tap_64p(struct fixed31_32 ratio) const uint16_t *get_filter_6tap_64p(struct fixed31_32 ratio) { - if (ratio.value < dal_fixed31_32_one.value) + if (ratio.value < dc_fixpt_one.value) return filter_6tap_64p_upscale; - else if (ratio.value < dal_fixed31_32_from_fraction(4, 3).value) + else if (ratio.value < dc_fixpt_from_fraction(4, 3).value) return filter_6tap_64p_117; - else if (ratio.value < dal_fixed31_32_from_fraction(5, 3).value) + else if (ratio.value < dc_fixpt_from_fraction(5, 3).value) return filter_6tap_64p_150; else return filter_6tap_64p_183; @@ -1086,11 +1086,11 @@ const uint16_t *get_filter_6tap_64p(struct fixed31_32 ratio) const uint16_t *get_filter_7tap_64p(struct fixed31_32 ratio) { - if (ratio.value < dal_fixed31_32_one.value) + if (ratio.value < dc_fixpt_one.value) return filter_7tap_64p_upscale; - else if (ratio.value < dal_fixed31_32_from_fraction(4, 3).value) + else if (ratio.value < dc_fixpt_from_fraction(4, 3).value) return filter_7tap_64p_117; - else if (ratio.value < dal_fixed31_32_from_fraction(5, 3).value) + else if (ratio.value < dc_fixpt_from_fraction(5, 3).value) return filter_7tap_64p_150; else return filter_7tap_64p_183; @@ -1098,11 +1098,11 @@ const uint16_t *get_filter_7tap_64p(struct fixed31_32 ratio) const uint16_t *get_filter_8tap_64p(struct fixed31_32 ratio) { - if (ratio.value < dal_fixed31_32_one.value) + if (ratio.value < dc_fixpt_one.value) return filter_8tap_64p_upscale; - else if (ratio.value < dal_fixed31_32_from_fraction(4, 3).value) + else if (ratio.value < dc_fixpt_from_fraction(4, 3).value) return filter_8tap_64p_117; - else if (ratio.value < dal_fixed31_32_from_fraction(5, 3).value) + else if (ratio.value < dc_fixpt_from_fraction(5, 3).value) return filter_8tap_64p_150; else return filter_8tap_64p_183; diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_stream_encoder.c b/drivers/gpu/drm/amd/display/dc/dce/dce_stream_encoder.c index e265a0abe361..0a6d483dc046 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_stream_encoder.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_stream_encoder.c @@ -683,11 +683,11 @@ static void dce110_stream_encoder_set_mst_bandwidth( struct fixed31_32 avg_time_slots_per_mtp) { struct dce110_stream_encoder *enc110 = DCE110STRENC_FROM_STRENC(enc); - uint32_t x = dal_fixed31_32_floor( + uint32_t x = dc_fixpt_floor( avg_time_slots_per_mtp); - uint32_t y = dal_fixed31_32_ceil( - dal_fixed31_32_shl( - dal_fixed31_32_sub_int( + uint32_t y = dc_fixpt_ceil( + dc_fixpt_shl( + dc_fixpt_sub_int( avg_time_slots_per_mtp, x), 26)); diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_transform.c b/drivers/gpu/drm/amd/display/dc/dce/dce_transform.c index 832c5daada35..a02e719d7794 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_transform.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_transform.c @@ -41,7 +41,7 @@ #define DC_LOGGER \ xfm_dce->base.ctx->logger -#define IDENTITY_RATIO(ratio) (dal_fixed31_32_u2d19(ratio) == (1 << 19)) +#define IDENTITY_RATIO(ratio) (dc_fixpt_u2d19(ratio) == (1 << 19)) #define GAMUT_MATRIX_SIZE 12 #define SCL_PHASES 16 @@ -256,27 +256,27 @@ static void calculate_inits( struct fixed31_32 v_init; inits->h_int_scale_ratio = - dal_fixed31_32_u2d19(data->ratios.horz) << 5; + dc_fixpt_u2d19(data->ratios.horz) << 5; inits->v_int_scale_ratio = - dal_fixed31_32_u2d19(data->ratios.vert) << 5; + dc_fixpt_u2d19(data->ratios.vert) << 5; h_init = - dal_fixed31_32_div_int( - dal_fixed31_32_add( + dc_fixpt_div_int( + dc_fixpt_add( data->ratios.horz, - dal_fixed31_32_from_int(data->taps.h_taps + 1)), + dc_fixpt_from_int(data->taps.h_taps + 1)), 2); - inits->h_init.integer = dal_fixed31_32_floor(h_init); - inits->h_init.fraction = dal_fixed31_32_u0d19(h_init) << 5; + inits->h_init.integer = dc_fixpt_floor(h_init); + inits->h_init.fraction = dc_fixpt_u0d19(h_init) << 5; v_init = - dal_fixed31_32_div_int( - dal_fixed31_32_add( + dc_fixpt_div_int( + dc_fixpt_add( data->ratios.vert, - dal_fixed31_32_from_int(data->taps.v_taps + 1)), + dc_fixpt_from_int(data->taps.v_taps + 1)), 2); - inits->v_init.integer = dal_fixed31_32_floor(v_init); - inits->v_init.fraction = dal_fixed31_32_u0d19(v_init) << 5; + inits->v_init.integer = dc_fixpt_floor(v_init); + inits->v_init.fraction = dc_fixpt_u0d19(v_init) << 5; } static void program_scl_ratios_inits( diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c index 2288d0aa773b..a92fb0aa2ff3 100644 --- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c +++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c @@ -509,19 +509,19 @@ dce110_translate_regamma_to_hw_format(const struct dc_transfer_func *output_tf, rgb_resulted[hw_points - 1].green = output_tf->tf_pts.green[start_index]; rgb_resulted[hw_points - 1].blue = output_tf->tf_pts.blue[start_index]; - arr_points[0].x = dal_fixed31_32_pow(dal_fixed31_32_from_int(2), - dal_fixed31_32_from_int(region_start)); - arr_points[1].x = dal_fixed31_32_pow(dal_fixed31_32_from_int(2), - dal_fixed31_32_from_int(region_end)); + arr_points[0].x = dc_fixpt_pow(dc_fixpt_from_int(2), + dc_fixpt_from_int(region_start)); + arr_points[1].x = dc_fixpt_pow(dc_fixpt_from_int(2), + dc_fixpt_from_int(region_end)); y_r = rgb_resulted[0].red; y_g = rgb_resulted[0].green; y_b = rgb_resulted[0].blue; - y1_min = dal_fixed31_32_min(y_r, dal_fixed31_32_min(y_g, y_b)); + y1_min = dc_fixpt_min(y_r, dc_fixpt_min(y_g, y_b)); arr_points[0].y = y1_min; - arr_points[0].slope = dal_fixed31_32_div(arr_points[0].y, + arr_points[0].slope = dc_fixpt_div(arr_points[0].y, arr_points[0].x); y_r = rgb_resulted[hw_points - 1].red; @@ -531,21 +531,21 @@ dce110_translate_regamma_to_hw_format(const struct dc_transfer_func *output_tf, /* see comment above, m_arrPoints[1].y should be the Y value for the * region end (m_numOfHwPoints), not last HW point(m_numOfHwPoints - 1) */ - y3_max = dal_fixed31_32_max(y_r, dal_fixed31_32_max(y_g, y_b)); + y3_max = dc_fixpt_max(y_r, dc_fixpt_max(y_g, y_b)); arr_points[1].y = y3_max; - arr_points[1].slope = dal_fixed31_32_zero; + arr_points[1].slope = dc_fixpt_zero; if (output_tf->tf == TRANSFER_FUNCTION_PQ) { /* for PQ, we want to have a straight line from last HW X point, * and the slope to be such that we hit 1.0 at 10000 nits. */ - const struct fixed31_32 end_value = dal_fixed31_32_from_int(125); + const struct fixed31_32 end_value = dc_fixpt_from_int(125); - arr_points[1].slope = dal_fixed31_32_div( - dal_fixed31_32_sub(dal_fixed31_32_one, arr_points[1].y), - dal_fixed31_32_sub(end_value, arr_points[1].x)); + arr_points[1].slope = dc_fixpt_div( + dc_fixpt_sub(dc_fixpt_one, arr_points[1].y), + dc_fixpt_sub(end_value, arr_points[1].x)); } regamma_params->hw_points_num = hw_points; @@ -569,16 +569,16 @@ dce110_translate_regamma_to_hw_format(const struct dc_transfer_func *output_tf, i = 1; while (i != hw_points + 1) { - if (dal_fixed31_32_lt(rgb_plus_1->red, rgb->red)) + if (dc_fixpt_lt(rgb_plus_1->red, rgb->red)) rgb_plus_1->red = rgb->red; - if (dal_fixed31_32_lt(rgb_plus_1->green, rgb->green)) + if (dc_fixpt_lt(rgb_plus_1->green, rgb->green)) rgb_plus_1->green = rgb->green; - if (dal_fixed31_32_lt(rgb_plus_1->blue, rgb->blue)) + if (dc_fixpt_lt(rgb_plus_1->blue, rgb->blue)) rgb_plus_1->blue = rgb->blue; - rgb->delta_red = dal_fixed31_32_sub(rgb_plus_1->red, rgb->red); - rgb->delta_green = dal_fixed31_32_sub(rgb_plus_1->green, rgb->green); - rgb->delta_blue = dal_fixed31_32_sub(rgb_plus_1->blue, rgb->blue); + rgb->delta_red = dc_fixpt_sub(rgb_plus_1->red, rgb->red); + rgb->delta_green = dc_fixpt_sub(rgb_plus_1->green, rgb->green); + rgb->delta_blue = dc_fixpt_sub(rgb_plus_1->blue, rgb->blue); ++rgb_plus_1; ++rgb; @@ -2269,74 +2269,6 @@ static void program_gamut_remap(struct pipe_ctx *pipe_ctx) pipe_ctx->plane_res.xfm->funcs->transform_set_gamut_remap(pipe_ctx->plane_res.xfm, &adjust); } - -/** - * TODO REMOVE, USE UPDATE INSTEAD - */ -static void set_plane_config( - const struct dc *dc, - struct pipe_ctx *pipe_ctx, - struct resource_context *res_ctx) -{ - struct mem_input *mi = pipe_ctx->plane_res.mi; - struct dc_plane_state *plane_state = pipe_ctx->plane_state; - struct xfm_grph_csc_adjustment adjust; - struct out_csc_color_matrix tbl_entry; - unsigned int i; - - memset(&adjust, 0, sizeof(adjust)); - memset(&tbl_entry, 0, sizeof(tbl_entry)); - adjust.gamut_adjust_type = GRAPHICS_GAMUT_ADJUST_TYPE_BYPASS; - - dce_enable_fe_clock(dc->hwseq, mi->inst, true); - - set_default_colors(pipe_ctx); - if (pipe_ctx->stream->csc_color_matrix.enable_adjustment == true) { - tbl_entry.color_space = - pipe_ctx->stream->output_color_space; - - for (i = 0; i < 12; i++) - tbl_entry.regval[i] = - pipe_ctx->stream->csc_color_matrix.matrix[i]; - - pipe_ctx->plane_res.xfm->funcs->opp_set_csc_adjustment - (pipe_ctx->plane_res.xfm, &tbl_entry); - } - - if (pipe_ctx->stream->gamut_remap_matrix.enable_remap == true) { - adjust.gamut_adjust_type = GRAPHICS_GAMUT_ADJUST_TYPE_SW; - - for (i = 0; i < CSC_TEMPERATURE_MATRIX_SIZE; i++) - adjust.temperature_matrix[i] = - pipe_ctx->stream->gamut_remap_matrix.matrix[i]; - } - - pipe_ctx->plane_res.xfm->funcs->transform_set_gamut_remap(pipe_ctx->plane_res.xfm, &adjust); - - pipe_ctx->plane_res.scl_data.lb_params.alpha_en = pipe_ctx->bottom_pipe != 0; - program_scaler(dc, pipe_ctx); - - program_surface_visibility(dc, pipe_ctx); - - mi->funcs->mem_input_program_surface_config( - mi, - plane_state->format, - &plane_state->tiling_info, - &plane_state->plane_size, - plane_state->rotation, - NULL, - false); - if (mi->funcs->set_blank) - mi->funcs->set_blank(mi, pipe_ctx->plane_state->visible); - - if (dc->config.gpu_vm_support) - mi->funcs->mem_input_program_pte_vm( - pipe_ctx->plane_res.mi, - plane_state->format, - &plane_state->tiling_info, - plane_state->rotation); -} - static void update_plane_addr(const struct dc *dc, struct pipe_ctx *pipe_ctx) { @@ -3023,7 +2955,6 @@ static const struct hw_sequencer_funcs dce110_funcs = { .init_hw = init_hw, .apply_ctx_to_hw = dce110_apply_ctx_to_hw, .apply_ctx_for_surface = dce110_apply_ctx_for_surface, - .set_plane_config = set_plane_config, .update_plane_addr = update_plane_addr, .update_pending_status = dce110_update_pending_status, .set_input_transfer_func = dce110_set_input_transfer_func, diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_transform_v.c b/drivers/gpu/drm/amd/display/dc/dce110/dce110_transform_v.c index 8ba3c12fc608..a7dce060204f 100644 --- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_transform_v.c +++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_transform_v.c @@ -373,13 +373,13 @@ static void calculate_inits( struct rect *chroma_viewport) { inits->h_int_scale_ratio_luma = - dal_fixed31_32_u2d19(data->ratios.horz) << 5; + dc_fixpt_u2d19(data->ratios.horz) << 5; inits->v_int_scale_ratio_luma = - dal_fixed31_32_u2d19(data->ratios.vert) << 5; + dc_fixpt_u2d19(data->ratios.vert) << 5; inits->h_int_scale_ratio_chroma = - dal_fixed31_32_u2d19(data->ratios.horz_c) << 5; + dc_fixpt_u2d19(data->ratios.horz_c) << 5; inits->v_int_scale_ratio_chroma = - dal_fixed31_32_u2d19(data->ratios.vert_c) << 5; + dc_fixpt_u2d19(data->ratios.vert_c) << 5; inits->h_init_luma.integer = 1; inits->v_init_luma.integer = 1; diff --git a/drivers/gpu/drm/amd/display/dc/dce120/dce120_resource.c b/drivers/gpu/drm/amd/display/dc/dce120/dce120_resource.c index fda01574d1ba..2d58daccc005 100644 --- a/drivers/gpu/drm/amd/display/dc/dce120/dce120_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dce120/dce120_resource.c @@ -814,14 +814,25 @@ static void bw_calcs_data_update_from_pplib(struct dc *dc) dm_pp_notify_wm_clock_changes(dc->ctx, &clk_ranges); } +static uint32_t read_pipe_fuses(struct dc_context *ctx) +{ + uint32_t value = dm_read_reg_soc15(ctx, mmCC_DC_PIPE_DIS, 0); + /* VG20 support max 6 pipes */ + value = value & 0x3f; + return value; +} + static bool construct( uint8_t num_virtual_links, struct dc *dc, struct dce110_resource_pool *pool) { unsigned int i; + int j; struct dc_context *ctx = dc->ctx; struct irq_service_init_data irq_init_data; + bool harvest_enabled = ASICREV_IS_VEGA20_P(ctx->asic_id.hw_internal_rev); + uint32_t pipe_fuses; ctx->dc_bios->regs = &bios_regs; @@ -915,28 +926,41 @@ static bool construct( if (!pool->base.irqs) goto irqs_create_fail; + /* retrieve valid pipe fuses */ + if (harvest_enabled) + pipe_fuses = read_pipe_fuses(ctx); + + /* index to valid pipe resource */ + j = 0; for (i = 0; i < pool->base.pipe_count; i++) { - pool->base.timing_generators[i] = + if (harvest_enabled) { + if ((pipe_fuses & (1 << i)) != 0) { + dm_error("DC: skip invalid pipe %d!\n", i); + continue; + } + } + + pool->base.timing_generators[j] = dce120_timing_generator_create( ctx, i, &dce120_tg_offsets[i]); - if (pool->base.timing_generators[i] == NULL) { + if (pool->base.timing_generators[j] == NULL) { BREAK_TO_DEBUGGER(); dm_error("DC: failed to create tg!\n"); goto controller_create_fail; } - pool->base.mis[i] = dce120_mem_input_create(ctx, i); + pool->base.mis[j] = dce120_mem_input_create(ctx, i); - if (pool->base.mis[i] == NULL) { + if (pool->base.mis[j] == NULL) { BREAK_TO_DEBUGGER(); dm_error( "DC: failed to create memory input!\n"); goto controller_create_fail; } - pool->base.ipps[i] = dce120_ipp_create(ctx, i); + pool->base.ipps[j] = dce120_ipp_create(ctx, i); if (pool->base.ipps[i] == NULL) { BREAK_TO_DEBUGGER(); dm_error( @@ -944,7 +968,7 @@ static bool construct( goto controller_create_fail; } - pool->base.transforms[i] = dce120_transform_create(ctx, i); + pool->base.transforms[j] = dce120_transform_create(ctx, i); if (pool->base.transforms[i] == NULL) { BREAK_TO_DEBUGGER(); dm_error( @@ -952,16 +976,23 @@ static bool construct( goto res_create_fail; } - pool->base.opps[i] = dce120_opp_create( + pool->base.opps[j] = dce120_opp_create( ctx, i); - if (pool->base.opps[i] == NULL) { + if (pool->base.opps[j] == NULL) { BREAK_TO_DEBUGGER(); dm_error( "DC: failed to create output pixel processor!\n"); } + + /* check next valid pipe */ + j++; } + /* valid pipe num */ + pool->base.pipe_count = j; + pool->base.timing_generator_count = j; + if (!resource_construct(num_virtual_links, dc, &pool->base, &res_create_funcs)) goto res_create_fail; diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/Makefile b/drivers/gpu/drm/amd/display/dc/dcn10/Makefile index 5c69743a4b4f..84f52c63d95c 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/Makefile +++ b/drivers/gpu/drm/amd/display/dc/dcn10/Makefile @@ -26,7 +26,7 @@ DCN10 = dcn10_resource.o dcn10_ipp.o dcn10_hw_sequencer.o \ dcn10_dpp.o dcn10_opp.o dcn10_optc.o \ dcn10_hubp.o dcn10_mpc.o \ dcn10_dpp_dscl.o dcn10_dpp_cm.o dcn10_cm_common.o \ - dcn10_hubbub.o dcn10_stream_encoder.o + dcn10_hubbub.o dcn10_stream_encoder.o dcn10_link_encoder.o AMD_DAL_DCN10 = $(addprefix $(AMDDALPATH)/dc/dcn10/,$(DCN10)) diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_cm_common.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_cm_common.c index 96d5878e9ccd..5d95a997fd9f 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_cm_common.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_cm_common.c @@ -169,7 +169,7 @@ bool cm_helper_convert_to_custom_float( } if (fixpoint == true) - arr_points[1].custom_float_y = dal_fixed31_32_clamp_u0d14(arr_points[1].y); + arr_points[1].custom_float_y = dc_fixpt_clamp_u0d14(arr_points[1].y); else if (!convert_to_custom_float_format(arr_points[1].y, &fmt, &arr_points[1].custom_float_y)) { BREAK_TO_DEBUGGER(); @@ -327,19 +327,19 @@ bool cm_helper_translate_curve_to_hw_format( rgb_resulted[hw_points - 1].green = output_tf->tf_pts.green[start_index]; rgb_resulted[hw_points - 1].blue = output_tf->tf_pts.blue[start_index]; - arr_points[0].x = dal_fixed31_32_pow(dal_fixed31_32_from_int(2), - dal_fixed31_32_from_int(region_start)); - arr_points[1].x = dal_fixed31_32_pow(dal_fixed31_32_from_int(2), - dal_fixed31_32_from_int(region_end)); + arr_points[0].x = dc_fixpt_pow(dc_fixpt_from_int(2), + dc_fixpt_from_int(region_start)); + arr_points[1].x = dc_fixpt_pow(dc_fixpt_from_int(2), + dc_fixpt_from_int(region_end)); y_r = rgb_resulted[0].red; y_g = rgb_resulted[0].green; y_b = rgb_resulted[0].blue; - y1_min = dal_fixed31_32_min(y_r, dal_fixed31_32_min(y_g, y_b)); + y1_min = dc_fixpt_min(y_r, dc_fixpt_min(y_g, y_b)); arr_points[0].y = y1_min; - arr_points[0].slope = dal_fixed31_32_div(arr_points[0].y, arr_points[0].x); + arr_points[0].slope = dc_fixpt_div(arr_points[0].y, arr_points[0].x); y_r = rgb_resulted[hw_points - 1].red; y_g = rgb_resulted[hw_points - 1].green; y_b = rgb_resulted[hw_points - 1].blue; @@ -347,22 +347,22 @@ bool cm_helper_translate_curve_to_hw_format( /* see comment above, m_arrPoints[1].y should be the Y value for the * region end (m_numOfHwPoints), not last HW point(m_numOfHwPoints - 1) */ - y3_max = dal_fixed31_32_max(y_r, dal_fixed31_32_max(y_g, y_b)); + y3_max = dc_fixpt_max(y_r, dc_fixpt_max(y_g, y_b)); arr_points[1].y = y3_max; - arr_points[1].slope = dal_fixed31_32_zero; + arr_points[1].slope = dc_fixpt_zero; if (output_tf->tf == TRANSFER_FUNCTION_PQ) { /* for PQ, we want to have a straight line from last HW X point, * and the slope to be such that we hit 1.0 at 10000 nits. */ const struct fixed31_32 end_value = - dal_fixed31_32_from_int(125); + dc_fixpt_from_int(125); - arr_points[1].slope = dal_fixed31_32_div( - dal_fixed31_32_sub(dal_fixed31_32_one, arr_points[1].y), - dal_fixed31_32_sub(end_value, arr_points[1].x)); + arr_points[1].slope = dc_fixpt_div( + dc_fixpt_sub(dc_fixpt_one, arr_points[1].y), + dc_fixpt_sub(end_value, arr_points[1].x)); } lut_params->hw_points_num = hw_points; @@ -386,24 +386,24 @@ bool cm_helper_translate_curve_to_hw_format( i = 1; while (i != hw_points + 1) { - if (dal_fixed31_32_lt(rgb_plus_1->red, rgb->red)) + if (dc_fixpt_lt(rgb_plus_1->red, rgb->red)) rgb_plus_1->red = rgb->red; - if (dal_fixed31_32_lt(rgb_plus_1->green, rgb->green)) + if (dc_fixpt_lt(rgb_plus_1->green, rgb->green)) rgb_plus_1->green = rgb->green; - if (dal_fixed31_32_lt(rgb_plus_1->blue, rgb->blue)) + if (dc_fixpt_lt(rgb_plus_1->blue, rgb->blue)) rgb_plus_1->blue = rgb->blue; - rgb->delta_red = dal_fixed31_32_sub(rgb_plus_1->red, rgb->red); - rgb->delta_green = dal_fixed31_32_sub(rgb_plus_1->green, rgb->green); - rgb->delta_blue = dal_fixed31_32_sub(rgb_plus_1->blue, rgb->blue); + rgb->delta_red = dc_fixpt_sub(rgb_plus_1->red, rgb->red); + rgb->delta_green = dc_fixpt_sub(rgb_plus_1->green, rgb->green); + rgb->delta_blue = dc_fixpt_sub(rgb_plus_1->blue, rgb->blue); if (fixpoint == true) { - rgb->delta_red_reg = dal_fixed31_32_clamp_u0d10(rgb->delta_red); - rgb->delta_green_reg = dal_fixed31_32_clamp_u0d10(rgb->delta_green); - rgb->delta_blue_reg = dal_fixed31_32_clamp_u0d10(rgb->delta_blue); - rgb->red_reg = dal_fixed31_32_clamp_u0d14(rgb->red); - rgb->green_reg = dal_fixed31_32_clamp_u0d14(rgb->green); - rgb->blue_reg = dal_fixed31_32_clamp_u0d14(rgb->blue); + rgb->delta_red_reg = dc_fixpt_clamp_u0d10(rgb->delta_red); + rgb->delta_green_reg = dc_fixpt_clamp_u0d10(rgb->delta_green); + rgb->delta_blue_reg = dc_fixpt_clamp_u0d10(rgb->delta_blue); + rgb->red_reg = dc_fixpt_clamp_u0d14(rgb->red); + rgb->green_reg = dc_fixpt_clamp_u0d14(rgb->green); + rgb->blue_reg = dc_fixpt_clamp_u0d14(rgb->blue); } ++rgb_plus_1; @@ -489,19 +489,19 @@ bool cm_helper_translate_curve_to_degamma_hw_format( rgb_resulted[hw_points - 1].green = output_tf->tf_pts.green[start_index]; rgb_resulted[hw_points - 1].blue = output_tf->tf_pts.blue[start_index]; - arr_points[0].x = dal_fixed31_32_pow(dal_fixed31_32_from_int(2), - dal_fixed31_32_from_int(region_start)); - arr_points[1].x = dal_fixed31_32_pow(dal_fixed31_32_from_int(2), - dal_fixed31_32_from_int(region_end)); + arr_points[0].x = dc_fixpt_pow(dc_fixpt_from_int(2), + dc_fixpt_from_int(region_start)); + arr_points[1].x = dc_fixpt_pow(dc_fixpt_from_int(2), + dc_fixpt_from_int(region_end)); y_r = rgb_resulted[0].red; y_g = rgb_resulted[0].green; y_b = rgb_resulted[0].blue; - y1_min = dal_fixed31_32_min(y_r, dal_fixed31_32_min(y_g, y_b)); + y1_min = dc_fixpt_min(y_r, dc_fixpt_min(y_g, y_b)); arr_points[0].y = y1_min; - arr_points[0].slope = dal_fixed31_32_div(arr_points[0].y, arr_points[0].x); + arr_points[0].slope = dc_fixpt_div(arr_points[0].y, arr_points[0].x); y_r = rgb_resulted[hw_points - 1].red; y_g = rgb_resulted[hw_points - 1].green; y_b = rgb_resulted[hw_points - 1].blue; @@ -509,22 +509,22 @@ bool cm_helper_translate_curve_to_degamma_hw_format( /* see comment above, m_arrPoints[1].y should be the Y value for the * region end (m_numOfHwPoints), not last HW point(m_numOfHwPoints - 1) */ - y3_max = dal_fixed31_32_max(y_r, dal_fixed31_32_max(y_g, y_b)); + y3_max = dc_fixpt_max(y_r, dc_fixpt_max(y_g, y_b)); arr_points[1].y = y3_max; - arr_points[1].slope = dal_fixed31_32_zero; + arr_points[1].slope = dc_fixpt_zero; if (output_tf->tf == TRANSFER_FUNCTION_PQ) { /* for PQ, we want to have a straight line from last HW X point, * and the slope to be such that we hit 1.0 at 10000 nits. */ const struct fixed31_32 end_value = - dal_fixed31_32_from_int(125); + dc_fixpt_from_int(125); - arr_points[1].slope = dal_fixed31_32_div( - dal_fixed31_32_sub(dal_fixed31_32_one, arr_points[1].y), - dal_fixed31_32_sub(end_value, arr_points[1].x)); + arr_points[1].slope = dc_fixpt_div( + dc_fixpt_sub(dc_fixpt_one, arr_points[1].y), + dc_fixpt_sub(end_value, arr_points[1].x)); } lut_params->hw_points_num = hw_points; @@ -548,16 +548,16 @@ bool cm_helper_translate_curve_to_degamma_hw_format( i = 1; while (i != hw_points + 1) { - if (dal_fixed31_32_lt(rgb_plus_1->red, rgb->red)) + if (dc_fixpt_lt(rgb_plus_1->red, rgb->red)) rgb_plus_1->red = rgb->red; - if (dal_fixed31_32_lt(rgb_plus_1->green, rgb->green)) + if (dc_fixpt_lt(rgb_plus_1->green, rgb->green)) rgb_plus_1->green = rgb->green; - if (dal_fixed31_32_lt(rgb_plus_1->blue, rgb->blue)) + if (dc_fixpt_lt(rgb_plus_1->blue, rgb->blue)) rgb_plus_1->blue = rgb->blue; - rgb->delta_red = dal_fixed31_32_sub(rgb_plus_1->red, rgb->red); - rgb->delta_green = dal_fixed31_32_sub(rgb_plus_1->green, rgb->green); - rgb->delta_blue = dal_fixed31_32_sub(rgb_plus_1->blue, rgb->blue); + rgb->delta_red = dc_fixpt_sub(rgb_plus_1->red, rgb->red); + rgb->delta_green = dc_fixpt_sub(rgb_plus_1->green, rgb->green); + rgb->delta_blue = dc_fixpt_sub(rgb_plus_1->blue, rgb->blue); ++rgb_plus_1; ++rgb; diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.c index 8c4d9e523331..46a35c7f01df 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.c @@ -130,7 +130,7 @@ void dpp_set_gamut_remap_bypass(struct dcn10_dpp *dpp) /* Gamut remap in bypass */ } -#define IDENTITY_RATIO(ratio) (dal_fixed31_32_u2d19(ratio) == (1 << 19)) +#define IDENTITY_RATIO(ratio) (dc_fixpt_u2d19(ratio) == (1 << 19)) bool dpp_get_optimal_number_of_taps( @@ -152,6 +152,11 @@ bool dpp_get_optimal_number_of_taps( scl_data->format == PIXEL_FORMAT_FP16) return false; + if (scl_data->viewport.width > scl_data->h_active && + dpp->ctx->dc->debug.max_downscale_src_width != 0 && + scl_data->viewport.width > dpp->ctx->dc->debug.max_downscale_src_width) + return false; + /* TODO: add lb check */ /* No support for programming ratio of 4, drop to 3.99999.. */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp_cm.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp_cm.c index 4f373c97804f..116977eb24e2 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp_cm.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp_cm.c @@ -811,13 +811,13 @@ void dpp1_program_input_lut( REG_UPDATE(CM_IGAM_LUT_RW_INDEX, CM_IGAM_LUT_RW_INDEX, 0); for (i = 0; i < gamma->num_entries; i++) { REG_SET(CM_IGAM_LUT_SEQ_COLOR, 0, CM_IGAM_LUT_SEQ_COLOR, - dal_fixed31_32_round( + dc_fixpt_round( gamma->entries.red[i])); REG_SET(CM_IGAM_LUT_SEQ_COLOR, 0, CM_IGAM_LUT_SEQ_COLOR, - dal_fixed31_32_round( + dc_fixpt_round( gamma->entries.green[i])); REG_SET(CM_IGAM_LUT_SEQ_COLOR, 0, CM_IGAM_LUT_SEQ_COLOR, - dal_fixed31_32_round( + dc_fixpt_round( gamma->entries.blue[i])); } // Power off LUT memory diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp_dscl.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp_dscl.c index 3eb824debf43..4ddd6273d5a5 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp_dscl.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp_dscl.c @@ -169,7 +169,7 @@ static enum dscl_mode_sel dpp1_dscl_get_dscl_mode( const struct scaler_data *data, bool dbg_always_scale) { - const long long one = dal_fixed31_32_one.value; + const long long one = dc_fixpt_one.value; if (dpp_base->caps->dscl_data_proc_format == DSCL_DATA_PRCESSING_FIXED_FORMAT) { /* DSCL is processing data in fixed format */ @@ -464,8 +464,8 @@ static enum lb_memory_config dpp1_dscl_find_lb_memory_config(struct dcn10_dpp *d int num_part_y, num_part_c; int vtaps = scl_data->taps.v_taps; int vtaps_c = scl_data->taps.v_taps_c; - int ceil_vratio = dal_fixed31_32_ceil(scl_data->ratios.vert); - int ceil_vratio_c = dal_fixed31_32_ceil(scl_data->ratios.vert_c); + int ceil_vratio = dc_fixpt_ceil(scl_data->ratios.vert); + int ceil_vratio_c = dc_fixpt_ceil(scl_data->ratios.vert_c); enum lb_memory_config mem_cfg = LB_MEMORY_CONFIG_0; if (dpp->base.ctx->dc->debug.use_max_lb) @@ -565,52 +565,52 @@ static void dpp1_dscl_set_manual_ratio_init( uint32_t init_int = 0; REG_SET(SCL_HORZ_FILTER_SCALE_RATIO, 0, - SCL_H_SCALE_RATIO, dal_fixed31_32_u2d19(data->ratios.horz) << 5); + SCL_H_SCALE_RATIO, dc_fixpt_u2d19(data->ratios.horz) << 5); REG_SET(SCL_VERT_FILTER_SCALE_RATIO, 0, - SCL_V_SCALE_RATIO, dal_fixed31_32_u2d19(data->ratios.vert) << 5); + SCL_V_SCALE_RATIO, dc_fixpt_u2d19(data->ratios.vert) << 5); REG_SET(SCL_HORZ_FILTER_SCALE_RATIO_C, 0, - SCL_H_SCALE_RATIO_C, dal_fixed31_32_u2d19(data->ratios.horz_c) << 5); + SCL_H_SCALE_RATIO_C, dc_fixpt_u2d19(data->ratios.horz_c) << 5); REG_SET(SCL_VERT_FILTER_SCALE_RATIO_C, 0, - SCL_V_SCALE_RATIO_C, dal_fixed31_32_u2d19(data->ratios.vert_c) << 5); + SCL_V_SCALE_RATIO_C, dc_fixpt_u2d19(data->ratios.vert_c) << 5); /* * 0.24 format for fraction, first five bits zeroed */ - init_frac = dal_fixed31_32_u0d19(data->inits.h) << 5; - init_int = dal_fixed31_32_floor(data->inits.h); + init_frac = dc_fixpt_u0d19(data->inits.h) << 5; + init_int = dc_fixpt_floor(data->inits.h); REG_SET_2(SCL_HORZ_FILTER_INIT, 0, SCL_H_INIT_FRAC, init_frac, SCL_H_INIT_INT, init_int); - init_frac = dal_fixed31_32_u0d19(data->inits.h_c) << 5; - init_int = dal_fixed31_32_floor(data->inits.h_c); + init_frac = dc_fixpt_u0d19(data->inits.h_c) << 5; + init_int = dc_fixpt_floor(data->inits.h_c); REG_SET_2(SCL_HORZ_FILTER_INIT_C, 0, SCL_H_INIT_FRAC_C, init_frac, SCL_H_INIT_INT_C, init_int); - init_frac = dal_fixed31_32_u0d19(data->inits.v) << 5; - init_int = dal_fixed31_32_floor(data->inits.v); + init_frac = dc_fixpt_u0d19(data->inits.v) << 5; + init_int = dc_fixpt_floor(data->inits.v); REG_SET_2(SCL_VERT_FILTER_INIT, 0, SCL_V_INIT_FRAC, init_frac, SCL_V_INIT_INT, init_int); - init_frac = dal_fixed31_32_u0d19(data->inits.v_bot) << 5; - init_int = dal_fixed31_32_floor(data->inits.v_bot); + init_frac = dc_fixpt_u0d19(data->inits.v_bot) << 5; + init_int = dc_fixpt_floor(data->inits.v_bot); REG_SET_2(SCL_VERT_FILTER_INIT_BOT, 0, SCL_V_INIT_FRAC_BOT, init_frac, SCL_V_INIT_INT_BOT, init_int); - init_frac = dal_fixed31_32_u0d19(data->inits.v_c) << 5; - init_int = dal_fixed31_32_floor(data->inits.v_c); + init_frac = dc_fixpt_u0d19(data->inits.v_c) << 5; + init_int = dc_fixpt_floor(data->inits.v_c); REG_SET_2(SCL_VERT_FILTER_INIT_C, 0, SCL_V_INIT_FRAC_C, init_frac, SCL_V_INIT_INT_C, init_int); - init_frac = dal_fixed31_32_u0d19(data->inits.v_c_bot) << 5; - init_int = dal_fixed31_32_floor(data->inits.v_c_bot); + init_frac = dc_fixpt_u0d19(data->inits.v_c_bot) << 5; + init_int = dc_fixpt_floor(data->inits.v_c_bot); REG_SET_2(SCL_VERT_FILTER_INIT_BOT_C, 0, SCL_V_INIT_FRAC_BOT_C, init_frac, SCL_V_INIT_INT_BOT_C, init_int); diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.c index b9fb14a3224b..943143efbb82 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.c @@ -476,6 +476,14 @@ void hubbub1_toggle_watermark_change_req(struct hubbub *hubbub) DCHUBBUB_ARB_WATERMARK_CHANGE_REQUEST, watermark_change_req); } +void hubbub1_soft_reset(struct hubbub *hubbub, bool reset) +{ + uint32_t reset_en = reset ? 1 : 0; + + REG_UPDATE(DCHUBBUB_SOFT_RESET, + DCHUBBUB_GLOBAL_SOFT_RESET, reset_en); +} + static bool hubbub1_dcc_support_swizzle( enum swizzle_mode_values swizzle, unsigned int bytes_per_element, diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.h index f479f54e5bb2..6315a0e6b0d6 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.h +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.h @@ -48,7 +48,8 @@ SR(DCHUBBUB_ARB_DF_REQ_OUTSTAND),\ SR(DCHUBBUB_GLOBAL_TIMER_CNTL), \ SR(DCHUBBUB_TEST_DEBUG_INDEX), \ - SR(DCHUBBUB_TEST_DEBUG_DATA) + SR(DCHUBBUB_TEST_DEBUG_DATA),\ + SR(DCHUBBUB_SOFT_RESET) #define HUBBUB_SR_WATERMARK_REG_LIST()\ SR(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_A),\ @@ -105,6 +106,7 @@ struct dcn_hubbub_registers { uint32_t DCHUBBUB_SDPIF_AGP_BOT; uint32_t DCHUBBUB_SDPIF_AGP_TOP; uint32_t DCHUBBUB_CRC_CTRL; + uint32_t DCHUBBUB_SOFT_RESET; }; /* set field name */ @@ -114,6 +116,7 @@ struct dcn_hubbub_registers { #define HUBBUB_MASK_SH_LIST_DCN(mask_sh)\ HUBBUB_SF(DCHUBBUB_GLOBAL_TIMER_CNTL, DCHUBBUB_GLOBAL_TIMER_ENABLE, mask_sh), \ + HUBBUB_SF(DCHUBBUB_SOFT_RESET, DCHUBBUB_GLOBAL_SOFT_RESET, mask_sh), \ HUBBUB_SF(DCHUBBUB_ARB_WATERMARK_CHANGE_CNTL, DCHUBBUB_ARB_WATERMARK_CHANGE_REQUEST, mask_sh), \ HUBBUB_SF(DCHUBBUB_ARB_WATERMARK_CHANGE_CNTL, DCHUBBUB_ARB_WATERMARK_CHANGE_DONE_INTERRUPT_DISABLE, mask_sh), \ HUBBUB_SF(DCHUBBUB_ARB_DRAM_STATE_CNTL, DCHUBBUB_ARB_ALLOW_SELF_REFRESH_FORCE_VALUE, mask_sh), \ @@ -143,6 +146,7 @@ struct dcn_hubbub_registers { type DCHUBBUB_ARB_SAT_LEVEL;\ type DCHUBBUB_ARB_MIN_REQ_OUTSTAND;\ type DCHUBBUB_GLOBAL_TIMER_REFDIV;\ + type DCHUBBUB_GLOBAL_SOFT_RESET; \ type SDPIF_FB_TOP;\ type SDPIF_FB_BASE;\ type SDPIF_FB_OFFSET;\ @@ -201,6 +205,7 @@ void hubbub1_toggle_watermark_change_req( void hubbub1_wm_read_state(struct hubbub *hubbub, struct dcn_hubbub_wm *wm); +void hubbub1_soft_reset(struct hubbub *hubbub, bool reset); void hubbub1_construct(struct hubbub *hubbub, struct dc_context *ctx, const struct dcn_hubbub_registers *hubbub_regs, diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c index 0cbc83edd37f..d2ab78b35a7a 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c @@ -78,6 +78,27 @@ static void hubp1_disconnect(struct hubp *hubp) CURSOR_ENABLE, 0); } +static void hubp1_disable_control(struct hubp *hubp, bool disable_hubp) +{ + struct dcn10_hubp *hubp1 = TO_DCN10_HUBP(hubp); + uint32_t disable = disable_hubp ? 1 : 0; + + REG_UPDATE(DCHUBP_CNTL, + HUBP_DISABLE, disable); +} + +static unsigned int hubp1_get_underflow_status(struct hubp *hubp) +{ + uint32_t hubp_underflow = 0; + struct dcn10_hubp *hubp1 = TO_DCN10_HUBP(hubp); + + REG_GET(DCHUBP_CNTL, + HUBP_UNDERFLOW_STATUS, + &hubp_underflow); + + return hubp_underflow; +} + static void hubp1_set_hubp_blank_en(struct hubp *hubp, bool blank) { struct dcn10_hubp *hubp1 = TO_DCN10_HUBP(hubp); @@ -1054,8 +1075,8 @@ void hubp1_cursor_set_position( ASSERT(param->h_scale_ratio.value); if (param->h_scale_ratio.value) - dst_x_offset = dal_fixed31_32_floor(dal_fixed31_32_div( - dal_fixed31_32_from_int(dst_x_offset), + dst_x_offset = dc_fixpt_floor(dc_fixpt_div( + dc_fixpt_from_int(dst_x_offset), param->h_scale_ratio)); if (src_x_offset >= (int)param->viewport_width) @@ -1117,6 +1138,9 @@ static struct hubp_funcs dcn10_hubp_funcs = { .hubp_clk_cntl = hubp1_clk_cntl, .hubp_vtg_sel = hubp1_vtg_sel, .hubp_read_state = hubp1_read_state, + .hubp_disable_control = hubp1_disable_control, + .hubp_get_underflow_status = hubp1_get_underflow_status, + }; /*****************************************/ diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.h index fe9b8c4a91ca..af384034398f 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.h +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.h @@ -253,6 +253,7 @@ HUBP_SF(HUBP0_DCHUBP_CNTL, HUBP_UNDERFLOW_STATUS, mask_sh),\ HUBP_SF(HUBP0_DCHUBP_CNTL, HUBP_NO_OUTSTANDING_REQ, mask_sh),\ HUBP_SF(HUBP0_DCHUBP_CNTL, HUBP_VTG_SEL, mask_sh),\ + HUBP_SF(HUBP0_DCHUBP_CNTL, HUBP_DISABLE, mask_sh),\ HUBP_SF(HUBP0_DCSURF_ADDR_CONFIG, NUM_PIPES, mask_sh),\ HUBP_SF(HUBP0_DCSURF_ADDR_CONFIG, NUM_BANKS, mask_sh),\ HUBP_SF(HUBP0_DCSURF_ADDR_CONFIG, PIPE_INTERLEAVE, mask_sh),\ @@ -421,6 +422,7 @@ #define DCN_HUBP_REG_FIELD_LIST(type) \ type HUBP_BLANK_EN;\ + type HUBP_DISABLE;\ type HUBP_TTU_DISABLE;\ type HUBP_NO_OUTSTANDING_REQ;\ type HUBP_VTG_SEL;\ @@ -723,4 +725,5 @@ void hubp1_read_state(struct hubp *hubp); enum cursor_pitch hubp1_get_cursor_pitch(unsigned int pitch); + #endif diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c index 572fa601a0eb..f8e0576af6e0 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c @@ -127,24 +127,26 @@ static void dcn10_log_hubp_states(struct dc *dc) hubp->funcs->hubp_read_state(hubp); - DTN_INFO("[%2d]: %5xh %6xh %5d %6d %2xh %2xh %6xh" - " %6d %8d %7d %8xh", - hubp->inst, - s->pixel_format, - s->inuse_addr_hi, - s->viewport_width, - s->viewport_height, - s->rotation_angle, - s->h_mirror_en, - s->sw_mode, - s->dcc_en, - s->blank_en, - s->ttu_disable, - s->underflow_status); - DTN_INFO_MICRO_SEC(s->min_ttu_vblank); - DTN_INFO_MICRO_SEC(s->qos_level_low_wm); - DTN_INFO_MICRO_SEC(s->qos_level_high_wm); - DTN_INFO("\n"); + if (!s->blank_en) { + DTN_INFO("[%2d]: %5xh %6xh %5d %6d %2xh %2xh %6xh" + " %6d %8d %7d %8xh", + hubp->inst, + s->pixel_format, + s->inuse_addr_hi, + s->viewport_width, + s->viewport_height, + s->rotation_angle, + s->h_mirror_en, + s->sw_mode, + s->dcc_en, + s->blank_en, + s->ttu_disable, + s->underflow_status); + DTN_INFO_MICRO_SEC(s->min_ttu_vblank); + DTN_INFO_MICRO_SEC(s->qos_level_low_wm); + DTN_INFO_MICRO_SEC(s->qos_level_high_wm); + DTN_INFO("\n"); + } } DTN_INFO("\n=========RQ========\n"); @@ -155,16 +157,17 @@ static void dcn10_log_hubp_states(struct dc *dc) struct dcn_hubp_state *s = &(TO_DCN10_HUBP(pool->hubps[i])->state); struct _vcs_dpi_display_rq_regs_st *rq_regs = &s->rq_regs; - DTN_INFO("[%2d]: %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh\n", - i, rq_regs->drq_expansion_mode, rq_regs->prq_expansion_mode, rq_regs->mrq_expansion_mode, - rq_regs->crq_expansion_mode, rq_regs->plane1_base_address, rq_regs->rq_regs_l.chunk_size, - rq_regs->rq_regs_l.min_chunk_size, rq_regs->rq_regs_l.meta_chunk_size, - rq_regs->rq_regs_l.min_meta_chunk_size, rq_regs->rq_regs_l.dpte_group_size, - rq_regs->rq_regs_l.mpte_group_size, rq_regs->rq_regs_l.swath_height, - rq_regs->rq_regs_l.pte_row_height_linear, rq_regs->rq_regs_c.chunk_size, rq_regs->rq_regs_c.min_chunk_size, - rq_regs->rq_regs_c.meta_chunk_size, rq_regs->rq_regs_c.min_meta_chunk_size, - rq_regs->rq_regs_c.dpte_group_size, rq_regs->rq_regs_c.mpte_group_size, - rq_regs->rq_regs_c.swath_height, rq_regs->rq_regs_c.pte_row_height_linear); + if (!s->blank_en) + DTN_INFO("[%2d]: %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh\n", + pool->hubps[i]->inst, rq_regs->drq_expansion_mode, rq_regs->prq_expansion_mode, rq_regs->mrq_expansion_mode, + rq_regs->crq_expansion_mode, rq_regs->plane1_base_address, rq_regs->rq_regs_l.chunk_size, + rq_regs->rq_regs_l.min_chunk_size, rq_regs->rq_regs_l.meta_chunk_size, + rq_regs->rq_regs_l.min_meta_chunk_size, rq_regs->rq_regs_l.dpte_group_size, + rq_regs->rq_regs_l.mpte_group_size, rq_regs->rq_regs_l.swath_height, + rq_regs->rq_regs_l.pte_row_height_linear, rq_regs->rq_regs_c.chunk_size, rq_regs->rq_regs_c.min_chunk_size, + rq_regs->rq_regs_c.meta_chunk_size, rq_regs->rq_regs_c.min_meta_chunk_size, + rq_regs->rq_regs_c.dpte_group_size, rq_regs->rq_regs_c.mpte_group_size, + rq_regs->rq_regs_c.swath_height, rq_regs->rq_regs_c.pte_row_height_linear); } DTN_INFO("========DLG========\n"); @@ -179,27 +182,28 @@ static void dcn10_log_hubp_states(struct dc *dc) struct dcn_hubp_state *s = &(TO_DCN10_HUBP(pool->hubps[i])->state); struct _vcs_dpi_display_dlg_regs_st *dlg_regs = &s->dlg_attr; - DTN_INFO("[%2d]: %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh" - "% 8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh" - " %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh\n", - i, dlg_regs->refcyc_h_blank_end, dlg_regs->dlg_vblank_end, dlg_regs->min_dst_y_next_start, - dlg_regs->refcyc_per_htotal, dlg_regs->refcyc_x_after_scaler, dlg_regs->dst_y_after_scaler, - dlg_regs->dst_y_prefetch, dlg_regs->dst_y_per_vm_vblank, dlg_regs->dst_y_per_row_vblank, - dlg_regs->dst_y_per_vm_flip, dlg_regs->dst_y_per_row_flip, dlg_regs->ref_freq_to_pix_freq, - dlg_regs->vratio_prefetch, dlg_regs->vratio_prefetch_c, dlg_regs->refcyc_per_pte_group_vblank_l, - dlg_regs->refcyc_per_pte_group_vblank_c, dlg_regs->refcyc_per_meta_chunk_vblank_l, - dlg_regs->refcyc_per_meta_chunk_vblank_c, dlg_regs->refcyc_per_pte_group_flip_l, - dlg_regs->refcyc_per_pte_group_flip_c, dlg_regs->refcyc_per_meta_chunk_flip_l, - dlg_regs->refcyc_per_meta_chunk_flip_c, dlg_regs->dst_y_per_pte_row_nom_l, - dlg_regs->dst_y_per_pte_row_nom_c, dlg_regs->refcyc_per_pte_group_nom_l, - dlg_regs->refcyc_per_pte_group_nom_c, dlg_regs->dst_y_per_meta_row_nom_l, - dlg_regs->dst_y_per_meta_row_nom_c, dlg_regs->refcyc_per_meta_chunk_nom_l, - dlg_regs->refcyc_per_meta_chunk_nom_c, dlg_regs->refcyc_per_line_delivery_pre_l, - dlg_regs->refcyc_per_line_delivery_pre_c, dlg_regs->refcyc_per_line_delivery_l, - dlg_regs->refcyc_per_line_delivery_c, dlg_regs->chunk_hdl_adjust_cur0, dlg_regs->dst_y_offset_cur1, - dlg_regs->chunk_hdl_adjust_cur1, dlg_regs->vready_after_vcount0, dlg_regs->dst_y_delta_drq_limit, - dlg_regs->xfc_reg_transfer_delay, dlg_regs->xfc_reg_precharge_delay, - dlg_regs->xfc_reg_remote_surface_flip_latency); + if (!s->blank_en) + DTN_INFO("[%2d]: %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh" + "% 8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh" + " %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh\n", + pool->hubps[i]->inst, dlg_regs->refcyc_h_blank_end, dlg_regs->dlg_vblank_end, dlg_regs->min_dst_y_next_start, + dlg_regs->refcyc_per_htotal, dlg_regs->refcyc_x_after_scaler, dlg_regs->dst_y_after_scaler, + dlg_regs->dst_y_prefetch, dlg_regs->dst_y_per_vm_vblank, dlg_regs->dst_y_per_row_vblank, + dlg_regs->dst_y_per_vm_flip, dlg_regs->dst_y_per_row_flip, dlg_regs->ref_freq_to_pix_freq, + dlg_regs->vratio_prefetch, dlg_regs->vratio_prefetch_c, dlg_regs->refcyc_per_pte_group_vblank_l, + dlg_regs->refcyc_per_pte_group_vblank_c, dlg_regs->refcyc_per_meta_chunk_vblank_l, + dlg_regs->refcyc_per_meta_chunk_vblank_c, dlg_regs->refcyc_per_pte_group_flip_l, + dlg_regs->refcyc_per_pte_group_flip_c, dlg_regs->refcyc_per_meta_chunk_flip_l, + dlg_regs->refcyc_per_meta_chunk_flip_c, dlg_regs->dst_y_per_pte_row_nom_l, + dlg_regs->dst_y_per_pte_row_nom_c, dlg_regs->refcyc_per_pte_group_nom_l, + dlg_regs->refcyc_per_pte_group_nom_c, dlg_regs->dst_y_per_meta_row_nom_l, + dlg_regs->dst_y_per_meta_row_nom_c, dlg_regs->refcyc_per_meta_chunk_nom_l, + dlg_regs->refcyc_per_meta_chunk_nom_c, dlg_regs->refcyc_per_line_delivery_pre_l, + dlg_regs->refcyc_per_line_delivery_pre_c, dlg_regs->refcyc_per_line_delivery_l, + dlg_regs->refcyc_per_line_delivery_c, dlg_regs->chunk_hdl_adjust_cur0, dlg_regs->dst_y_offset_cur1, + dlg_regs->chunk_hdl_adjust_cur1, dlg_regs->vready_after_vcount0, dlg_regs->dst_y_delta_drq_limit, + dlg_regs->xfc_reg_transfer_delay, dlg_regs->xfc_reg_precharge_delay, + dlg_regs->xfc_reg_remote_surface_flip_latency); } DTN_INFO("========TTU========\n"); @@ -210,14 +214,15 @@ static void dcn10_log_hubp_states(struct dc *dc) struct dcn_hubp_state *s = &(TO_DCN10_HUBP(pool->hubps[i])->state); struct _vcs_dpi_display_ttu_regs_st *ttu_regs = &s->ttu_attr; - DTN_INFO("[%2d]: %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh\n", - i, ttu_regs->qos_level_low_wm, ttu_regs->qos_level_high_wm, ttu_regs->min_ttu_vblank, - ttu_regs->qos_level_flip, ttu_regs->refcyc_per_req_delivery_pre_l, ttu_regs->refcyc_per_req_delivery_l, - ttu_regs->refcyc_per_req_delivery_pre_c, ttu_regs->refcyc_per_req_delivery_c, ttu_regs->refcyc_per_req_delivery_cur0, - ttu_regs->refcyc_per_req_delivery_pre_cur0, ttu_regs->refcyc_per_req_delivery_cur1, - ttu_regs->refcyc_per_req_delivery_pre_cur1, ttu_regs->qos_level_fixed_l, ttu_regs->qos_ramp_disable_l, - ttu_regs->qos_level_fixed_c, ttu_regs->qos_ramp_disable_c, ttu_regs->qos_level_fixed_cur0, - ttu_regs->qos_ramp_disable_cur0, ttu_regs->qos_level_fixed_cur1, ttu_regs->qos_ramp_disable_cur1); + if (!s->blank_en) + DTN_INFO("[%2d]: %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh %8xh\n", + pool->hubps[i]->inst, ttu_regs->qos_level_low_wm, ttu_regs->qos_level_high_wm, ttu_regs->min_ttu_vblank, + ttu_regs->qos_level_flip, ttu_regs->refcyc_per_req_delivery_pre_l, ttu_regs->refcyc_per_req_delivery_l, + ttu_regs->refcyc_per_req_delivery_pre_c, ttu_regs->refcyc_per_req_delivery_c, ttu_regs->refcyc_per_req_delivery_cur0, + ttu_regs->refcyc_per_req_delivery_pre_cur0, ttu_regs->refcyc_per_req_delivery_cur1, + ttu_regs->refcyc_per_req_delivery_pre_cur1, ttu_regs->qos_level_fixed_l, ttu_regs->qos_ramp_disable_l, + ttu_regs->qos_level_fixed_c, ttu_regs->qos_ramp_disable_c, ttu_regs->qos_level_fixed_cur0, + ttu_regs->qos_ramp_disable_cur0, ttu_regs->qos_level_fixed_cur1, ttu_regs->qos_ramp_disable_cur1); } DTN_INFO("\n"); } @@ -321,6 +326,12 @@ void dcn10_log_hw_state(struct dc *dc) s.h_total, s.v_total, s.underflow_occurred_status); + + // Clear underflow for debug purposes + // We want to keep underflow sticky bit on for the longevity tests outside of test environment. + // This function is called only from Windows or Diags test environment, hence it's safe to clear + // it from here without affecting the original intent. + tg->funcs->clear_optc_underflow(tg); } DTN_INFO("\n"); @@ -747,6 +758,90 @@ static void reset_back_end_for_pipe( pipe_ctx->pipe_idx, pipe_ctx->stream_res.tg->inst); } +static bool dcn10_hw_wa_force_recovery(struct dc *dc) +{ + struct hubp *hubp ; + unsigned int i; + bool need_recover = true; + + if (!dc->debug.recovery_enabled) + return false; + + for (i = 0; i < dc->res_pool->pipe_count; i++) { + struct pipe_ctx *pipe_ctx = + &dc->current_state->res_ctx.pipe_ctx[i]; + if (pipe_ctx != NULL) { + hubp = pipe_ctx->plane_res.hubp; + if (hubp != NULL) { + if (hubp->funcs->hubp_get_underflow_status(hubp) != 0) { + /* one pipe underflow, we will reset all the pipes*/ + need_recover = true; + } + } + } + } + if (!need_recover) + return false; + /* + DCHUBP_CNTL:HUBP_BLANK_EN=1 + DCHUBBUB_SOFT_RESET:DCHUBBUB_GLOBAL_SOFT_RESET=1 + DCHUBP_CNTL:HUBP_DISABLE=1 + DCHUBP_CNTL:HUBP_DISABLE=0 + DCHUBBUB_SOFT_RESET:DCHUBBUB_GLOBAL_SOFT_RESET=0 + DCSURF_PRIMARY_SURFACE_ADDRESS + DCHUBP_CNTL:HUBP_BLANK_EN=0 + */ + + for (i = 0; i < dc->res_pool->pipe_count; i++) { + struct pipe_ctx *pipe_ctx = + &dc->current_state->res_ctx.pipe_ctx[i]; + if (pipe_ctx != NULL) { + hubp = pipe_ctx->plane_res.hubp; + /*DCHUBP_CNTL:HUBP_BLANK_EN=1*/ + if (hubp != NULL) + hubp->funcs->set_hubp_blank_en(hubp, true); + } + } + /*DCHUBBUB_SOFT_RESET:DCHUBBUB_GLOBAL_SOFT_RESET=1*/ + hubbub1_soft_reset(dc->res_pool->hubbub, true); + + for (i = 0; i < dc->res_pool->pipe_count; i++) { + struct pipe_ctx *pipe_ctx = + &dc->current_state->res_ctx.pipe_ctx[i]; + if (pipe_ctx != NULL) { + hubp = pipe_ctx->plane_res.hubp; + /*DCHUBP_CNTL:HUBP_DISABLE=1*/ + if (hubp != NULL) + hubp->funcs->hubp_disable_control(hubp, true); + } + } + for (i = 0; i < dc->res_pool->pipe_count; i++) { + struct pipe_ctx *pipe_ctx = + &dc->current_state->res_ctx.pipe_ctx[i]; + if (pipe_ctx != NULL) { + hubp = pipe_ctx->plane_res.hubp; + /*DCHUBP_CNTL:HUBP_DISABLE=0*/ + if (hubp != NULL) + hubp->funcs->hubp_disable_control(hubp, true); + } + } + /*DCHUBBUB_SOFT_RESET:DCHUBBUB_GLOBAL_SOFT_RESET=0*/ + hubbub1_soft_reset(dc->res_pool->hubbub, false); + for (i = 0; i < dc->res_pool->pipe_count; i++) { + struct pipe_ctx *pipe_ctx = + &dc->current_state->res_ctx.pipe_ctx[i]; + if (pipe_ctx != NULL) { + hubp = pipe_ctx->plane_res.hubp; + /*DCHUBP_CNTL:HUBP_BLANK_EN=0*/ + if (hubp != NULL) + hubp->funcs->set_hubp_blank_en(hubp, true); + } + } + return true; + +} + + static void dcn10_verify_allow_pstate_change_high(struct dc *dc) { static bool should_log_hw_state; /* prevent hw state log by default */ @@ -755,13 +850,17 @@ static void dcn10_verify_allow_pstate_change_high(struct dc *dc) if (should_log_hw_state) { dcn10_log_hw_state(dc); } - BREAK_TO_DEBUGGER(); + if (dcn10_hw_wa_force_recovery(dc)) { + /*check again*/ + if (!hubbub1_verify_allow_pstate_change_high(dc->res_pool->hubbub)) + BREAK_TO_DEBUGGER(); + } } } /* trigger HW to start disconnect plane from stream on the next vsync */ -static void plane_atomic_disconnect(struct dc *dc, struct pipe_ctx *pipe_ctx) +void hwss1_plane_atomic_disconnect(struct dc *dc, struct pipe_ctx *pipe_ctx) { struct hubp *hubp = pipe_ctx->plane_res.hubp; int dpp_id = pipe_ctx->plane_res.dpp->inst; @@ -944,7 +1043,7 @@ static void dcn10_init_hw(struct dc *dc) dc->res_pool->opps[i]->mpcc_disconnect_pending[pipe_ctx->plane_res.mpcc_inst] = true; pipe_ctx->stream_res.opp = dc->res_pool->opps[i]; - plane_atomic_disconnect(dc, pipe_ctx); + hwss1_plane_atomic_disconnect(dc, pipe_ctx); } for (i = 0; i < dc->res_pool->pipe_count; i++) { @@ -1685,22 +1784,22 @@ static uint16_t fixed_point_to_int_frac( uint16_t result; - uint16_t d = (uint16_t)dal_fixed31_32_floor( - dal_fixed31_32_abs( + uint16_t d = (uint16_t)dc_fixpt_floor( + dc_fixpt_abs( arg)); if (d <= (uint16_t)(1 << integer_bits) - (1 / (uint16_t)divisor)) - numerator = (uint16_t)dal_fixed31_32_floor( - dal_fixed31_32_mul_int( + numerator = (uint16_t)dc_fixpt_floor( + dc_fixpt_mul_int( arg, divisor)); else { - numerator = dal_fixed31_32_floor( - dal_fixed31_32_sub( - dal_fixed31_32_from_int( + numerator = dc_fixpt_floor( + dc_fixpt_sub( + dc_fixpt_from_int( 1LL << integer_bits), - dal_fixed31_32_recip( - dal_fixed31_32_from_int( + dc_fixpt_recip( + dc_fixpt_from_int( divisor)))); } @@ -1710,8 +1809,8 @@ static uint16_t fixed_point_to_int_frac( result = (uint16_t)( (1 << (integer_bits + fractional_bits + 1)) + numerator); - if ((result != 0) && dal_fixed31_32_lt( - arg, dal_fixed31_32_zero)) + if ((result != 0) && dc_fixpt_lt( + arg, dc_fixpt_zero)) result |= 1 << (integer_bits + fractional_bits); return result; @@ -1725,8 +1824,8 @@ void build_prescale_params(struct dc_bias_and_scale *bias_and_scale, && plane_state->input_csc_color_matrix.enable_adjustment && plane_state->coeff_reduction_factor.value != 0) { bias_and_scale->scale_blue = fixed_point_to_int_frac( - dal_fixed31_32_mul(plane_state->coeff_reduction_factor, - dal_fixed31_32_from_fraction(256, 255)), + dc_fixpt_mul(plane_state->coeff_reduction_factor, + dc_fixpt_from_fraction(256, 255)), 2, 13); bias_and_scale->scale_red = bias_and_scale->scale_blue; @@ -1995,7 +2094,7 @@ static void dcn10_blank_pixel_data( static void set_hdr_multiplier(struct pipe_ctx *pipe_ctx) { - struct fixed31_32 multiplier = dal_fixed31_32_from_fraction( + struct fixed31_32 multiplier = dc_fixpt_from_fraction( pipe_ctx->plane_state->sdr_white_level, 80); uint32_t hw_mult = 0x1f000; // 1.0 default multiplier struct custom_float_format fmt; @@ -2179,7 +2278,7 @@ static void dcn10_apply_ctx_for_surface( old_pipe_ctx->plane_state && old_pipe_ctx->stream_res.tg == tg) { - plane_atomic_disconnect(dc, old_pipe_ctx); + hwss1_plane_atomic_disconnect(dc, old_pipe_ctx); removed_pipe[i] = true; DC_LOG_DC( @@ -2487,15 +2586,6 @@ static void set_static_screen_control(struct pipe_ctx **pipe_ctx, set_static_screen_control(pipe_ctx[i]->stream_res.tg, value); } -static void set_plane_config( - const struct dc *dc, - struct pipe_ctx *pipe_ctx, - struct resource_context *res_ctx) -{ - /* TODO */ - program_gamut_remap(pipe_ctx); -} - static void dcn10_config_stereo_parameters( struct dc_stream_state *stream, struct crtc_stereo_flags *flags) { @@ -2673,7 +2763,6 @@ static const struct hw_sequencer_funcs dcn10_funcs = { .init_hw = dcn10_init_hw, .apply_ctx_to_hw = dce110_apply_ctx_to_hw, .apply_ctx_for_surface = dcn10_apply_ctx_for_surface, - .set_plane_config = set_plane_config, .update_plane_addr = dcn10_update_plane_addr, .update_dchub = dcn10_update_dchub, .update_pending_status = dcn10_update_pending_status, diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.h index 6c526b5095d9..44f734b73f9e 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.h +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.h @@ -37,4 +37,6 @@ extern void fill_display_configs( bool is_rgb_cspace(enum dc_color_space output_color_space); +void hwss1_plane_atomic_disconnect(struct dc *dc, struct pipe_ctx *pipe_ctx); + #endif /* __DC_HWSS_DCN10_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.c new file mode 100644 index 000000000000..21fa40ac0786 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.c @@ -0,0 +1,1362 @@ +/* + * Copyright 2012-15 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#include "reg_helper.h" + +#include "core_types.h" +#include "link_encoder.h" +#include "dcn10_link_encoder.h" +#include "stream_encoder.h" +#include "i2caux_interface.h" +#include "dc_bios_types.h" + +#include "gpio_service_interface.h" + +#define CTX \ + enc10->base.ctx +#define DC_LOGGER \ + enc10->base.ctx->logger + +#define REG(reg)\ + (enc10->link_regs->reg) + +#undef FN +#define FN(reg_name, field_name) \ + enc10->link_shift->field_name, enc10->link_mask->field_name + + +/* + * @brief + * Trigger Source Select + * ASIC-dependent, actual values for register programming + */ +#define DCN10_DIG_FE_SOURCE_SELECT_INVALID 0x0 +#define DCN10_DIG_FE_SOURCE_SELECT_DIGA 0x1 +#define DCN10_DIG_FE_SOURCE_SELECT_DIGB 0x2 +#define DCN10_DIG_FE_SOURCE_SELECT_DIGC 0x4 +#define DCN10_DIG_FE_SOURCE_SELECT_DIGD 0x08 +#define DCN10_DIG_FE_SOURCE_SELECT_DIGE 0x10 +#define DCN10_DIG_FE_SOURCE_SELECT_DIGF 0x20 +#define DCN10_DIG_FE_SOURCE_SELECT_DIGG 0x40 + +enum { + DP_MST_UPDATE_MAX_RETRY = 50 +}; + + + +static void aux_initialize(struct dcn10_link_encoder *enc10); + + +static const struct link_encoder_funcs dcn10_lnk_enc_funcs = { + .validate_output_with_stream = + dcn10_link_encoder_validate_output_with_stream, + .hw_init = dcn10_link_encoder_hw_init, + .setup = dcn10_link_encoder_setup, + .enable_tmds_output = dcn10_link_encoder_enable_tmds_output, + .enable_dp_output = dcn10_link_encoder_enable_dp_output, + .enable_dp_mst_output = dcn10_link_encoder_enable_dp_mst_output, + .disable_output = dcn10_link_encoder_disable_output, + .dp_set_lane_settings = dcn10_link_encoder_dp_set_lane_settings, + .dp_set_phy_pattern = dcn10_link_encoder_dp_set_phy_pattern, + .update_mst_stream_allocation_table = + dcn10_link_encoder_update_mst_stream_allocation_table, + .psr_program_dp_dphy_fast_training = + dcn10_psr_program_dp_dphy_fast_training, + .psr_program_secondary_packet = dcn10_psr_program_secondary_packet, + .connect_dig_be_to_fe = dcn10_link_encoder_connect_dig_be_to_fe, + .enable_hpd = dcn10_link_encoder_enable_hpd, + .disable_hpd = dcn10_link_encoder_disable_hpd, + .is_dig_enabled = dcn10_is_dig_enabled, + .destroy = dcn10_link_encoder_destroy +}; + +static enum bp_result link_transmitter_control( + struct dcn10_link_encoder *enc10, + struct bp_transmitter_control *cntl) +{ + enum bp_result result; + struct dc_bios *bp = enc10->base.ctx->dc_bios; + + result = bp->funcs->transmitter_control(bp, cntl); + + return result; +} + +static void enable_phy_bypass_mode( + struct dcn10_link_encoder *enc10, + bool enable) +{ + /* This register resides in DP back end block; + * transmitter is used for the offset + */ + REG_UPDATE(DP_DPHY_CNTL, DPHY_BYPASS, enable); + +} + +static void disable_prbs_symbols( + struct dcn10_link_encoder *enc10, + bool disable) +{ + /* This register resides in DP back end block; + * transmitter is used for the offset + */ + REG_UPDATE_4(DP_DPHY_CNTL, + DPHY_ATEST_SEL_LANE0, disable, + DPHY_ATEST_SEL_LANE1, disable, + DPHY_ATEST_SEL_LANE2, disable, + DPHY_ATEST_SEL_LANE3, disable); +} + +static void disable_prbs_mode( + struct dcn10_link_encoder *enc10) +{ + REG_UPDATE(DP_DPHY_PRBS_CNTL, DPHY_PRBS_EN, 0); +} + +static void program_pattern_symbols( + struct dcn10_link_encoder *enc10, + uint16_t pattern_symbols[8]) +{ + /* This register resides in DP back end block; + * transmitter is used for the offset + */ + REG_SET_3(DP_DPHY_SYM0, 0, + DPHY_SYM1, pattern_symbols[0], + DPHY_SYM2, pattern_symbols[1], + DPHY_SYM3, pattern_symbols[2]); + + /* This register resides in DP back end block; + * transmitter is used for the offset + */ + REG_SET_3(DP_DPHY_SYM1, 0, + DPHY_SYM4, pattern_symbols[3], + DPHY_SYM5, pattern_symbols[4], + DPHY_SYM6, pattern_symbols[5]); + + /* This register resides in DP back end block; + * transmitter is used for the offset + */ + REG_SET_2(DP_DPHY_SYM2, 0, + DPHY_SYM7, pattern_symbols[6], + DPHY_SYM8, pattern_symbols[7]); +} + +static void set_dp_phy_pattern_d102( + struct dcn10_link_encoder *enc10) +{ + /* Disable PHY Bypass mode to setup the test pattern */ + enable_phy_bypass_mode(enc10, false); + + /* For 10-bit PRBS or debug symbols + * please use the following sequence: + * + * Enable debug symbols on the lanes + */ + disable_prbs_symbols(enc10, true); + + /* Disable PRBS mode */ + disable_prbs_mode(enc10); + + /* Program debug symbols to be output */ + { + uint16_t pattern_symbols[8] = { + 0x2AA, 0x2AA, 0x2AA, 0x2AA, + 0x2AA, 0x2AA, 0x2AA, 0x2AA + }; + + program_pattern_symbols(enc10, pattern_symbols); + } + + /* Enable phy bypass mode to enable the test pattern */ + + enable_phy_bypass_mode(enc10, true); +} + +static void set_link_training_complete( + struct dcn10_link_encoder *enc10, + bool complete) +{ + /* This register resides in DP back end block; + * transmitter is used for the offset + */ + REG_UPDATE(DP_LINK_CNTL, DP_LINK_TRAINING_COMPLETE, complete); + +} + +void dcn10_link_encoder_set_dp_phy_pattern_training_pattern( + struct link_encoder *enc, + uint32_t index) +{ + struct dcn10_link_encoder *enc10 = TO_DCN10_LINK_ENC(enc); + /* Write Training Pattern */ + + REG_WRITE(DP_DPHY_TRAINING_PATTERN_SEL, index); + + /* Set HW Register Training Complete to false */ + + set_link_training_complete(enc10, false); + + /* Disable PHY Bypass mode to output Training Pattern */ + + enable_phy_bypass_mode(enc10, false); + + /* Disable PRBS mode */ + disable_prbs_mode(enc10); +} + +static void setup_panel_mode( + struct dcn10_link_encoder *enc10, + enum dp_panel_mode panel_mode) +{ + uint32_t value; + + ASSERT(REG(DP_DPHY_INTERNAL_CTRL)); + value = REG_READ(DP_DPHY_INTERNAL_CTRL); + + switch (panel_mode) { + case DP_PANEL_MODE_EDP: + value = 0x1; + break; + case DP_PANEL_MODE_SPECIAL: + value = 0x11; + break; + default: + value = 0x0; + break; + } + + REG_WRITE(DP_DPHY_INTERNAL_CTRL, value); +} + +static void set_dp_phy_pattern_symbol_error( + struct dcn10_link_encoder *enc10) +{ + /* Disable PHY Bypass mode to setup the test pattern */ + enable_phy_bypass_mode(enc10, false); + + /* program correct panel mode*/ + setup_panel_mode(enc10, DP_PANEL_MODE_DEFAULT); + + /* A PRBS23 pattern is used for most DP electrical measurements. */ + + /* Enable PRBS symbols on the lanes */ + disable_prbs_symbols(enc10, false); + + /* For PRBS23 Set bit DPHY_PRBS_SEL=1 and Set bit DPHY_PRBS_EN=1 */ + REG_UPDATE_2(DP_DPHY_PRBS_CNTL, + DPHY_PRBS_SEL, 1, + DPHY_PRBS_EN, 1); + + /* Enable phy bypass mode to enable the test pattern */ + enable_phy_bypass_mode(enc10, true); +} + +static void set_dp_phy_pattern_prbs7( + struct dcn10_link_encoder *enc10) +{ + /* Disable PHY Bypass mode to setup the test pattern */ + enable_phy_bypass_mode(enc10, false); + + /* A PRBS7 pattern is used for most DP electrical measurements. */ + + /* Enable PRBS symbols on the lanes */ + disable_prbs_symbols(enc10, false); + + /* For PRBS7 Set bit DPHY_PRBS_SEL=0 and Set bit DPHY_PRBS_EN=1 */ + REG_UPDATE_2(DP_DPHY_PRBS_CNTL, + DPHY_PRBS_SEL, 0, + DPHY_PRBS_EN, 1); + + /* Enable phy bypass mode to enable the test pattern */ + enable_phy_bypass_mode(enc10, true); +} + +static void set_dp_phy_pattern_80bit_custom( + struct dcn10_link_encoder *enc10, + const uint8_t *pattern) +{ + /* Disable PHY Bypass mode to setup the test pattern */ + enable_phy_bypass_mode(enc10, false); + + /* Enable debug symbols on the lanes */ + + disable_prbs_symbols(enc10, true); + + /* Enable PHY bypass mode to enable the test pattern */ + /* TODO is it really needed ? */ + + enable_phy_bypass_mode(enc10, true); + + /* Program 80 bit custom pattern */ + { + uint16_t pattern_symbols[8]; + + pattern_symbols[0] = + ((pattern[1] & 0x03) << 8) | pattern[0]; + pattern_symbols[1] = + ((pattern[2] & 0x0f) << 6) | ((pattern[1] >> 2) & 0x3f); + pattern_symbols[2] = + ((pattern[3] & 0x3f) << 4) | ((pattern[2] >> 4) & 0x0f); + pattern_symbols[3] = + (pattern[4] << 2) | ((pattern[3] >> 6) & 0x03); + pattern_symbols[4] = + ((pattern[6] & 0x03) << 8) | pattern[5]; + pattern_symbols[5] = + ((pattern[7] & 0x0f) << 6) | ((pattern[6] >> 2) & 0x3f); + pattern_symbols[6] = + ((pattern[8] & 0x3f) << 4) | ((pattern[7] >> 4) & 0x0f); + pattern_symbols[7] = + (pattern[9] << 2) | ((pattern[8] >> 6) & 0x03); + + program_pattern_symbols(enc10, pattern_symbols); + } + + /* Enable phy bypass mode to enable the test pattern */ + + enable_phy_bypass_mode(enc10, true); +} + +static void set_dp_phy_pattern_hbr2_compliance_cp2520_2( + struct dcn10_link_encoder *enc10, + unsigned int cp2520_pattern) +{ + + /* previously there is a register DP_HBR2_EYE_PATTERN + * that is enabled to get the pattern. + * But it does not work with the latest spec change, + * so we are programming the following registers manually. + * + * The following settings have been confirmed + * by Nick Chorney and Sandra Liu + */ + + /* Disable PHY Bypass mode to setup the test pattern */ + + enable_phy_bypass_mode(enc10, false); + + /* Setup DIG encoder in DP SST mode */ + enc10->base.funcs->setup(&enc10->base, SIGNAL_TYPE_DISPLAY_PORT); + + /* ensure normal panel mode. */ + setup_panel_mode(enc10, DP_PANEL_MODE_DEFAULT); + + /* no vbid after BS (SR) + * DP_LINK_FRAMING_CNTL changed history Sandra Liu + * 11000260 / 11000104 / 110000FC + */ + REG_UPDATE_3(DP_LINK_FRAMING_CNTL, + DP_IDLE_BS_INTERVAL, 0xFC, + DP_VBID_DISABLE, 1, + DP_VID_ENHANCED_FRAME_MODE, 1); + + /* swap every BS with SR */ + REG_UPDATE(DP_DPHY_SCRAM_CNTL, DPHY_SCRAMBLER_BS_COUNT, 0); + + /* select cp2520 patterns */ + if (REG(DP_DPHY_HBR2_PATTERN_CONTROL)) + REG_UPDATE(DP_DPHY_HBR2_PATTERN_CONTROL, + DP_DPHY_HBR2_PATTERN_CONTROL, cp2520_pattern); + else + /* pre-DCE11 can only generate CP2520 pattern 2 */ + ASSERT(cp2520_pattern == 2); + + /* set link training complete */ + set_link_training_complete(enc10, true); + + /* disable video stream */ + REG_UPDATE(DP_VID_STREAM_CNTL, DP_VID_STREAM_ENABLE, 0); + + /* Disable PHY Bypass mode to setup the test pattern */ + enable_phy_bypass_mode(enc10, false); +} + +static void set_dp_phy_pattern_passthrough_mode( + struct dcn10_link_encoder *enc10, + enum dp_panel_mode panel_mode) +{ + /* program correct panel mode */ + setup_panel_mode(enc10, panel_mode); + + /* restore LINK_FRAMING_CNTL and DPHY_SCRAMBLER_BS_COUNT + * in case we were doing HBR2 compliance pattern before + */ + REG_UPDATE_3(DP_LINK_FRAMING_CNTL, + DP_IDLE_BS_INTERVAL, 0x2000, + DP_VBID_DISABLE, 0, + DP_VID_ENHANCED_FRAME_MODE, 1); + + REG_UPDATE(DP_DPHY_SCRAM_CNTL, DPHY_SCRAMBLER_BS_COUNT, 0x1FF); + + /* set link training complete */ + set_link_training_complete(enc10, true); + + /* Disable PHY Bypass mode to setup the test pattern */ + enable_phy_bypass_mode(enc10, false); + + /* Disable PRBS mode */ + disable_prbs_mode(enc10); +} + +/* return value is bit-vector */ +static uint8_t get_frontend_source( + enum engine_id engine) +{ + switch (engine) { + case ENGINE_ID_DIGA: + return DCN10_DIG_FE_SOURCE_SELECT_DIGA; + case ENGINE_ID_DIGB: + return DCN10_DIG_FE_SOURCE_SELECT_DIGB; + case ENGINE_ID_DIGC: + return DCN10_DIG_FE_SOURCE_SELECT_DIGC; + case ENGINE_ID_DIGD: + return DCN10_DIG_FE_SOURCE_SELECT_DIGD; + case ENGINE_ID_DIGE: + return DCN10_DIG_FE_SOURCE_SELECT_DIGE; + case ENGINE_ID_DIGF: + return DCN10_DIG_FE_SOURCE_SELECT_DIGF; + case ENGINE_ID_DIGG: + return DCN10_DIG_FE_SOURCE_SELECT_DIGG; + default: + ASSERT_CRITICAL(false); + return DCN10_DIG_FE_SOURCE_SELECT_INVALID; + } +} + +static void configure_encoder( + struct dcn10_link_encoder *enc10, + const struct dc_link_settings *link_settings) +{ + /* set number of lanes */ + + REG_SET(DP_CONFIG, 0, + DP_UDI_LANES, link_settings->lane_count - LANE_COUNT_ONE); + + /* setup scrambler */ + REG_UPDATE(DP_DPHY_SCRAM_CNTL, DPHY_SCRAMBLER_ADVANCE, 1); +} + +void dcn10_psr_program_dp_dphy_fast_training(struct link_encoder *enc, + bool exit_link_training_required) +{ + struct dcn10_link_encoder *enc10 = TO_DCN10_LINK_ENC(enc); + + if (exit_link_training_required) + REG_UPDATE(DP_DPHY_FAST_TRAINING, + DPHY_RX_FAST_TRAINING_CAPABLE, 1); + else { + REG_UPDATE(DP_DPHY_FAST_TRAINING, + DPHY_RX_FAST_TRAINING_CAPABLE, 0); + /*In DCE 11, we are able to pre-program a Force SR register + * to be able to trigger SR symbol after 5 idle patterns + * transmitted. Upon PSR Exit, DMCU can trigger + * DPHY_LOAD_BS_COUNT_START = 1. Upon writing 1 to + * DPHY_LOAD_BS_COUNT_START and the internal counter + * reaches DPHY_LOAD_BS_COUNT, the next BS symbol will be + * replaced by SR symbol once. + */ + + REG_UPDATE(DP_DPHY_BS_SR_SWAP_CNTL, DPHY_LOAD_BS_COUNT, 0x5); + } +} + +void dcn10_psr_program_secondary_packet(struct link_encoder *enc, + unsigned int sdp_transmit_line_num_deadline) +{ + struct dcn10_link_encoder *enc10 = TO_DCN10_LINK_ENC(enc); + + REG_UPDATE_2(DP_SEC_CNTL1, + DP_SEC_GSP0_LINE_NUM, sdp_transmit_line_num_deadline, + DP_SEC_GSP0_PRIORITY, 1); +} + +bool dcn10_is_dig_enabled(struct link_encoder *enc) +{ + struct dcn10_link_encoder *enc10 = TO_DCN10_LINK_ENC(enc); + uint32_t value; + + REG_GET(DIG_BE_EN_CNTL, DIG_ENABLE, &value); + return value; +} + +static void link_encoder_disable(struct dcn10_link_encoder *enc10) +{ + /* reset training pattern */ + REG_SET(DP_DPHY_TRAINING_PATTERN_SEL, 0, + DPHY_TRAINING_PATTERN_SEL, 0); + + /* reset training complete */ + REG_UPDATE(DP_LINK_CNTL, DP_LINK_TRAINING_COMPLETE, 0); + + /* reset panel mode */ + setup_panel_mode(enc10, DP_PANEL_MODE_DEFAULT); +} + +static void hpd_initialize( + struct dcn10_link_encoder *enc10) +{ + /* Associate HPD with DIG_BE */ + enum hpd_source_id hpd_source = enc10->base.hpd_source; + + REG_UPDATE(DIG_BE_CNTL, DIG_HPD_SELECT, hpd_source); +} + +bool dcn10_link_encoder_validate_dvi_output( + const struct dcn10_link_encoder *enc10, + enum signal_type connector_signal, + enum signal_type signal, + const struct dc_crtc_timing *crtc_timing) +{ + uint32_t max_pixel_clock = TMDS_MAX_PIXEL_CLOCK; + + if (signal == SIGNAL_TYPE_DVI_DUAL_LINK) + max_pixel_clock *= 2; + + /* This handles the case of HDMI downgrade to DVI we don't want to + * we don't want to cap the pixel clock if the DDI is not DVI. + */ + if (connector_signal != SIGNAL_TYPE_DVI_DUAL_LINK && + connector_signal != SIGNAL_TYPE_DVI_SINGLE_LINK) + max_pixel_clock = enc10->base.features.max_hdmi_pixel_clock; + + /* DVI only support RGB pixel encoding */ + if (crtc_timing->pixel_encoding != PIXEL_ENCODING_RGB) + return false; + + /*connect DVI via adpater's HDMI connector*/ + if ((connector_signal == SIGNAL_TYPE_DVI_SINGLE_LINK || + connector_signal == SIGNAL_TYPE_HDMI_TYPE_A) && + signal != SIGNAL_TYPE_HDMI_TYPE_A && + crtc_timing->pix_clk_khz > TMDS_MAX_PIXEL_CLOCK) + return false; + if (crtc_timing->pix_clk_khz < TMDS_MIN_PIXEL_CLOCK) + return false; + + if (crtc_timing->pix_clk_khz > max_pixel_clock) + return false; + + /* DVI supports 6/8bpp single-link and 10/16bpp dual-link */ + switch (crtc_timing->display_color_depth) { + case COLOR_DEPTH_666: + case COLOR_DEPTH_888: + break; + case COLOR_DEPTH_101010: + case COLOR_DEPTH_161616: + if (signal != SIGNAL_TYPE_DVI_DUAL_LINK) + return false; + break; + default: + return false; + } + + return true; +} + +static bool dcn10_link_encoder_validate_hdmi_output( + const struct dcn10_link_encoder *enc10, + const struct dc_crtc_timing *crtc_timing, + int adjusted_pix_clk_khz) +{ + enum dc_color_depth max_deep_color = + enc10->base.features.max_hdmi_deep_color; + + if (max_deep_color < crtc_timing->display_color_depth) + return false; + + if (crtc_timing->display_color_depth < COLOR_DEPTH_888) + return false; + if (adjusted_pix_clk_khz < TMDS_MIN_PIXEL_CLOCK) + return false; + + if ((adjusted_pix_clk_khz == 0) || + (adjusted_pix_clk_khz > enc10->base.features.max_hdmi_pixel_clock)) + return false; + + /* DCE11 HW does not support 420 */ + if (!enc10->base.features.ycbcr420_supported && + crtc_timing->pixel_encoding == PIXEL_ENCODING_YCBCR420) + return false; + + if (!enc10->base.features.flags.bits.HDMI_6GB_EN && + adjusted_pix_clk_khz >= 300000) + return false; + return true; +} + +bool dcn10_link_encoder_validate_dp_output( + const struct dcn10_link_encoder *enc10, + const struct dc_crtc_timing *crtc_timing) +{ + /* default RGB only */ + if (crtc_timing->pixel_encoding == PIXEL_ENCODING_RGB) + return true; + + if (enc10->base.features.flags.bits.IS_YCBCR_CAPABLE) + return true; + + /* for DCE 8.x or later DP Y-only feature, + * we need ASIC cap + FeatureSupportDPYonly, not support 666 + */ + if (crtc_timing->flags.Y_ONLY && + enc10->base.features.flags.bits.IS_YCBCR_CAPABLE && + crtc_timing->display_color_depth != COLOR_DEPTH_666) + return true; + + return false; +} + +void dcn10_link_encoder_construct( + struct dcn10_link_encoder *enc10, + const struct encoder_init_data *init_data, + const struct encoder_feature_support *enc_features, + const struct dcn10_link_enc_registers *link_regs, + const struct dcn10_link_enc_aux_registers *aux_regs, + const struct dcn10_link_enc_hpd_registers *hpd_regs, + const struct dcn10_link_enc_shift *link_shift, + const struct dcn10_link_enc_mask *link_mask) +{ + struct bp_encoder_cap_info bp_cap_info = {0}; + const struct dc_vbios_funcs *bp_funcs = init_data->ctx->dc_bios->funcs; + enum bp_result result = BP_RESULT_OK; + + enc10->base.funcs = &dcn10_lnk_enc_funcs; + enc10->base.ctx = init_data->ctx; + enc10->base.id = init_data->encoder; + + enc10->base.hpd_source = init_data->hpd_source; + enc10->base.connector = init_data->connector; + + enc10->base.preferred_engine = ENGINE_ID_UNKNOWN; + + enc10->base.features = *enc_features; + + enc10->base.transmitter = init_data->transmitter; + + /* set the flag to indicate whether driver poll the I2C data pin + * while doing the DP sink detect + */ + +/* if (dal_adapter_service_is_feature_supported(as, + FEATURE_DP_SINK_DETECT_POLL_DATA_PIN)) + enc10->base.features.flags.bits. + DP_SINK_DETECT_POLL_DATA_PIN = true;*/ + + enc10->base.output_signals = + SIGNAL_TYPE_DVI_SINGLE_LINK | + SIGNAL_TYPE_DVI_DUAL_LINK | + SIGNAL_TYPE_LVDS | + SIGNAL_TYPE_DISPLAY_PORT | + SIGNAL_TYPE_DISPLAY_PORT_MST | + SIGNAL_TYPE_EDP | + SIGNAL_TYPE_HDMI_TYPE_A; + + /* For DCE 8.0 and 8.1, by design, UNIPHY is hardwired to DIG_BE. + * SW always assign DIG_FE 1:1 mapped to DIG_FE for non-MST UNIPHY. + * SW assign DIG_FE to non-MST UNIPHY first and MST last. So prefer + * DIG is per UNIPHY and used by SST DP, eDP, HDMI, DVI and LVDS. + * Prefer DIG assignment is decided by board design. + * For DCE 8.0, there are only max 6 UNIPHYs, we assume board design + * and VBIOS will filter out 7 UNIPHY for DCE 8.0. + * By this, adding DIGG should not hurt DCE 8.0. + * This will let DCE 8.1 share DCE 8.0 as much as possible + */ + + enc10->link_regs = link_regs; + enc10->aux_regs = aux_regs; + enc10->hpd_regs = hpd_regs; + enc10->link_shift = link_shift; + enc10->link_mask = link_mask; + + switch (enc10->base.transmitter) { + case TRANSMITTER_UNIPHY_A: + enc10->base.preferred_engine = ENGINE_ID_DIGA; + break; + case TRANSMITTER_UNIPHY_B: + enc10->base.preferred_engine = ENGINE_ID_DIGB; + break; + case TRANSMITTER_UNIPHY_C: + enc10->base.preferred_engine = ENGINE_ID_DIGC; + break; + case TRANSMITTER_UNIPHY_D: + enc10->base.preferred_engine = ENGINE_ID_DIGD; + break; + case TRANSMITTER_UNIPHY_E: + enc10->base.preferred_engine = ENGINE_ID_DIGE; + break; + case TRANSMITTER_UNIPHY_F: + enc10->base.preferred_engine = ENGINE_ID_DIGF; + break; + case TRANSMITTER_UNIPHY_G: + enc10->base.preferred_engine = ENGINE_ID_DIGG; + break; + default: + ASSERT_CRITICAL(false); + enc10->base.preferred_engine = ENGINE_ID_UNKNOWN; + } + + /* default to one to mirror Windows behavior */ + enc10->base.features.flags.bits.HDMI_6GB_EN = 1; + + result = bp_funcs->get_encoder_cap_info(enc10->base.ctx->dc_bios, + enc10->base.id, &bp_cap_info); + + /* Override features with DCE-specific values */ + if (result == BP_RESULT_OK) { + enc10->base.features.flags.bits.IS_HBR2_CAPABLE = + bp_cap_info.DP_HBR2_EN; + enc10->base.features.flags.bits.IS_HBR3_CAPABLE = + bp_cap_info.DP_HBR3_EN; + enc10->base.features.flags.bits.HDMI_6GB_EN = bp_cap_info.HDMI_6GB_EN; + } else { + DC_LOG_WARNING("%s: Failed to get encoder_cap_info from VBIOS with error code %d!\n", + __func__, + result); + } +} + +bool dcn10_link_encoder_validate_output_with_stream( + struct link_encoder *enc, + const struct dc_stream_state *stream) +{ + struct dcn10_link_encoder *enc10 = TO_DCN10_LINK_ENC(enc); + bool is_valid; + + switch (stream->signal) { + case SIGNAL_TYPE_DVI_SINGLE_LINK: + case SIGNAL_TYPE_DVI_DUAL_LINK: + is_valid = dcn10_link_encoder_validate_dvi_output( + enc10, + stream->sink->link->connector_signal, + stream->signal, + &stream->timing); + break; + case SIGNAL_TYPE_HDMI_TYPE_A: + is_valid = dcn10_link_encoder_validate_hdmi_output( + enc10, + &stream->timing, + stream->phy_pix_clk); + break; + case SIGNAL_TYPE_DISPLAY_PORT: + case SIGNAL_TYPE_DISPLAY_PORT_MST: + is_valid = dcn10_link_encoder_validate_dp_output( + enc10, &stream->timing); + break; + case SIGNAL_TYPE_EDP: + is_valid = (stream->timing.pixel_encoding == PIXEL_ENCODING_RGB) ? true : false; + break; + case SIGNAL_TYPE_VIRTUAL: + is_valid = true; + break; + default: + is_valid = false; + break; + } + + return is_valid; +} + +void dcn10_link_encoder_hw_init( + struct link_encoder *enc) +{ + struct dcn10_link_encoder *enc10 = TO_DCN10_LINK_ENC(enc); + struct bp_transmitter_control cntl = { 0 }; + enum bp_result result; + + cntl.action = TRANSMITTER_CONTROL_INIT; + cntl.engine_id = ENGINE_ID_UNKNOWN; + cntl.transmitter = enc10->base.transmitter; + cntl.connector_obj_id = enc10->base.connector; + cntl.lanes_number = LANE_COUNT_FOUR; + cntl.coherent = false; + cntl.hpd_sel = enc10->base.hpd_source; + + if (enc10->base.connector.id == CONNECTOR_ID_EDP) + cntl.signal = SIGNAL_TYPE_EDP; + + result = link_transmitter_control(enc10, &cntl); + + if (result != BP_RESULT_OK) { + DC_LOG_ERROR("%s: Failed to execute VBIOS command table!\n", + __func__); + BREAK_TO_DEBUGGER(); + return; + } + + if (enc10->base.connector.id == CONNECTOR_ID_LVDS) { + cntl.action = TRANSMITTER_CONTROL_BACKLIGHT_BRIGHTNESS; + + result = link_transmitter_control(enc10, &cntl); + + ASSERT(result == BP_RESULT_OK); + + } + aux_initialize(enc10); + + /* reinitialize HPD. + * hpd_initialize() will pass DIG_FE id to HW context. + * All other routine within HW context will use fe_engine_offset + * as DIG_FE id even caller pass DIG_FE id. + * So this routine must be called first. + */ + hpd_initialize(enc10); +} + +void dcn10_link_encoder_destroy(struct link_encoder **enc) +{ + kfree(TO_DCN10_LINK_ENC(*enc)); + *enc = NULL; +} + +void dcn10_link_encoder_setup( + struct link_encoder *enc, + enum signal_type signal) +{ + struct dcn10_link_encoder *enc10 = TO_DCN10_LINK_ENC(enc); + + switch (signal) { + case SIGNAL_TYPE_EDP: + case SIGNAL_TYPE_DISPLAY_PORT: + /* DP SST */ + REG_UPDATE(DIG_BE_CNTL, DIG_MODE, 0); + break; + case SIGNAL_TYPE_LVDS: + /* LVDS */ + REG_UPDATE(DIG_BE_CNTL, DIG_MODE, 1); + break; + case SIGNAL_TYPE_DVI_SINGLE_LINK: + case SIGNAL_TYPE_DVI_DUAL_LINK: + /* TMDS-DVI */ + REG_UPDATE(DIG_BE_CNTL, DIG_MODE, 2); + break; + case SIGNAL_TYPE_HDMI_TYPE_A: + /* TMDS-HDMI */ + REG_UPDATE(DIG_BE_CNTL, DIG_MODE, 3); + break; + case SIGNAL_TYPE_DISPLAY_PORT_MST: + /* DP MST */ + REG_UPDATE(DIG_BE_CNTL, DIG_MODE, 5); + break; + default: + ASSERT_CRITICAL(false); + /* invalid mode ! */ + break; + } + +} + +/* TODO: still need depth or just pass in adjusted pixel clock? */ +void dcn10_link_encoder_enable_tmds_output( + struct link_encoder *enc, + enum clock_source_id clock_source, + enum dc_color_depth color_depth, + enum signal_type signal, + uint32_t pixel_clock) +{ + struct dcn10_link_encoder *enc10 = TO_DCN10_LINK_ENC(enc); + struct bp_transmitter_control cntl = { 0 }; + enum bp_result result; + + /* Enable the PHY */ + + cntl.action = TRANSMITTER_CONTROL_ENABLE; + cntl.engine_id = enc->preferred_engine; + cntl.transmitter = enc10->base.transmitter; + cntl.pll_id = clock_source; + cntl.signal = signal; + if (cntl.signal == SIGNAL_TYPE_DVI_DUAL_LINK) + cntl.lanes_number = 8; + else + cntl.lanes_number = 4; + + cntl.hpd_sel = enc10->base.hpd_source; + + cntl.pixel_clock = pixel_clock; + cntl.color_depth = color_depth; + + result = link_transmitter_control(enc10, &cntl); + + if (result != BP_RESULT_OK) { + DC_LOG_ERROR("%s: Failed to execute VBIOS command table!\n", + __func__); + BREAK_TO_DEBUGGER(); + } +} + +/* enables DP PHY output */ +void dcn10_link_encoder_enable_dp_output( + struct link_encoder *enc, + const struct dc_link_settings *link_settings, + enum clock_source_id clock_source) +{ + struct dcn10_link_encoder *enc10 = TO_DCN10_LINK_ENC(enc); + struct bp_transmitter_control cntl = { 0 }; + enum bp_result result; + + /* Enable the PHY */ + + /* number_of_lanes is used for pixel clock adjust, + * but it's not passed to asic_control. + * We need to set number of lanes manually. + */ + configure_encoder(enc10, link_settings); + + cntl.action = TRANSMITTER_CONTROL_ENABLE; + cntl.engine_id = enc->preferred_engine; + cntl.transmitter = enc10->base.transmitter; + cntl.pll_id = clock_source; + cntl.signal = SIGNAL_TYPE_DISPLAY_PORT; + cntl.lanes_number = link_settings->lane_count; + cntl.hpd_sel = enc10->base.hpd_source; + cntl.pixel_clock = link_settings->link_rate + * LINK_RATE_REF_FREQ_IN_KHZ; + /* TODO: check if undefined works */ + cntl.color_depth = COLOR_DEPTH_UNDEFINED; + + result = link_transmitter_control(enc10, &cntl); + + if (result != BP_RESULT_OK) { + DC_LOG_ERROR("%s: Failed to execute VBIOS command table!\n", + __func__); + BREAK_TO_DEBUGGER(); + } +} + +/* enables DP PHY output in MST mode */ +void dcn10_link_encoder_enable_dp_mst_output( + struct link_encoder *enc, + const struct dc_link_settings *link_settings, + enum clock_source_id clock_source) +{ + struct dcn10_link_encoder *enc10 = TO_DCN10_LINK_ENC(enc); + struct bp_transmitter_control cntl = { 0 }; + enum bp_result result; + + /* Enable the PHY */ + + /* number_of_lanes is used for pixel clock adjust, + * but it's not passed to asic_control. + * We need to set number of lanes manually. + */ + configure_encoder(enc10, link_settings); + + cntl.action = TRANSMITTER_CONTROL_ENABLE; + cntl.engine_id = ENGINE_ID_UNKNOWN; + cntl.transmitter = enc10->base.transmitter; + cntl.pll_id = clock_source; + cntl.signal = SIGNAL_TYPE_DISPLAY_PORT_MST; + cntl.lanes_number = link_settings->lane_count; + cntl.hpd_sel = enc10->base.hpd_source; + cntl.pixel_clock = link_settings->link_rate + * LINK_RATE_REF_FREQ_IN_KHZ; + /* TODO: check if undefined works */ + cntl.color_depth = COLOR_DEPTH_UNDEFINED; + + result = link_transmitter_control(enc10, &cntl); + + if (result != BP_RESULT_OK) { + DC_LOG_ERROR("%s: Failed to execute VBIOS command table!\n", + __func__); + BREAK_TO_DEBUGGER(); + } +} +/* + * @brief + * Disable transmitter and its encoder + */ +void dcn10_link_encoder_disable_output( + struct link_encoder *enc, + enum signal_type signal) +{ + struct dcn10_link_encoder *enc10 = TO_DCN10_LINK_ENC(enc); + struct bp_transmitter_control cntl = { 0 }; + enum bp_result result; + + if (!dcn10_is_dig_enabled(enc)) { + /* OF_SKIP_POWER_DOWN_INACTIVE_ENCODER */ + return; + } + /* Power-down RX and disable GPU PHY should be paired. + * Disabling PHY without powering down RX may cause + * symbol lock loss, on which we will get DP Sink interrupt. + */ + + /* There is a case for the DP active dongles + * where we want to disable the PHY but keep RX powered, + * for those we need to ignore DP Sink interrupt + * by checking lane count that has been set + * on the last do_enable_output(). + */ + + /* disable transmitter */ + cntl.action = TRANSMITTER_CONTROL_DISABLE; + cntl.transmitter = enc10->base.transmitter; + cntl.hpd_sel = enc10->base.hpd_source; + cntl.signal = signal; + cntl.connector_obj_id = enc10->base.connector; + + result = link_transmitter_control(enc10, &cntl); + + if (result != BP_RESULT_OK) { + DC_LOG_ERROR("%s: Failed to execute VBIOS command table!\n", + __func__); + BREAK_TO_DEBUGGER(); + return; + } + + /* disable encoder */ + if (dc_is_dp_signal(signal)) + link_encoder_disable(enc10); +} + +void dcn10_link_encoder_dp_set_lane_settings( + struct link_encoder *enc, + const struct link_training_settings *link_settings) +{ + struct dcn10_link_encoder *enc10 = TO_DCN10_LINK_ENC(enc); + union dpcd_training_lane_set training_lane_set = { { 0 } }; + int32_t lane = 0; + struct bp_transmitter_control cntl = { 0 }; + + if (!link_settings) { + BREAK_TO_DEBUGGER(); + return; + } + + cntl.action = TRANSMITTER_CONTROL_SET_VOLTAGE_AND_PREEMPASIS; + cntl.transmitter = enc10->base.transmitter; + cntl.connector_obj_id = enc10->base.connector; + cntl.lanes_number = link_settings->link_settings.lane_count; + cntl.hpd_sel = enc10->base.hpd_source; + cntl.pixel_clock = link_settings->link_settings.link_rate * + LINK_RATE_REF_FREQ_IN_KHZ; + + for (lane = 0; lane < link_settings->link_settings.lane_count; lane++) { + /* translate lane settings */ + + training_lane_set.bits.VOLTAGE_SWING_SET = + link_settings->lane_settings[lane].VOLTAGE_SWING; + training_lane_set.bits.PRE_EMPHASIS_SET = + link_settings->lane_settings[lane].PRE_EMPHASIS; + + /* post cursor 2 setting only applies to HBR2 link rate */ + if (link_settings->link_settings.link_rate == LINK_RATE_HIGH2) { + /* this is passed to VBIOS + * to program post cursor 2 level + */ + training_lane_set.bits.POST_CURSOR2_SET = + link_settings->lane_settings[lane].POST_CURSOR2; + } + + cntl.lane_select = lane; + cntl.lane_settings = training_lane_set.raw; + + /* call VBIOS table to set voltage swing and pre-emphasis */ + link_transmitter_control(enc10, &cntl); + } +} + +/* set DP PHY test and training patterns */ +void dcn10_link_encoder_dp_set_phy_pattern( + struct link_encoder *enc, + const struct encoder_set_dp_phy_pattern_param *param) +{ + struct dcn10_link_encoder *enc10 = TO_DCN10_LINK_ENC(enc); + + switch (param->dp_phy_pattern) { + case DP_TEST_PATTERN_TRAINING_PATTERN1: + dcn10_link_encoder_set_dp_phy_pattern_training_pattern(enc, 0); + break; + case DP_TEST_PATTERN_TRAINING_PATTERN2: + dcn10_link_encoder_set_dp_phy_pattern_training_pattern(enc, 1); + break; + case DP_TEST_PATTERN_TRAINING_PATTERN3: + dcn10_link_encoder_set_dp_phy_pattern_training_pattern(enc, 2); + break; + case DP_TEST_PATTERN_TRAINING_PATTERN4: + dcn10_link_encoder_set_dp_phy_pattern_training_pattern(enc, 3); + break; + case DP_TEST_PATTERN_D102: + set_dp_phy_pattern_d102(enc10); + break; + case DP_TEST_PATTERN_SYMBOL_ERROR: + set_dp_phy_pattern_symbol_error(enc10); + break; + case DP_TEST_PATTERN_PRBS7: + set_dp_phy_pattern_prbs7(enc10); + break; + case DP_TEST_PATTERN_80BIT_CUSTOM: + set_dp_phy_pattern_80bit_custom( + enc10, param->custom_pattern); + break; + case DP_TEST_PATTERN_CP2520_1: + set_dp_phy_pattern_hbr2_compliance_cp2520_2(enc10, 1); + break; + case DP_TEST_PATTERN_CP2520_2: + set_dp_phy_pattern_hbr2_compliance_cp2520_2(enc10, 2); + break; + case DP_TEST_PATTERN_CP2520_3: + set_dp_phy_pattern_hbr2_compliance_cp2520_2(enc10, 3); + break; + case DP_TEST_PATTERN_VIDEO_MODE: { + set_dp_phy_pattern_passthrough_mode( + enc10, param->dp_panel_mode); + break; + } + + default: + /* invalid phy pattern */ + ASSERT_CRITICAL(false); + break; + } +} + +static void fill_stream_allocation_row_info( + const struct link_mst_stream_allocation *stream_allocation, + uint32_t *src, + uint32_t *slots) +{ + const struct stream_encoder *stream_enc = stream_allocation->stream_enc; + + if (stream_enc) { + *src = stream_enc->id; + *slots = stream_allocation->slot_count; + } else { + *src = 0; + *slots = 0; + } +} + +/* programs DP MST VC payload allocation */ +void dcn10_link_encoder_update_mst_stream_allocation_table( + struct link_encoder *enc, + const struct link_mst_stream_allocation_table *table) +{ + struct dcn10_link_encoder *enc10 = TO_DCN10_LINK_ENC(enc); + uint32_t value0 = 0; + uint32_t value1 = 0; + uint32_t value2 = 0; + uint32_t slots = 0; + uint32_t src = 0; + uint32_t retries = 0; + + /* For CZ, there are only 3 pipes. So Virtual channel is up 3.*/ + + /* --- Set MSE Stream Attribute - + * Setup VC Payload Table on Tx Side, + * Issue allocation change trigger + * to commit payload on both tx and rx side + */ + + /* we should clean-up table each time */ + + if (table->stream_count >= 1) { + fill_stream_allocation_row_info( + &table->stream_allocations[0], + &src, + &slots); + } else { + src = 0; + slots = 0; + } + + REG_UPDATE_2(DP_MSE_SAT0, + DP_MSE_SAT_SRC0, src, + DP_MSE_SAT_SLOT_COUNT0, slots); + + if (table->stream_count >= 2) { + fill_stream_allocation_row_info( + &table->stream_allocations[1], + &src, + &slots); + } else { + src = 0; + slots = 0; + } + + REG_UPDATE_2(DP_MSE_SAT0, + DP_MSE_SAT_SRC1, src, + DP_MSE_SAT_SLOT_COUNT1, slots); + + if (table->stream_count >= 3) { + fill_stream_allocation_row_info( + &table->stream_allocations[2], + &src, + &slots); + } else { + src = 0; + slots = 0; + } + + REG_UPDATE_2(DP_MSE_SAT1, + DP_MSE_SAT_SRC2, src, + DP_MSE_SAT_SLOT_COUNT2, slots); + + if (table->stream_count >= 4) { + fill_stream_allocation_row_info( + &table->stream_allocations[3], + &src, + &slots); + } else { + src = 0; + slots = 0; + } + + REG_UPDATE_2(DP_MSE_SAT1, + DP_MSE_SAT_SRC3, src, + DP_MSE_SAT_SLOT_COUNT3, slots); + + /* --- wait for transaction finish */ + + /* send allocation change trigger (ACT) ? + * this step first sends the ACT, + * then double buffers the SAT into the hardware + * making the new allocation active on the DP MST mode link + */ + + /* DP_MSE_SAT_UPDATE: + * 0 - No Action + * 1 - Update SAT with trigger + * 2 - Update SAT without trigger + */ + REG_UPDATE(DP_MSE_SAT_UPDATE, + DP_MSE_SAT_UPDATE, 1); + + /* wait for update to complete + * (i.e. DP_MSE_SAT_UPDATE field is reset to 0) + * then wait for the transmission + * of at least 16 MTP headers on immediate local link. + * i.e. DP_MSE_16_MTP_KEEPOUT field (read only) is reset to 0 + * a value of 1 indicates that DP MST mode + * is in the 16 MTP keepout region after a VC has been added. + * MST stream bandwidth (VC rate) can be configured + * after this bit is cleared + */ + do { + udelay(10); + + value0 = REG_READ(DP_MSE_SAT_UPDATE); + + REG_GET(DP_MSE_SAT_UPDATE, + DP_MSE_SAT_UPDATE, &value1); + + REG_GET(DP_MSE_SAT_UPDATE, + DP_MSE_16_MTP_KEEPOUT, &value2); + + /* bit field DP_MSE_SAT_UPDATE is set to 1 already */ + if (!value1 && !value2) + break; + ++retries; + } while (retries < DP_MST_UPDATE_MAX_RETRY); +} + +void dcn10_link_encoder_connect_dig_be_to_fe( + struct link_encoder *enc, + enum engine_id engine, + bool connect) +{ + struct dcn10_link_encoder *enc10 = TO_DCN10_LINK_ENC(enc); + uint32_t field; + + if (engine != ENGINE_ID_UNKNOWN) { + + REG_GET(DIG_BE_CNTL, DIG_FE_SOURCE_SELECT, &field); + + if (connect) + field |= get_frontend_source(engine); + else + field &= ~get_frontend_source(engine); + + REG_UPDATE(DIG_BE_CNTL, DIG_FE_SOURCE_SELECT, field); + } +} + + +#define HPD_REG(reg)\ + (enc10->hpd_regs->reg) + +#define HPD_REG_READ(reg_name) \ + dm_read_reg(CTX, HPD_REG(reg_name)) + +#define HPD_REG_UPDATE_N(reg_name, n, ...) \ + generic_reg_update_ex(CTX, \ + HPD_REG(reg_name), \ + HPD_REG_READ(reg_name), \ + n, __VA_ARGS__) + +#define HPD_REG_UPDATE(reg_name, field, val) \ + HPD_REG_UPDATE_N(reg_name, 1, \ + FN(reg_name, field), val) + +void dcn10_link_encoder_enable_hpd(struct link_encoder *enc) +{ + struct dcn10_link_encoder *enc10 = TO_DCN10_LINK_ENC(enc); + + HPD_REG_UPDATE(DC_HPD_CONTROL, + DC_HPD_EN, 1); +} + +void dcn10_link_encoder_disable_hpd(struct link_encoder *enc) +{ + struct dcn10_link_encoder *enc10 = TO_DCN10_LINK_ENC(enc); + + HPD_REG_UPDATE(DC_HPD_CONTROL, + DC_HPD_EN, 0); +} + + +#define AUX_REG(reg)\ + (enc10->aux_regs->reg) + +#define AUX_REG_READ(reg_name) \ + dm_read_reg(CTX, AUX_REG(reg_name)) + +#define AUX_REG_UPDATE_N(reg_name, n, ...) \ + generic_reg_update_ex(CTX, \ + AUX_REG(reg_name), \ + AUX_REG_READ(reg_name), \ + n, __VA_ARGS__) + +#define AUX_REG_UPDATE(reg_name, field, val) \ + AUX_REG_UPDATE_N(reg_name, 1, \ + FN(reg_name, field), val) + +#define AUX_REG_UPDATE_2(reg, f1, v1, f2, v2) \ + AUX_REG_UPDATE_N(reg, 2,\ + FN(reg, f1), v1,\ + FN(reg, f2), v2) + +static void aux_initialize( + struct dcn10_link_encoder *enc10) +{ + enum hpd_source_id hpd_source = enc10->base.hpd_source; + + AUX_REG_UPDATE_2(AUX_CONTROL, + AUX_HPD_SEL, hpd_source, + AUX_LS_READ_EN, 0); + + /* 1/4 window (the maximum allowed) */ + AUX_REG_UPDATE(AUX_DPHY_RX_CONTROL0, + AUX_RX_RECEIVE_WINDOW, 1); +} diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.h new file mode 100644 index 000000000000..2a97cdb2cfbb --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.h @@ -0,0 +1,330 @@ +/* + * Copyright 2012-15 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef __DC_LINK_ENCODER__DCN10_H__ +#define __DC_LINK_ENCODER__DCN10_H__ + +#include "link_encoder.h" + +#define TO_DCN10_LINK_ENC(link_encoder)\ + container_of(link_encoder, struct dcn10_link_encoder, base) + + +#define AUX_REG_LIST(id)\ + SRI(AUX_CONTROL, DP_AUX, id), \ + SRI(AUX_DPHY_RX_CONTROL0, DP_AUX, id) + +#define HPD_REG_LIST(id)\ + SRI(DC_HPD_CONTROL, HPD, id) + +#define LE_DCN_COMMON_REG_LIST(id) \ + SRI(DIG_BE_CNTL, DIG, id), \ + SRI(DIG_BE_EN_CNTL, DIG, id), \ + SRI(DP_CONFIG, DP, id), \ + SRI(DP_DPHY_CNTL, DP, id), \ + SRI(DP_DPHY_PRBS_CNTL, DP, id), \ + SRI(DP_DPHY_SCRAM_CNTL, DP, id),\ + SRI(DP_DPHY_SYM0, DP, id), \ + SRI(DP_DPHY_SYM1, DP, id), \ + SRI(DP_DPHY_SYM2, DP, id), \ + SRI(DP_DPHY_TRAINING_PATTERN_SEL, DP, id), \ + SRI(DP_LINK_CNTL, DP, id), \ + SRI(DP_LINK_FRAMING_CNTL, DP, id), \ + SRI(DP_MSE_SAT0, DP, id), \ + SRI(DP_MSE_SAT1, DP, id), \ + SRI(DP_MSE_SAT2, DP, id), \ + SRI(DP_MSE_SAT_UPDATE, DP, id), \ + SRI(DP_SEC_CNTL, DP, id), \ + SRI(DP_VID_STREAM_CNTL, DP, id), \ + SRI(DP_DPHY_FAST_TRAINING, DP, id), \ + SRI(DP_SEC_CNTL1, DP, id), \ + SRI(DP_DPHY_BS_SR_SWAP_CNTL, DP, id), \ + SRI(DP_DPHY_INTERNAL_CTRL, DP, id), \ + SRI(DP_DPHY_HBR2_PATTERN_CONTROL, DP, id) + +#define LE_DCN10_REG_LIST(id)\ + LE_DCN_COMMON_REG_LIST(id) + +struct dcn10_link_enc_aux_registers { + uint32_t AUX_CONTROL; + uint32_t AUX_DPHY_RX_CONTROL0; +}; + +struct dcn10_link_enc_hpd_registers { + uint32_t DC_HPD_CONTROL; +}; + +struct dcn10_link_enc_registers { + uint32_t DIG_BE_CNTL; + uint32_t DIG_BE_EN_CNTL; + uint32_t DP_CONFIG; + uint32_t DP_DPHY_CNTL; + uint32_t DP_DPHY_INTERNAL_CTRL; + uint32_t DP_DPHY_PRBS_CNTL; + uint32_t DP_DPHY_SCRAM_CNTL; + uint32_t DP_DPHY_SYM0; + uint32_t DP_DPHY_SYM1; + uint32_t DP_DPHY_SYM2; + uint32_t DP_DPHY_TRAINING_PATTERN_SEL; + uint32_t DP_LINK_CNTL; + uint32_t DP_LINK_FRAMING_CNTL; + uint32_t DP_MSE_SAT0; + uint32_t DP_MSE_SAT1; + uint32_t DP_MSE_SAT2; + uint32_t DP_MSE_SAT_UPDATE; + uint32_t DP_SEC_CNTL; + uint32_t DP_VID_STREAM_CNTL; + uint32_t DP_DPHY_FAST_TRAINING; + uint32_t DP_DPHY_BS_SR_SWAP_CNTL; + uint32_t DP_DPHY_HBR2_PATTERN_CONTROL; + uint32_t DP_SEC_CNTL1; +}; + +#define LE_SF(reg_name, field_name, post_fix)\ + .field_name = reg_name ## __ ## field_name ## post_fix + +#define LINK_ENCODER_MASK_SH_LIST_DCN10(mask_sh)\ + LE_SF(DIG0_DIG_BE_EN_CNTL, DIG_ENABLE, mask_sh),\ + LE_SF(DIG0_DIG_BE_CNTL, DIG_HPD_SELECT, mask_sh),\ + LE_SF(DIG0_DIG_BE_CNTL, DIG_MODE, mask_sh),\ + LE_SF(DIG0_DIG_BE_CNTL, DIG_FE_SOURCE_SELECT, mask_sh),\ + LE_SF(DP0_DP_DPHY_CNTL, DPHY_BYPASS, mask_sh),\ + LE_SF(DP0_DP_DPHY_CNTL, DPHY_ATEST_SEL_LANE0, mask_sh),\ + LE_SF(DP0_DP_DPHY_CNTL, DPHY_ATEST_SEL_LANE1, mask_sh),\ + LE_SF(DP0_DP_DPHY_CNTL, DPHY_ATEST_SEL_LANE2, mask_sh),\ + LE_SF(DP0_DP_DPHY_CNTL, DPHY_ATEST_SEL_LANE3, mask_sh),\ + LE_SF(DP0_DP_DPHY_PRBS_CNTL, DPHY_PRBS_EN, mask_sh),\ + LE_SF(DP0_DP_DPHY_PRBS_CNTL, DPHY_PRBS_SEL, mask_sh),\ + LE_SF(DP0_DP_DPHY_SYM0, DPHY_SYM1, mask_sh),\ + LE_SF(DP0_DP_DPHY_SYM0, DPHY_SYM2, mask_sh),\ + LE_SF(DP0_DP_DPHY_SYM0, DPHY_SYM3, mask_sh),\ + LE_SF(DP0_DP_DPHY_SYM1, DPHY_SYM4, mask_sh),\ + LE_SF(DP0_DP_DPHY_SYM1, DPHY_SYM5, mask_sh),\ + LE_SF(DP0_DP_DPHY_SYM1, DPHY_SYM6, mask_sh),\ + LE_SF(DP0_DP_DPHY_SYM2, DPHY_SYM7, mask_sh),\ + LE_SF(DP0_DP_DPHY_SYM2, DPHY_SYM8, mask_sh),\ + LE_SF(DP0_DP_DPHY_SCRAM_CNTL, DPHY_SCRAMBLER_BS_COUNT, mask_sh),\ + LE_SF(DP0_DP_DPHY_SCRAM_CNTL, DPHY_SCRAMBLER_ADVANCE, mask_sh),\ + LE_SF(DP0_DP_DPHY_FAST_TRAINING, DPHY_RX_FAST_TRAINING_CAPABLE, mask_sh),\ + LE_SF(DP0_DP_DPHY_BS_SR_SWAP_CNTL, DPHY_LOAD_BS_COUNT, mask_sh),\ + LE_SF(DP0_DP_DPHY_TRAINING_PATTERN_SEL, DPHY_TRAINING_PATTERN_SEL, mask_sh),\ + LE_SF(DP0_DP_DPHY_HBR2_PATTERN_CONTROL, DP_DPHY_HBR2_PATTERN_CONTROL, mask_sh),\ + LE_SF(DP0_DP_LINK_CNTL, DP_LINK_TRAINING_COMPLETE, mask_sh),\ + LE_SF(DP0_DP_LINK_FRAMING_CNTL, DP_IDLE_BS_INTERVAL, mask_sh),\ + LE_SF(DP0_DP_LINK_FRAMING_CNTL, DP_VBID_DISABLE, mask_sh),\ + LE_SF(DP0_DP_LINK_FRAMING_CNTL, DP_VID_ENHANCED_FRAME_MODE, mask_sh),\ + LE_SF(DP0_DP_VID_STREAM_CNTL, DP_VID_STREAM_ENABLE, mask_sh),\ + LE_SF(DP0_DP_CONFIG, DP_UDI_LANES, mask_sh),\ + LE_SF(DP0_DP_SEC_CNTL1, DP_SEC_GSP0_LINE_NUM, mask_sh),\ + LE_SF(DP0_DP_SEC_CNTL1, DP_SEC_GSP0_PRIORITY, mask_sh),\ + LE_SF(DP0_DP_MSE_SAT0, DP_MSE_SAT_SRC0, mask_sh),\ + LE_SF(DP0_DP_MSE_SAT0, DP_MSE_SAT_SRC1, mask_sh),\ + LE_SF(DP0_DP_MSE_SAT0, DP_MSE_SAT_SLOT_COUNT0, mask_sh),\ + LE_SF(DP0_DP_MSE_SAT0, DP_MSE_SAT_SLOT_COUNT1, mask_sh),\ + LE_SF(DP0_DP_MSE_SAT1, DP_MSE_SAT_SRC2, mask_sh),\ + LE_SF(DP0_DP_MSE_SAT1, DP_MSE_SAT_SRC3, mask_sh),\ + LE_SF(DP0_DP_MSE_SAT1, DP_MSE_SAT_SLOT_COUNT2, mask_sh),\ + LE_SF(DP0_DP_MSE_SAT1, DP_MSE_SAT_SLOT_COUNT3, mask_sh),\ + LE_SF(DP0_DP_MSE_SAT_UPDATE, DP_MSE_SAT_UPDATE, mask_sh),\ + LE_SF(DP0_DP_MSE_SAT_UPDATE, DP_MSE_16_MTP_KEEPOUT, mask_sh),\ + LE_SF(DP_AUX0_AUX_CONTROL, AUX_HPD_SEL, mask_sh),\ + LE_SF(DP_AUX0_AUX_CONTROL, AUX_LS_READ_EN, mask_sh),\ + LE_SF(DP_AUX0_AUX_DPHY_RX_CONTROL0, AUX_RX_RECEIVE_WINDOW, mask_sh),\ + LE_SF(HPD0_DC_HPD_CONTROL, DC_HPD_EN, mask_sh) + +#define DCN_LINK_ENCODER_REG_FIELD_LIST(type) \ + type DIG_ENABLE;\ + type DIG_HPD_SELECT;\ + type DIG_MODE;\ + type DIG_FE_SOURCE_SELECT;\ + type DPHY_BYPASS;\ + type DPHY_ATEST_SEL_LANE0;\ + type DPHY_ATEST_SEL_LANE1;\ + type DPHY_ATEST_SEL_LANE2;\ + type DPHY_ATEST_SEL_LANE3;\ + type DPHY_PRBS_EN;\ + type DPHY_PRBS_SEL;\ + type DPHY_SYM1;\ + type DPHY_SYM2;\ + type DPHY_SYM3;\ + type DPHY_SYM4;\ + type DPHY_SYM5;\ + type DPHY_SYM6;\ + type DPHY_SYM7;\ + type DPHY_SYM8;\ + type DPHY_SCRAMBLER_BS_COUNT;\ + type DPHY_SCRAMBLER_ADVANCE;\ + type DPHY_RX_FAST_TRAINING_CAPABLE;\ + type DPHY_LOAD_BS_COUNT;\ + type DPHY_TRAINING_PATTERN_SEL;\ + type DP_DPHY_HBR2_PATTERN_CONTROL;\ + type DP_LINK_TRAINING_COMPLETE;\ + type DP_IDLE_BS_INTERVAL;\ + type DP_VBID_DISABLE;\ + type DP_VID_ENHANCED_FRAME_MODE;\ + type DP_VID_STREAM_ENABLE;\ + type DP_UDI_LANES;\ + type DP_SEC_GSP0_LINE_NUM;\ + type DP_SEC_GSP0_PRIORITY;\ + type DP_MSE_SAT_SRC0;\ + type DP_MSE_SAT_SRC1;\ + type DP_MSE_SAT_SRC2;\ + type DP_MSE_SAT_SRC3;\ + type DP_MSE_SAT_SLOT_COUNT0;\ + type DP_MSE_SAT_SLOT_COUNT1;\ + type DP_MSE_SAT_SLOT_COUNT2;\ + type DP_MSE_SAT_SLOT_COUNT3;\ + type DP_MSE_SAT_UPDATE;\ + type DP_MSE_16_MTP_KEEPOUT;\ + type AUX_HPD_SEL;\ + type AUX_LS_READ_EN;\ + type AUX_RX_RECEIVE_WINDOW;\ + type DC_HPD_EN + +struct dcn10_link_enc_shift { + DCN_LINK_ENCODER_REG_FIELD_LIST(uint8_t); +}; + +struct dcn10_link_enc_mask { + DCN_LINK_ENCODER_REG_FIELD_LIST(uint32_t); +}; + +struct dcn10_link_encoder { + struct link_encoder base; + const struct dcn10_link_enc_registers *link_regs; + const struct dcn10_link_enc_aux_registers *aux_regs; + const struct dcn10_link_enc_hpd_registers *hpd_regs; + const struct dcn10_link_enc_shift *link_shift; + const struct dcn10_link_enc_mask *link_mask; +}; + + +void dcn10_link_encoder_construct( + struct dcn10_link_encoder *enc10, + const struct encoder_init_data *init_data, + const struct encoder_feature_support *enc_features, + const struct dcn10_link_enc_registers *link_regs, + const struct dcn10_link_enc_aux_registers *aux_regs, + const struct dcn10_link_enc_hpd_registers *hpd_regs, + const struct dcn10_link_enc_shift *link_shift, + const struct dcn10_link_enc_mask *link_mask); + +bool dcn10_link_encoder_validate_dvi_output( + const struct dcn10_link_encoder *enc10, + enum signal_type connector_signal, + enum signal_type signal, + const struct dc_crtc_timing *crtc_timing); + +bool dcn10_link_encoder_validate_rgb_output( + const struct dcn10_link_encoder *enc10, + const struct dc_crtc_timing *crtc_timing); + +bool dcn10_link_encoder_validate_dp_output( + const struct dcn10_link_encoder *enc10, + const struct dc_crtc_timing *crtc_timing); + +bool dcn10_link_encoder_validate_wireless_output( + const struct dcn10_link_encoder *enc10, + const struct dc_crtc_timing *crtc_timing); + +bool dcn10_link_encoder_validate_output_with_stream( + struct link_encoder *enc, + const struct dc_stream_state *stream); + +/****************** HW programming ************************/ + +/* initialize HW */ /* why do we initialze aux in here? */ +void dcn10_link_encoder_hw_init(struct link_encoder *enc); + +void dcn10_link_encoder_destroy(struct link_encoder **enc); + +/* program DIG_MODE in DIG_BE */ +/* TODO can this be combined with enable_output? */ +void dcn10_link_encoder_setup( + struct link_encoder *enc, + enum signal_type signal); + +/* enables TMDS PHY output */ +/* TODO: still need depth or just pass in adjusted pixel clock? */ +void dcn10_link_encoder_enable_tmds_output( + struct link_encoder *enc, + enum clock_source_id clock_source, + enum dc_color_depth color_depth, + enum signal_type signal, + uint32_t pixel_clock); + +/* enables DP PHY output */ +void dcn10_link_encoder_enable_dp_output( + struct link_encoder *enc, + const struct dc_link_settings *link_settings, + enum clock_source_id clock_source); + +/* enables DP PHY output in MST mode */ +void dcn10_link_encoder_enable_dp_mst_output( + struct link_encoder *enc, + const struct dc_link_settings *link_settings, + enum clock_source_id clock_source); + +/* disable PHY output */ +void dcn10_link_encoder_disable_output( + struct link_encoder *enc, + enum signal_type signal); + +/* set DP lane settings */ +void dcn10_link_encoder_dp_set_lane_settings( + struct link_encoder *enc, + const struct link_training_settings *link_settings); + +void dcn10_link_encoder_dp_set_phy_pattern( + struct link_encoder *enc, + const struct encoder_set_dp_phy_pattern_param *param); + +/* programs DP MST VC payload allocation */ +void dcn10_link_encoder_update_mst_stream_allocation_table( + struct link_encoder *enc, + const struct link_mst_stream_allocation_table *table); + +void dcn10_link_encoder_connect_dig_be_to_fe( + struct link_encoder *enc, + enum engine_id engine, + bool connect); + +void dcn10_link_encoder_set_dp_phy_pattern_training_pattern( + struct link_encoder *enc, + uint32_t index); + +void dcn10_link_encoder_enable_hpd(struct link_encoder *enc); + +void dcn10_link_encoder_disable_hpd(struct link_encoder *enc); + +void dcn10_psr_program_dp_dphy_fast_training(struct link_encoder *enc, + bool exit_link_training_required); + +void dcn10_psr_program_secondary_packet(struct link_encoder *enc, + unsigned int sdp_transmit_line_num_deadline); + +bool dcn10_is_dig_enabled(struct link_encoder *enc); + +#endif /* __DC_LINK_ENCODER__DCN10_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c index c734b7fa5835..f2fbce0e3fc5 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c @@ -360,7 +360,7 @@ void optc1_program_timing( } -static void optc1_set_blank_data_double_buffer(struct timing_generator *optc, bool enable) +void optc1_set_blank_data_double_buffer(struct timing_generator *optc, bool enable) { struct optc *optc1 = DCN10TG_FROM_TG(optc); @@ -1257,20 +1257,20 @@ void optc1_read_otg_state(struct optc *optc1, OPTC_UNDERFLOW_OCCURRED_STATUS, &s->underflow_occurred_status); } -static void optc1_clear_optc_underflow(struct timing_generator *optc) +void optc1_clear_optc_underflow(struct timing_generator *optc) { struct optc *optc1 = DCN10TG_FROM_TG(optc); REG_UPDATE(OPTC_INPUT_GLOBAL_CONTROL, OPTC_UNDERFLOW_CLEAR, 1); } -static void optc1_tg_init(struct timing_generator *optc) +void optc1_tg_init(struct timing_generator *optc) { optc1_set_blank_data_double_buffer(optc, true); optc1_clear_optc_underflow(optc); } -static bool optc1_is_tg_enabled(struct timing_generator *optc) +bool optc1_is_tg_enabled(struct timing_generator *optc) { struct optc *optc1 = DCN10TG_FROM_TG(optc); uint32_t otg_enabled = 0; @@ -1281,7 +1281,7 @@ static bool optc1_is_tg_enabled(struct timing_generator *optc) } -static bool optc1_is_optc_underflow_occurred(struct timing_generator *optc) +bool optc1_is_optc_underflow_occurred(struct timing_generator *optc) { struct optc *optc1 = DCN10TG_FROM_TG(optc); uint32_t underflow_occurred = 0; diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.h index 89e09e5327a2..c62052f46460 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.h +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.h @@ -497,4 +497,14 @@ void optc1_program_stereo(struct timing_generator *optc, bool optc1_is_stereo_left_eye(struct timing_generator *optc); +void optc1_clear_optc_underflow(struct timing_generator *optc); + +void optc1_tg_init(struct timing_generator *optc); + +bool optc1_is_tg_enabled(struct timing_generator *optc); + +bool optc1_is_optc_underflow_occurred(struct timing_generator *optc); + +void optc1_set_blank_data_double_buffer(struct timing_generator *optc, bool enable); + #endif /* __DC_TIMING_GENERATOR_DCN10_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c index 2c0a3150bf2d..df5cb2d1d164 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c @@ -38,7 +38,7 @@ #include "dcn10/dcn10_hw_sequencer.h" #include "dce110/dce110_hw_sequencer.h" #include "dcn10/dcn10_opp.h" -#include "dce/dce_link_encoder.h" +#include "dcn10/dcn10_link_encoder.h" #include "dcn10/dcn10_stream_encoder.h" #include "dce/dce_clocks.h" #include "dce/dce_clock_source.h" @@ -214,13 +214,11 @@ static const struct dce_aduio_mask audio_mask = { AUX_REG_LIST(id)\ } -static const struct dce110_link_enc_aux_registers link_enc_aux_regs[] = { +static const struct dcn10_link_enc_aux_registers link_enc_aux_regs[] = { aux_regs(0), aux_regs(1), aux_regs(2), - aux_regs(3), - aux_regs(4), - aux_regs(5) + aux_regs(3) }; #define hpd_regs(id)\ @@ -228,13 +226,11 @@ static const struct dce110_link_enc_aux_registers link_enc_aux_regs[] = { HPD_REG_LIST(id)\ } -static const struct dce110_link_enc_hpd_registers link_enc_hpd_regs[] = { +static const struct dcn10_link_enc_hpd_registers link_enc_hpd_regs[] = { hpd_regs(0), hpd_regs(1), hpd_regs(2), - hpd_regs(3), - hpd_regs(4), - hpd_regs(5) + hpd_regs(3) }; #define link_regs(id)\ @@ -243,14 +239,19 @@ static const struct dce110_link_enc_hpd_registers link_enc_hpd_regs[] = { SRI(DP_DPHY_INTERNAL_CTRL, DP, id) \ } -static const struct dce110_link_enc_registers link_enc_regs[] = { +static const struct dcn10_link_enc_registers link_enc_regs[] = { link_regs(0), link_regs(1), link_regs(2), - link_regs(3), - link_regs(4), - link_regs(5), - link_regs(6), + link_regs(3) +}; + +static const struct dcn10_link_enc_shift le_shift = { + LINK_ENCODER_MASK_SH_LIST_DCN10(__SHIFT) +}; + +static const struct dcn10_link_enc_mask le_mask = { + LINK_ENCODER_MASK_SH_LIST_DCN10(_MASK) }; #define ipp_regs(id)\ @@ -446,6 +447,8 @@ static const struct dc_debug debug_defaults_drv = { .vsr_support = true, .performance_trace = false, .az_endpoint_mute_only = true, + .recovery_enabled = false, /*enable this by default after testing.*/ + .max_downscale_src_width = 3840, }; static const struct dc_debug debug_defaults_diags = { @@ -581,20 +584,22 @@ static const struct encoder_feature_support link_enc_feature = { struct link_encoder *dcn10_link_encoder_create( const struct encoder_init_data *enc_init_data) { - struct dce110_link_encoder *enc110 = - kzalloc(sizeof(struct dce110_link_encoder), GFP_KERNEL); + struct dcn10_link_encoder *enc10 = + kzalloc(sizeof(struct dcn10_link_encoder), GFP_KERNEL); - if (!enc110) + if (!enc10) return NULL; - dce110_link_encoder_construct(enc110, + dcn10_link_encoder_construct(enc10, enc_init_data, &link_enc_feature, &link_enc_regs[enc_init_data->transmitter], &link_enc_aux_regs[enc_init_data->channel - 1], - &link_enc_hpd_regs[enc_init_data->hpd_source]); + &link_enc_hpd_regs[enc_init_data->hpd_source], + &le_shift, + &le_mask); - return &enc110->base; + return &enc10->base; } struct clock_source *dcn10_clock_source_create( @@ -1021,6 +1026,7 @@ static bool construct( dc->caps.max_cursor_size = 256; dc->caps.max_slave_planes = 1; dc->caps.is_apu = true; + dc->caps.post_blend_color_processing = false; if (dc->ctx->dce_environment == DCE_ENV_PRODUCTION_DRV) dc->debug = debug_defaults_drv; diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.c index befd8639ad55..653b7b2efe2e 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.c @@ -603,11 +603,11 @@ void enc1_stream_encoder_set_mst_bandwidth( struct fixed31_32 avg_time_slots_per_mtp) { struct dcn10_stream_encoder *enc1 = DCN10STRENC_FROM_STRENC(enc); - uint32_t x = dal_fixed31_32_floor( + uint32_t x = dc_fixpt_floor( avg_time_slots_per_mtp); - uint32_t y = dal_fixed31_32_ceil( - dal_fixed31_32_shl( - dal_fixed31_32_sub_int( + uint32_t y = dc_fixpt_ceil( + dc_fixpt_shl( + dc_fixpt_sub_int( avg_time_slots_per_mtp, x), 26)); diff --git a/drivers/gpu/drm/amd/display/dc/dm_services.h b/drivers/gpu/drm/amd/display/dc/dm_services.h index 8eafe1af8a5e..4ff9b2bba178 100644 --- a/drivers/gpu/drm/amd/display/dc/dm_services.h +++ b/drivers/gpu/drm/amd/display/dc/dm_services.h @@ -355,10 +355,6 @@ void dm_perf_trace_timestamp(const char *func_name, unsigned int line); /* * Debug and verification hooks */ -bool dm_helpers_dc_conn_log( - struct dc_context *ctx, - struct log_entry *entry, - enum dc_log_type event); void dm_dtn_log_begin(struct dc_context *ctx); void dm_dtn_log_append_v(struct dc_context *ctx, const char *msg, ...); diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_mode_enums.h b/drivers/gpu/drm/amd/display/dc/dml/display_mode_enums.h index b1ad3553f900..47c19f8fe7d1 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/display_mode_enums.h +++ b/drivers/gpu/drm/amd/display/dc/dml/display_mode_enums.h @@ -108,4 +108,17 @@ enum output_standard { dm_std_uninitialized = 0, dm_std_cvtr2, dm_std_cvt }; +enum mpc_combine_affinity { + dm_mpc_always_when_possible, + dm_mpc_reduce_voltage, + dm_mpc_reduce_voltage_and_clocks +}; + +enum self_refresh_affinity { + dm_try_to_allow_self_refresh_and_mclk_switch, + dm_allow_self_refresh_and_mclk_switch, + dm_allow_self_refresh, + dm_neither_self_refresh_nor_mclk_switch +}; + #endif diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h b/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h index ce750edc1e5f..7fa0375939ae 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h +++ b/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h @@ -25,39 +25,39 @@ #ifndef __DISPLAY_MODE_STRUCTS_H__ #define __DISPLAY_MODE_STRUCTS_H__ -typedef struct _vcs_dpi_voltage_scaling_st voltage_scaling_st; -typedef struct _vcs_dpi_soc_bounding_box_st soc_bounding_box_st; -typedef struct _vcs_dpi_ip_params_st ip_params_st; -typedef struct _vcs_dpi_display_pipe_source_params_st display_pipe_source_params_st; -typedef struct _vcs_dpi_display_output_params_st display_output_params_st; -typedef struct _vcs_dpi_display_bandwidth_st display_bandwidth_st; -typedef struct _vcs_dpi_scaler_ratio_depth_st scaler_ratio_depth_st; -typedef struct _vcs_dpi_scaler_taps_st scaler_taps_st; -typedef struct _vcs_dpi_display_pipe_dest_params_st display_pipe_dest_params_st; -typedef struct _vcs_dpi_display_pipe_params_st display_pipe_params_st; -typedef struct _vcs_dpi_display_clocks_and_cfg_st display_clocks_and_cfg_st; -typedef struct _vcs_dpi_display_e2e_pipe_params_st display_e2e_pipe_params_st; -typedef struct _vcs_dpi_dchub_buffer_sizing_st dchub_buffer_sizing_st; -typedef struct _vcs_dpi_watermarks_perf_st watermarks_perf_st; -typedef struct _vcs_dpi_cstate_pstate_watermarks_st cstate_pstate_watermarks_st; -typedef struct _vcs_dpi_wm_calc_pipe_params_st wm_calc_pipe_params_st; -typedef struct _vcs_dpi_vratio_pre_st vratio_pre_st; -typedef struct _vcs_dpi_display_data_rq_misc_params_st display_data_rq_misc_params_st; -typedef struct _vcs_dpi_display_data_rq_sizing_params_st display_data_rq_sizing_params_st; -typedef struct _vcs_dpi_display_data_rq_dlg_params_st display_data_rq_dlg_params_st; -typedef struct _vcs_dpi_display_cur_rq_dlg_params_st display_cur_rq_dlg_params_st; -typedef struct _vcs_dpi_display_rq_dlg_params_st display_rq_dlg_params_st; -typedef struct _vcs_dpi_display_rq_sizing_params_st display_rq_sizing_params_st; -typedef struct _vcs_dpi_display_rq_misc_params_st display_rq_misc_params_st; -typedef struct _vcs_dpi_display_rq_params_st display_rq_params_st; -typedef struct _vcs_dpi_display_dlg_regs_st display_dlg_regs_st; -typedef struct _vcs_dpi_display_ttu_regs_st display_ttu_regs_st; -typedef struct _vcs_dpi_display_data_rq_regs_st display_data_rq_regs_st; -typedef struct _vcs_dpi_display_rq_regs_st display_rq_regs_st; -typedef struct _vcs_dpi_display_dlg_sys_params_st display_dlg_sys_params_st; -typedef struct _vcs_dpi_display_dlg_prefetch_param_st display_dlg_prefetch_param_st; -typedef struct _vcs_dpi_display_pipe_clock_st display_pipe_clock_st; -typedef struct _vcs_dpi_display_arb_params_st display_arb_params_st; +typedef struct _vcs_dpi_voltage_scaling_st voltage_scaling_st; +typedef struct _vcs_dpi_soc_bounding_box_st soc_bounding_box_st; +typedef struct _vcs_dpi_ip_params_st ip_params_st; +typedef struct _vcs_dpi_display_pipe_source_params_st display_pipe_source_params_st; +typedef struct _vcs_dpi_display_output_params_st display_output_params_st; +typedef struct _vcs_dpi_display_bandwidth_st display_bandwidth_st; +typedef struct _vcs_dpi_scaler_ratio_depth_st scaler_ratio_depth_st; +typedef struct _vcs_dpi_scaler_taps_st scaler_taps_st; +typedef struct _vcs_dpi_display_pipe_dest_params_st display_pipe_dest_params_st; +typedef struct _vcs_dpi_display_pipe_params_st display_pipe_params_st; +typedef struct _vcs_dpi_display_clocks_and_cfg_st display_clocks_and_cfg_st; +typedef struct _vcs_dpi_display_e2e_pipe_params_st display_e2e_pipe_params_st; +typedef struct _vcs_dpi_dchub_buffer_sizing_st dchub_buffer_sizing_st; +typedef struct _vcs_dpi_watermarks_perf_st watermarks_perf_st; +typedef struct _vcs_dpi_cstate_pstate_watermarks_st cstate_pstate_watermarks_st; +typedef struct _vcs_dpi_wm_calc_pipe_params_st wm_calc_pipe_params_st; +typedef struct _vcs_dpi_vratio_pre_st vratio_pre_st; +typedef struct _vcs_dpi_display_data_rq_misc_params_st display_data_rq_misc_params_st; +typedef struct _vcs_dpi_display_data_rq_sizing_params_st display_data_rq_sizing_params_st; +typedef struct _vcs_dpi_display_data_rq_dlg_params_st display_data_rq_dlg_params_st; +typedef struct _vcs_dpi_display_cur_rq_dlg_params_st display_cur_rq_dlg_params_st; +typedef struct _vcs_dpi_display_rq_dlg_params_st display_rq_dlg_params_st; +typedef struct _vcs_dpi_display_rq_sizing_params_st display_rq_sizing_params_st; +typedef struct _vcs_dpi_display_rq_misc_params_st display_rq_misc_params_st; +typedef struct _vcs_dpi_display_rq_params_st display_rq_params_st; +typedef struct _vcs_dpi_display_dlg_regs_st display_dlg_regs_st; +typedef struct _vcs_dpi_display_ttu_regs_st display_ttu_regs_st; +typedef struct _vcs_dpi_display_data_rq_regs_st display_data_rq_regs_st; +typedef struct _vcs_dpi_display_rq_regs_st display_rq_regs_st; +typedef struct _vcs_dpi_display_dlg_sys_params_st display_dlg_sys_params_st; +typedef struct _vcs_dpi_display_dlg_prefetch_param_st display_dlg_prefetch_param_st; +typedef struct _vcs_dpi_display_pipe_clock_st display_pipe_clock_st; +typedef struct _vcs_dpi_display_arb_params_st display_arb_params_st; struct _vcs_dpi_voltage_scaling_st { int state; @@ -72,89 +72,107 @@ struct _vcs_dpi_voltage_scaling_st { double dppclk_mhz; }; -struct _vcs_dpi_soc_bounding_box_st { - double sr_exit_time_us; - double sr_enter_plus_exit_time_us; - double urgent_latency_us; - double writeback_latency_us; - double ideal_dram_bw_after_urgent_percent; - unsigned int max_request_size_bytes; - double downspread_percent; - double dram_page_open_time_ns; - double dram_rw_turnaround_time_ns; - double dram_return_buffer_per_channel_bytes; - double dram_channel_width_bytes; +struct _vcs_dpi_soc_bounding_box_st { + double sr_exit_time_us; + double sr_enter_plus_exit_time_us; + double urgent_latency_us; + double urgent_latency_pixel_data_only_us; + double urgent_latency_pixel_mixed_with_vm_data_us; + double urgent_latency_vm_data_only_us; + double writeback_latency_us; + double ideal_dram_bw_after_urgent_percent; + double pct_ideal_dram_sdp_bw_after_urgent_pixel_only; // PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyPixelDataOnly + double pct_ideal_dram_sdp_bw_after_urgent_pixel_and_vm; + double pct_ideal_dram_sdp_bw_after_urgent_vm_only; + double max_avg_sdp_bw_use_normal_percent; + double max_avg_dram_bw_use_normal_percent; + unsigned int max_request_size_bytes; + double downspread_percent; + double dram_page_open_time_ns; + double dram_rw_turnaround_time_ns; + double dram_return_buffer_per_channel_bytes; + double dram_channel_width_bytes; double fabric_datapath_to_dcn_data_return_bytes; double dcn_downspread_percent; double dispclk_dppclk_vco_speed_mhz; double dfs_vco_period_ps; - unsigned int round_trip_ping_latency_dcfclk_cycles; - unsigned int urgent_out_of_order_return_per_channel_bytes; - unsigned int channel_interleave_bytes; - unsigned int num_banks; - unsigned int num_chans; - unsigned int vmm_page_size_bytes; - double dram_clock_change_latency_us; - double writeback_dram_clock_change_latency_us; - unsigned int return_bus_width_bytes; - unsigned int voltage_override; - double xfc_bus_transport_time_us; - double xfc_xbuf_latency_tolerance_us; + unsigned int urgent_out_of_order_return_per_channel_pixel_only_bytes; + unsigned int urgent_out_of_order_return_per_channel_pixel_and_vm_bytes; + unsigned int urgent_out_of_order_return_per_channel_vm_only_bytes; + unsigned int round_trip_ping_latency_dcfclk_cycles; + unsigned int urgent_out_of_order_return_per_channel_bytes; + unsigned int channel_interleave_bytes; + unsigned int num_banks; + unsigned int num_chans; + unsigned int vmm_page_size_bytes; + double dram_clock_change_latency_us; + double writeback_dram_clock_change_latency_us; + unsigned int return_bus_width_bytes; + unsigned int voltage_override; + double xfc_bus_transport_time_us; + double xfc_xbuf_latency_tolerance_us; + int use_urgent_burst_bw; struct _vcs_dpi_voltage_scaling_st clock_limits[7]; }; -struct _vcs_dpi_ip_params_st { - unsigned int max_inter_dcn_tile_repeaters; - unsigned int num_dsc; - unsigned int odm_capable; - unsigned int rob_buffer_size_kbytes; - unsigned int det_buffer_size_kbytes; - unsigned int dpte_buffer_size_in_pte_reqs; - unsigned int pde_proc_buffer_size_64k_reqs; - unsigned int dpp_output_buffer_pixels; - unsigned int opp_output_buffer_lines; - unsigned int pixel_chunk_size_kbytes; - unsigned char pte_enable; - unsigned int pte_chunk_size_kbytes; - unsigned int meta_chunk_size_kbytes; - unsigned int writeback_chunk_size_kbytes; - unsigned int line_buffer_size_bits; - unsigned int max_line_buffer_lines; - unsigned int writeback_luma_buffer_size_kbytes; - unsigned int writeback_chroma_buffer_size_kbytes; - unsigned int writeback_chroma_line_buffer_width_pixels; - unsigned int max_page_table_levels; - unsigned int max_num_dpp; - unsigned int max_num_otg; - unsigned int cursor_chunk_size; - unsigned int cursor_buffer_size; - unsigned int max_num_wb; - unsigned int max_dchub_pscl_bw_pix_per_clk; - unsigned int max_pscl_lb_bw_pix_per_clk; - unsigned int max_lb_vscl_bw_pix_per_clk; - unsigned int max_vscl_hscl_bw_pix_per_clk; - double max_hscl_ratio; - double max_vscl_ratio; - unsigned int hscl_mults; - unsigned int vscl_mults; - unsigned int max_hscl_taps; - unsigned int max_vscl_taps; - unsigned int xfc_supported; - unsigned int xfc_fill_constant_bytes; - double dispclk_ramp_margin_percent; - double xfc_fill_bw_overhead_percent; - double underscan_factor; - unsigned int min_vblank_lines; - unsigned int dppclk_delay_subtotal; - unsigned int dispclk_delay_subtotal; - unsigned int dcfclk_cstate_latency; - unsigned int dppclk_delay_scl; - unsigned int dppclk_delay_scl_lb_only; - unsigned int dppclk_delay_cnvc_formatter; - unsigned int dppclk_delay_cnvc_cursor; - unsigned int is_line_buffer_bpp_fixed; - unsigned int line_buffer_fixed_bpp; - unsigned int dcc_supported; +struct _vcs_dpi_ip_params_st { + bool gpuvm_enable; + bool hostvm_enable; + unsigned int gpuvm_max_page_table_levels; + unsigned int hostvm_max_page_table_levels; + unsigned int hostvm_cached_page_table_levels; + unsigned int pte_group_size_bytes; + unsigned int max_inter_dcn_tile_repeaters; + unsigned int num_dsc; + unsigned int odm_capable; + unsigned int rob_buffer_size_kbytes; + unsigned int det_buffer_size_kbytes; + unsigned int dpte_buffer_size_in_pte_reqs; + unsigned int pde_proc_buffer_size_64k_reqs; + unsigned int dpp_output_buffer_pixels; + unsigned int opp_output_buffer_lines; + unsigned int pixel_chunk_size_kbytes; + unsigned char pte_enable; + unsigned int pte_chunk_size_kbytes; + unsigned int meta_chunk_size_kbytes; + unsigned int writeback_chunk_size_kbytes; + unsigned int line_buffer_size_bits; + unsigned int max_line_buffer_lines; + unsigned int writeback_luma_buffer_size_kbytes; + unsigned int writeback_chroma_buffer_size_kbytes; + unsigned int writeback_chroma_line_buffer_width_pixels; + unsigned int max_page_table_levels; + unsigned int max_num_dpp; + unsigned int max_num_otg; + unsigned int cursor_chunk_size; + unsigned int cursor_buffer_size; + unsigned int max_num_wb; + unsigned int max_dchub_pscl_bw_pix_per_clk; + unsigned int max_pscl_lb_bw_pix_per_clk; + unsigned int max_lb_vscl_bw_pix_per_clk; + unsigned int max_vscl_hscl_bw_pix_per_clk; + double max_hscl_ratio; + double max_vscl_ratio; + unsigned int hscl_mults; + unsigned int vscl_mults; + unsigned int max_hscl_taps; + unsigned int max_vscl_taps; + unsigned int xfc_supported; + unsigned int xfc_fill_constant_bytes; + double dispclk_ramp_margin_percent; + double xfc_fill_bw_overhead_percent; + double underscan_factor; + unsigned int min_vblank_lines; + unsigned int dppclk_delay_subtotal; + unsigned int dispclk_delay_subtotal; + unsigned int dcfclk_cstate_latency; + unsigned int dppclk_delay_scl; + unsigned int dppclk_delay_scl_lb_only; + unsigned int dppclk_delay_cnvc_formatter; + unsigned int dppclk_delay_cnvc_cursor; + unsigned int is_line_buffer_bpp_fixed; + unsigned int line_buffer_fixed_bpp; + unsigned int dcc_supported; unsigned int IsLineBufferBppFixed; unsigned int LineBufferFixedBpp; @@ -169,41 +187,45 @@ struct _vcs_dpi_display_xfc_params_st { int xfc_slv_chunk_size_bytes; }; -struct _vcs_dpi_display_pipe_source_params_st { - int source_format; - unsigned char dcc; - unsigned int dcc_override; - unsigned int dcc_rate; - unsigned char dcc_use_global; - unsigned char vm; - unsigned char vm_levels_force_en; - unsigned int vm_levels_force; - int source_scan; - int sw_mode; - int macro_tile_size; - unsigned char is_display_sw; - unsigned int viewport_width; - unsigned int viewport_height; - unsigned int viewport_y_y; - unsigned int viewport_y_c; - unsigned int viewport_width_c; - unsigned int viewport_height_c; - unsigned int data_pitch; - unsigned int data_pitch_c; - unsigned int meta_pitch; - unsigned int meta_pitch_c; - unsigned int cur0_src_width; - int cur0_bpp; - unsigned int cur1_src_width; - int cur1_bpp; - int num_cursors; - unsigned char is_hsplit; - unsigned char dynamic_metadata_enable; - unsigned int dynamic_metadata_lines_before_active; - unsigned int dynamic_metadata_xmit_bytes; - unsigned int hsplit_grp; - unsigned char xfc_enable; - unsigned char xfc_slave; +struct _vcs_dpi_display_pipe_source_params_st { + int source_format; + unsigned char dcc; + unsigned int dcc_override; + unsigned int dcc_rate; + unsigned char dcc_use_global; + unsigned char vm; + bool gpuvm; // gpuvm enabled + bool hostvm; // hostvm enabled + bool gpuvm_levels_force_en; + unsigned int gpuvm_levels_force; + bool hostvm_levels_force_en; + unsigned int hostvm_levels_force; + int source_scan; + int sw_mode; + int macro_tile_size; + unsigned char is_display_sw; + unsigned int viewport_width; + unsigned int viewport_height; + unsigned int viewport_y_y; + unsigned int viewport_y_c; + unsigned int viewport_width_c; + unsigned int viewport_height_c; + unsigned int data_pitch; + unsigned int data_pitch_c; + unsigned int meta_pitch; + unsigned int meta_pitch_c; + unsigned int cur0_src_width; + int cur0_bpp; + unsigned int cur1_src_width; + int cur1_bpp; + int num_cursors; + unsigned char is_hsplit; + unsigned char dynamic_metadata_enable; + unsigned int dynamic_metadata_lines_before_active; + unsigned int dynamic_metadata_xmit_bytes; + unsigned int hsplit_grp; + unsigned char xfc_enable; + unsigned char xfc_slave; struct _vcs_dpi_display_xfc_params_st xfc_params; }; struct writeback_st { @@ -219,335 +241,335 @@ struct writeback_st { double wb_vratio; }; -struct _vcs_dpi_display_output_params_st { - int dp_lanes; - int output_bpp; - int dsc_enable; - int wb_enable; - int num_active_wb; - int opp_input_bpc; - int output_type; - int output_format; - int output_standard; - int dsc_slices; +struct _vcs_dpi_display_output_params_st { + int dp_lanes; + int output_bpp; + int dsc_enable; + int wb_enable; + int num_active_wb; + int output_bpc; + int output_type; + int output_format; + int output_standard; + int dsc_slices; struct writeback_st wb; }; -struct _vcs_dpi_display_bandwidth_st { - double total_bw_consumed_gbps; - double guaranteed_urgent_return_bw_gbps; -}; - -struct _vcs_dpi_scaler_ratio_depth_st { - double hscl_ratio; - double vscl_ratio; - double hscl_ratio_c; - double vscl_ratio_c; - double vinit; - double vinit_c; - double vinit_bot; - double vinit_bot_c; - int lb_depth; - int scl_enable; -}; - -struct _vcs_dpi_scaler_taps_st { - unsigned int htaps; - unsigned int vtaps; - unsigned int htaps_c; - unsigned int vtaps_c; -}; - -struct _vcs_dpi_display_pipe_dest_params_st { - unsigned int recout_width; - unsigned int recout_height; - unsigned int full_recout_width; - unsigned int full_recout_height; - unsigned int hblank_start; - unsigned int hblank_end; - unsigned int vblank_start; - unsigned int vblank_end; - unsigned int htotal; - unsigned int vtotal; - unsigned int vactive; - unsigned int hactive; - unsigned int vstartup_start; - unsigned int vupdate_offset; - unsigned int vupdate_width; - unsigned int vready_offset; - unsigned char interlaced; - unsigned char underscan; - double pixel_rate_mhz; - unsigned char synchronized_vblank_all_planes; - unsigned char otg_inst; - unsigned char odm_split_cnt; - unsigned char odm_combine; -}; - -struct _vcs_dpi_display_pipe_params_st { - display_pipe_source_params_st src; - display_pipe_dest_params_st dest; - scaler_ratio_depth_st scale_ratio_depth; - scaler_taps_st scale_taps; -}; - -struct _vcs_dpi_display_clocks_and_cfg_st { - int voltage; - double dppclk_mhz; - double refclk_mhz; - double dispclk_mhz; - double dcfclk_mhz; - double socclk_mhz; -}; - -struct _vcs_dpi_display_e2e_pipe_params_st { - display_pipe_params_st pipe; - display_output_params_st dout; - display_clocks_and_cfg_st clks_cfg; -}; - -struct _vcs_dpi_dchub_buffer_sizing_st { - unsigned int swath_width_y; - unsigned int swath_height_y; - unsigned int swath_height_c; - unsigned int detail_buffer_size_y; -}; - -struct _vcs_dpi_watermarks_perf_st { - double stutter_eff_in_active_region_percent; - double urgent_latency_supported_us; - double non_urgent_latency_supported_us; - double dram_clock_change_margin_us; - double dram_access_eff_percent; -}; - -struct _vcs_dpi_cstate_pstate_watermarks_st { - double cstate_exit_us; - double cstate_enter_plus_exit_us; - double pstate_change_us; -}; - -struct _vcs_dpi_wm_calc_pipe_params_st { - unsigned int num_dpp; - int voltage; - int output_type; - double dcfclk_mhz; - double socclk_mhz; - double dppclk_mhz; - double pixclk_mhz; - unsigned char interlace_en; - unsigned char pte_enable; - unsigned char dcc_enable; - double dcc_rate; - double bytes_per_pixel_c; - double bytes_per_pixel_y; - unsigned int swath_width_y; - unsigned int swath_height_y; - unsigned int swath_height_c; - unsigned int det_buffer_size_y; - double h_ratio; - double v_ratio; - unsigned int h_taps; - unsigned int h_total; - unsigned int v_total; - unsigned int v_active; - unsigned int e2e_index; - double display_pipe_line_delivery_time; - double read_bw; - unsigned int lines_in_det_y; - unsigned int lines_in_det_y_rounded_down_to_swath; - double full_det_buffering_time; - double dcfclk_deepsleep_mhz_per_plane; -}; - -struct _vcs_dpi_vratio_pre_st { - double vratio_pre_l; - double vratio_pre_c; -}; - -struct _vcs_dpi_display_data_rq_misc_params_st { - unsigned int full_swath_bytes; - unsigned int stored_swath_bytes; - unsigned int blk256_height; - unsigned int blk256_width; - unsigned int req_height; - unsigned int req_width; -}; - -struct _vcs_dpi_display_data_rq_sizing_params_st { - unsigned int chunk_bytes; - unsigned int min_chunk_bytes; - unsigned int meta_chunk_bytes; - unsigned int min_meta_chunk_bytes; - unsigned int mpte_group_bytes; - unsigned int dpte_group_bytes; -}; - -struct _vcs_dpi_display_data_rq_dlg_params_st { - unsigned int swath_width_ub; - unsigned int swath_height; - unsigned int req_per_swath_ub; - unsigned int meta_pte_bytes_per_frame_ub; - unsigned int dpte_req_per_row_ub; - unsigned int dpte_groups_per_row_ub; - unsigned int dpte_row_height; - unsigned int dpte_bytes_per_row_ub; - unsigned int meta_chunks_per_row_ub; - unsigned int meta_req_per_row_ub; - unsigned int meta_row_height; - unsigned int meta_bytes_per_row_ub; -}; - -struct _vcs_dpi_display_cur_rq_dlg_params_st { - unsigned char enable; - unsigned int swath_height; - unsigned int req_per_line; -}; - -struct _vcs_dpi_display_rq_dlg_params_st { - display_data_rq_dlg_params_st rq_l; - display_data_rq_dlg_params_st rq_c; - display_cur_rq_dlg_params_st rq_cur0; -}; - -struct _vcs_dpi_display_rq_sizing_params_st { - display_data_rq_sizing_params_st rq_l; - display_data_rq_sizing_params_st rq_c; -}; - -struct _vcs_dpi_display_rq_misc_params_st { - display_data_rq_misc_params_st rq_l; - display_data_rq_misc_params_st rq_c; -}; - -struct _vcs_dpi_display_rq_params_st { - unsigned char yuv420; - unsigned char yuv420_10bpc; - display_rq_misc_params_st misc; - display_rq_sizing_params_st sizing; - display_rq_dlg_params_st dlg; -}; - -struct _vcs_dpi_display_dlg_regs_st { - unsigned int refcyc_h_blank_end; - unsigned int dlg_vblank_end; - unsigned int min_dst_y_next_start; - unsigned int refcyc_per_htotal; - unsigned int refcyc_x_after_scaler; - unsigned int dst_y_after_scaler; - unsigned int dst_y_prefetch; - unsigned int dst_y_per_vm_vblank; - unsigned int dst_y_per_row_vblank; - unsigned int dst_y_per_vm_flip; - unsigned int dst_y_per_row_flip; - unsigned int ref_freq_to_pix_freq; - unsigned int vratio_prefetch; - unsigned int vratio_prefetch_c; - unsigned int refcyc_per_pte_group_vblank_l; - unsigned int refcyc_per_pte_group_vblank_c; - unsigned int refcyc_per_meta_chunk_vblank_l; - unsigned int refcyc_per_meta_chunk_vblank_c; - unsigned int refcyc_per_pte_group_flip_l; - unsigned int refcyc_per_pte_group_flip_c; - unsigned int refcyc_per_meta_chunk_flip_l; - unsigned int refcyc_per_meta_chunk_flip_c; - unsigned int dst_y_per_pte_row_nom_l; - unsigned int dst_y_per_pte_row_nom_c; - unsigned int refcyc_per_pte_group_nom_l; - unsigned int refcyc_per_pte_group_nom_c; - unsigned int dst_y_per_meta_row_nom_l; - unsigned int dst_y_per_meta_row_nom_c; - unsigned int refcyc_per_meta_chunk_nom_l; - unsigned int refcyc_per_meta_chunk_nom_c; - unsigned int refcyc_per_line_delivery_pre_l; - unsigned int refcyc_per_line_delivery_pre_c; - unsigned int refcyc_per_line_delivery_l; - unsigned int refcyc_per_line_delivery_c; - unsigned int chunk_hdl_adjust_cur0; - unsigned int chunk_hdl_adjust_cur1; - unsigned int vready_after_vcount0; - unsigned int dst_y_offset_cur0; - unsigned int dst_y_offset_cur1; - unsigned int xfc_reg_transfer_delay; - unsigned int xfc_reg_precharge_delay; - unsigned int xfc_reg_remote_surface_flip_latency; - unsigned int xfc_reg_prefetch_margin; - unsigned int dst_y_delta_drq_limit; -}; - -struct _vcs_dpi_display_ttu_regs_st { - unsigned int qos_level_low_wm; - unsigned int qos_level_high_wm; - unsigned int min_ttu_vblank; - unsigned int qos_level_flip; - unsigned int refcyc_per_req_delivery_l; - unsigned int refcyc_per_req_delivery_c; - unsigned int refcyc_per_req_delivery_cur0; - unsigned int refcyc_per_req_delivery_cur1; - unsigned int refcyc_per_req_delivery_pre_l; - unsigned int refcyc_per_req_delivery_pre_c; - unsigned int refcyc_per_req_delivery_pre_cur0; - unsigned int refcyc_per_req_delivery_pre_cur1; - unsigned int qos_level_fixed_l; - unsigned int qos_level_fixed_c; - unsigned int qos_level_fixed_cur0; - unsigned int qos_level_fixed_cur1; - unsigned int qos_ramp_disable_l; - unsigned int qos_ramp_disable_c; - unsigned int qos_ramp_disable_cur0; - unsigned int qos_ramp_disable_cur1; -}; - -struct _vcs_dpi_display_data_rq_regs_st { - unsigned int chunk_size; - unsigned int min_chunk_size; - unsigned int meta_chunk_size; - unsigned int min_meta_chunk_size; - unsigned int dpte_group_size; - unsigned int mpte_group_size; - unsigned int swath_height; - unsigned int pte_row_height_linear; -}; - -struct _vcs_dpi_display_rq_regs_st { - display_data_rq_regs_st rq_regs_l; - display_data_rq_regs_st rq_regs_c; - unsigned int drq_expansion_mode; - unsigned int prq_expansion_mode; - unsigned int mrq_expansion_mode; - unsigned int crq_expansion_mode; - unsigned int plane1_base_address; -}; - -struct _vcs_dpi_display_dlg_sys_params_st { - double t_mclk_wm_us; - double t_urg_wm_us; - double t_sr_wm_us; - double t_extra_us; - double mem_trip_us; - double t_srx_delay_us; - double deepsleep_dcfclk_mhz; - double total_flip_bw; - unsigned int total_flip_bytes; -}; - -struct _vcs_dpi_display_dlg_prefetch_param_st { - double prefetch_bw; - unsigned int flip_bytes; -}; - -struct _vcs_dpi_display_pipe_clock_st { - double dcfclk_mhz; - double dispclk_mhz; - double socclk_mhz; - double dscclk_mhz[6]; - double dppclk_mhz[6]; -}; - -struct _vcs_dpi_display_arb_params_st { - int max_req_outstanding; - int min_req_outstanding; - int sat_level_us; +struct _vcs_dpi_display_bandwidth_st { + double total_bw_consumed_gbps; + double guaranteed_urgent_return_bw_gbps; +}; + +struct _vcs_dpi_scaler_ratio_depth_st { + double hscl_ratio; + double vscl_ratio; + double hscl_ratio_c; + double vscl_ratio_c; + double vinit; + double vinit_c; + double vinit_bot; + double vinit_bot_c; + int lb_depth; + int scl_enable; +}; + +struct _vcs_dpi_scaler_taps_st { + unsigned int htaps; + unsigned int vtaps; + unsigned int htaps_c; + unsigned int vtaps_c; +}; + +struct _vcs_dpi_display_pipe_dest_params_st { + unsigned int recout_width; + unsigned int recout_height; + unsigned int full_recout_width; + unsigned int full_recout_height; + unsigned int hblank_start; + unsigned int hblank_end; + unsigned int vblank_start; + unsigned int vblank_end; + unsigned int htotal; + unsigned int vtotal; + unsigned int vactive; + unsigned int hactive; + unsigned int vstartup_start; + unsigned int vupdate_offset; + unsigned int vupdate_width; + unsigned int vready_offset; + unsigned char interlaced; + unsigned char underscan; + double pixel_rate_mhz; + unsigned char synchronized_vblank_all_planes; + unsigned char otg_inst; + unsigned char odm_split_cnt; + unsigned char odm_combine; +}; + +struct _vcs_dpi_display_pipe_params_st { + display_pipe_source_params_st src; + display_pipe_dest_params_st dest; + scaler_ratio_depth_st scale_ratio_depth; + scaler_taps_st scale_taps; +}; + +struct _vcs_dpi_display_clocks_and_cfg_st { + int voltage; + double dppclk_mhz; + double refclk_mhz; + double dispclk_mhz; + double dcfclk_mhz; + double socclk_mhz; +}; + +struct _vcs_dpi_display_e2e_pipe_params_st { + display_pipe_params_st pipe; + display_output_params_st dout; + display_clocks_and_cfg_st clks_cfg; +}; + +struct _vcs_dpi_dchub_buffer_sizing_st { + unsigned int swath_width_y; + unsigned int swath_height_y; + unsigned int swath_height_c; + unsigned int detail_buffer_size_y; +}; + +struct _vcs_dpi_watermarks_perf_st { + double stutter_eff_in_active_region_percent; + double urgent_latency_supported_us; + double non_urgent_latency_supported_us; + double dram_clock_change_margin_us; + double dram_access_eff_percent; +}; + +struct _vcs_dpi_cstate_pstate_watermarks_st { + double cstate_exit_us; + double cstate_enter_plus_exit_us; + double pstate_change_us; +}; + +struct _vcs_dpi_wm_calc_pipe_params_st { + unsigned int num_dpp; + int voltage; + int output_type; + double dcfclk_mhz; + double socclk_mhz; + double dppclk_mhz; + double pixclk_mhz; + unsigned char interlace_en; + unsigned char pte_enable; + unsigned char dcc_enable; + double dcc_rate; + double bytes_per_pixel_c; + double bytes_per_pixel_y; + unsigned int swath_width_y; + unsigned int swath_height_y; + unsigned int swath_height_c; + unsigned int det_buffer_size_y; + double h_ratio; + double v_ratio; + unsigned int h_taps; + unsigned int h_total; + unsigned int v_total; + unsigned int v_active; + unsigned int e2e_index; + double display_pipe_line_delivery_time; + double read_bw; + unsigned int lines_in_det_y; + unsigned int lines_in_det_y_rounded_down_to_swath; + double full_det_buffering_time; + double dcfclk_deepsleep_mhz_per_plane; +}; + +struct _vcs_dpi_vratio_pre_st { + double vratio_pre_l; + double vratio_pre_c; +}; + +struct _vcs_dpi_display_data_rq_misc_params_st { + unsigned int full_swath_bytes; + unsigned int stored_swath_bytes; + unsigned int blk256_height; + unsigned int blk256_width; + unsigned int req_height; + unsigned int req_width; +}; + +struct _vcs_dpi_display_data_rq_sizing_params_st { + unsigned int chunk_bytes; + unsigned int min_chunk_bytes; + unsigned int meta_chunk_bytes; + unsigned int min_meta_chunk_bytes; + unsigned int mpte_group_bytes; + unsigned int dpte_group_bytes; +}; + +struct _vcs_dpi_display_data_rq_dlg_params_st { + unsigned int swath_width_ub; + unsigned int swath_height; + unsigned int req_per_swath_ub; + unsigned int meta_pte_bytes_per_frame_ub; + unsigned int dpte_req_per_row_ub; + unsigned int dpte_groups_per_row_ub; + unsigned int dpte_row_height; + unsigned int dpte_bytes_per_row_ub; + unsigned int meta_chunks_per_row_ub; + unsigned int meta_req_per_row_ub; + unsigned int meta_row_height; + unsigned int meta_bytes_per_row_ub; +}; + +struct _vcs_dpi_display_cur_rq_dlg_params_st { + unsigned char enable; + unsigned int swath_height; + unsigned int req_per_line; +}; + +struct _vcs_dpi_display_rq_dlg_params_st { + display_data_rq_dlg_params_st rq_l; + display_data_rq_dlg_params_st rq_c; + display_cur_rq_dlg_params_st rq_cur0; +}; + +struct _vcs_dpi_display_rq_sizing_params_st { + display_data_rq_sizing_params_st rq_l; + display_data_rq_sizing_params_st rq_c; +}; + +struct _vcs_dpi_display_rq_misc_params_st { + display_data_rq_misc_params_st rq_l; + display_data_rq_misc_params_st rq_c; +}; + +struct _vcs_dpi_display_rq_params_st { + unsigned char yuv420; + unsigned char yuv420_10bpc; + display_rq_misc_params_st misc; + display_rq_sizing_params_st sizing; + display_rq_dlg_params_st dlg; +}; + +struct _vcs_dpi_display_dlg_regs_st { + unsigned int refcyc_h_blank_end; + unsigned int dlg_vblank_end; + unsigned int min_dst_y_next_start; + unsigned int refcyc_per_htotal; + unsigned int refcyc_x_after_scaler; + unsigned int dst_y_after_scaler; + unsigned int dst_y_prefetch; + unsigned int dst_y_per_vm_vblank; + unsigned int dst_y_per_row_vblank; + unsigned int dst_y_per_vm_flip; + unsigned int dst_y_per_row_flip; + unsigned int ref_freq_to_pix_freq; + unsigned int vratio_prefetch; + unsigned int vratio_prefetch_c; + unsigned int refcyc_per_pte_group_vblank_l; + unsigned int refcyc_per_pte_group_vblank_c; + unsigned int refcyc_per_meta_chunk_vblank_l; + unsigned int refcyc_per_meta_chunk_vblank_c; + unsigned int refcyc_per_pte_group_flip_l; + unsigned int refcyc_per_pte_group_flip_c; + unsigned int refcyc_per_meta_chunk_flip_l; + unsigned int refcyc_per_meta_chunk_flip_c; + unsigned int dst_y_per_pte_row_nom_l; + unsigned int dst_y_per_pte_row_nom_c; + unsigned int refcyc_per_pte_group_nom_l; + unsigned int refcyc_per_pte_group_nom_c; + unsigned int dst_y_per_meta_row_nom_l; + unsigned int dst_y_per_meta_row_nom_c; + unsigned int refcyc_per_meta_chunk_nom_l; + unsigned int refcyc_per_meta_chunk_nom_c; + unsigned int refcyc_per_line_delivery_pre_l; + unsigned int refcyc_per_line_delivery_pre_c; + unsigned int refcyc_per_line_delivery_l; + unsigned int refcyc_per_line_delivery_c; + unsigned int chunk_hdl_adjust_cur0; + unsigned int chunk_hdl_adjust_cur1; + unsigned int vready_after_vcount0; + unsigned int dst_y_offset_cur0; + unsigned int dst_y_offset_cur1; + unsigned int xfc_reg_transfer_delay; + unsigned int xfc_reg_precharge_delay; + unsigned int xfc_reg_remote_surface_flip_latency; + unsigned int xfc_reg_prefetch_margin; + unsigned int dst_y_delta_drq_limit; +}; + +struct _vcs_dpi_display_ttu_regs_st { + unsigned int qos_level_low_wm; + unsigned int qos_level_high_wm; + unsigned int min_ttu_vblank; + unsigned int qos_level_flip; + unsigned int refcyc_per_req_delivery_l; + unsigned int refcyc_per_req_delivery_c; + unsigned int refcyc_per_req_delivery_cur0; + unsigned int refcyc_per_req_delivery_cur1; + unsigned int refcyc_per_req_delivery_pre_l; + unsigned int refcyc_per_req_delivery_pre_c; + unsigned int refcyc_per_req_delivery_pre_cur0; + unsigned int refcyc_per_req_delivery_pre_cur1; + unsigned int qos_level_fixed_l; + unsigned int qos_level_fixed_c; + unsigned int qos_level_fixed_cur0; + unsigned int qos_level_fixed_cur1; + unsigned int qos_ramp_disable_l; + unsigned int qos_ramp_disable_c; + unsigned int qos_ramp_disable_cur0; + unsigned int qos_ramp_disable_cur1; +}; + +struct _vcs_dpi_display_data_rq_regs_st { + unsigned int chunk_size; + unsigned int min_chunk_size; + unsigned int meta_chunk_size; + unsigned int min_meta_chunk_size; + unsigned int dpte_group_size; + unsigned int mpte_group_size; + unsigned int swath_height; + unsigned int pte_row_height_linear; +}; + +struct _vcs_dpi_display_rq_regs_st { + display_data_rq_regs_st rq_regs_l; + display_data_rq_regs_st rq_regs_c; + unsigned int drq_expansion_mode; + unsigned int prq_expansion_mode; + unsigned int mrq_expansion_mode; + unsigned int crq_expansion_mode; + unsigned int plane1_base_address; +}; + +struct _vcs_dpi_display_dlg_sys_params_st { + double t_mclk_wm_us; + double t_urg_wm_us; + double t_sr_wm_us; + double t_extra_us; + double mem_trip_us; + double t_srx_delay_us; + double deepsleep_dcfclk_mhz; + double total_flip_bw; + unsigned int total_flip_bytes; +}; + +struct _vcs_dpi_display_dlg_prefetch_param_st { + double prefetch_bw; + unsigned int flip_bytes; +}; + +struct _vcs_dpi_display_pipe_clock_st { + double dcfclk_mhz; + double dispclk_mhz; + double socclk_mhz; + double dscclk_mhz[6]; + double dppclk_mhz[6]; +}; + +struct _vcs_dpi_display_arb_params_st { + int max_req_outstanding; + int min_req_outstanding; + int sat_level_us; }; #endif /*__DISPLAY_MODE_STRUCTS_H__*/ diff --git a/drivers/gpu/drm/amd/display/dc/dml/dml_inline_defs.h b/drivers/gpu/drm/amd/display/dc/dml/dml_inline_defs.h index f9cf08357989..e8ce08567cd8 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dml_inline_defs.h +++ b/drivers/gpu/drm/amd/display/dc/dml/dml_inline_defs.h @@ -35,6 +35,16 @@ static inline double dml_min(double a, double b) return (double) dcn_bw_min2(a, b); } +static inline double dml_min3(double a, double b, double c) +{ + return dml_min(dml_min(a, b), c); +} + +static inline double dml_min4(double a, double b, double c, double d) +{ + return dml_min(dml_min(a, b), dml_min(c, d)); +} + static inline double dml_max(double a, double b) { return (double) dcn_bw_max2(a, b); diff --git a/drivers/gpu/drm/amd/display/dc/gpio/hw_factory.c b/drivers/gpu/drm/amd/display/dc/gpio/hw_factory.c index 61fe484da1a0..0caee3523017 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/hw_factory.c +++ b/drivers/gpu/drm/amd/display/dc/gpio/hw_factory.c @@ -75,9 +75,7 @@ bool dal_hw_factory_init( return true; case DCE_VERSION_11_0: case DCE_VERSION_11_2: -#if defined(CONFIG_DRM_AMD_DC_VEGAM) case DCE_VERSION_11_22: -#endif dal_hw_factory_dce110_init(factory); return true; case DCE_VERSION_12_0: diff --git a/drivers/gpu/drm/amd/display/dc/gpio/hw_translate.c b/drivers/gpu/drm/amd/display/dc/gpio/hw_translate.c index 910ae2b7bf64..55c707488541 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/hw_translate.c +++ b/drivers/gpu/drm/amd/display/dc/gpio/hw_translate.c @@ -72,9 +72,7 @@ bool dal_hw_translate_init( case DCE_VERSION_10_0: case DCE_VERSION_11_0: case DCE_VERSION_11_2: -#if defined(CONFIG_DRM_AMD_DC_VEGAM) case DCE_VERSION_11_22: -#endif dal_hw_translate_dce110_init(translate); return true; case DCE_VERSION_12_0: diff --git a/drivers/gpu/drm/amd/display/dc/i2caux/i2caux.c b/drivers/gpu/drm/amd/display/dc/i2caux/i2caux.c index c3d7c320fdba..14dc8c94d862 100644 --- a/drivers/gpu/drm/amd/display/dc/i2caux/i2caux.c +++ b/drivers/gpu/drm/amd/display/dc/i2caux/i2caux.c @@ -83,9 +83,7 @@ struct i2caux *dal_i2caux_create( case DCE_VERSION_8_3: return dal_i2caux_dce80_create(ctx); case DCE_VERSION_11_2: -#if defined(CONFIG_DRM_AMD_DC_VEGAM) case DCE_VERSION_11_22: -#endif return dal_i2caux_dce112_create(ctx); case DCE_VERSION_11_0: return dal_i2caux_dce110_create(ctx); diff --git a/drivers/gpu/drm/amd/display/dc/inc/dce_calcs.h b/drivers/gpu/drm/amd/display/dc/inc/dce_calcs.h index 933ea7a1e18b..eece165206f9 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/dce_calcs.h +++ b/drivers/gpu/drm/amd/display/dc/inc/dce_calcs.h @@ -43,9 +43,7 @@ enum bw_calcs_version { BW_CALCS_VERSION_POLARIS10, BW_CALCS_VERSION_POLARIS11, BW_CALCS_VERSION_POLARIS12, -#if defined(CONFIG_DRM_AMD_DC_VEGAM) BW_CALCS_VERSION_VEGAM, -#endif BW_CALCS_VERSION_STONEY, BW_CALCS_VERSION_VEGA10 }; diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h b/drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h index 331f8ff57ed7..97df82cddf82 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h @@ -121,6 +121,8 @@ struct hubp_funcs { void (*hubp_clk_cntl)(struct hubp *hubp, bool enable); void (*hubp_vtg_sel)(struct hubp *hubp, uint32_t otg_inst); void (*hubp_read_state)(struct hubp *hubp); + void (*hubp_disable_control)(struct hubp *hubp, bool disable_hubp); + unsigned int (*hubp_get_underflow_status)(struct hubp *hubp); }; diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h b/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h index 29abf3ecb39c..63fc6c499789 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h @@ -83,11 +83,6 @@ struct hw_sequencer_funcs { int num_planes, struct dc_state *context); - void (*set_plane_config)( - const struct dc *dc, - struct pipe_ctx *pipe_ctx, - struct resource_context *res_ctx); - void (*program_gamut_remap)( struct pipe_ctx *pipe_ctx); diff --git a/drivers/gpu/drm/amd/display/dc/irq_types.h b/drivers/gpu/drm/amd/display/dc/irq_types.h index cc3b1bc6cedd..0b5f3a278c22 100644 --- a/drivers/gpu/drm/amd/display/dc/irq_types.h +++ b/drivers/gpu/drm/amd/display/dc/irq_types.h @@ -26,6 +26,8 @@ #ifndef __DAL_IRQ_TYPES_H__ #define __DAL_IRQ_TYPES_H__ +#include "os_types.h" + struct dc_context; typedef void (*interrupt_handler)(void *); diff --git a/drivers/gpu/drm/amd/display/include/dal_asic_id.h b/drivers/gpu/drm/amd/display/include/dal_asic_id.h index 1b987b6a347d..25029ed42d89 100644 --- a/drivers/gpu/drm/amd/display/include/dal_asic_id.h +++ b/drivers/gpu/drm/amd/display/include/dal_asic_id.h @@ -86,6 +86,7 @@ #define VI_POLARIS10_P_A0 80 #define VI_POLARIS11_M_A0 90 #define VI_POLARIS12_V_A0 100 +#define VI_VEGAM_A0 110 #define VI_UNKNOWN 0xFF @@ -98,14 +99,9 @@ (eChipRev < VI_POLARIS11_M_A0)) #define ASIC_REV_IS_POLARIS11_M(eChipRev) ((eChipRev >= VI_POLARIS11_M_A0) && \ (eChipRev < VI_POLARIS12_V_A0)) -#if defined(CONFIG_DRM_AMD_DC_VEGAM) -#define VI_VEGAM_A0 110 #define ASIC_REV_IS_POLARIS12_V(eChipRev) ((eChipRev >= VI_POLARIS12_V_A0) && \ (eChipRev < VI_VEGAM_A0)) #define ASIC_REV_IS_VEGAM(eChipRev) (eChipRev >= VI_VEGAM_A0) -#else -#define ASIC_REV_IS_POLARIS12_V(eChipRev) (eChipRev >= VI_POLARIS12_V_A0) -#endif /* DCE11 */ #define CZ_CARRIZO_A0 0x01 @@ -117,17 +113,19 @@ ((rev >= STONEY_A0) && (rev < CZ_UNKNOWN)) /* DCE12 */ +#define AI_UNKNOWN 0xFF #define AI_GREENLAND_P_A0 1 #define AI_GREENLAND_P_A1 2 #define AI_UNKNOWN 0xFF #define AI_VEGA12_P_A0 20 +#define AI_VEGA20_P_A0 40 #define ASICREV_IS_GREENLAND_M(eChipRev) (eChipRev < AI_VEGA12_P_A0) #define ASICREV_IS_GREENLAND_P(eChipRev) (eChipRev < AI_VEGA12_P_A0) -#define ASICREV_IS_VEGA12_P(eChipRev) ((eChipRev >= AI_VEGA12_P_A0) && (eChipRev < AI_UNKNOWN)) -#define ASICREV_IS_VEGA12_p(eChipRev) ((eChipRev >= AI_VEGA12_P_A0) && (eChipRev < AI_UNKNOWN)) +#define ASICREV_IS_VEGA12_P(eChipRev) ((eChipRev >= AI_VEGA12_P_A0) && (eChipRev < AI_VEGA20_P_A0)) +#define ASICREV_IS_VEGA20_P(eChipRev) ((eChipRev >= AI_VEGA20_P_A0) && (eChipRev < AI_UNKNOWN)) /* DCN1_0 */ #define INTERNAL_REV_RAVEN_A0 0x00 /* First spin of Raven */ diff --git a/drivers/gpu/drm/amd/display/include/dal_types.h b/drivers/gpu/drm/amd/display/include/dal_types.h index 5b1f8cef0c22..840142b65f8b 100644 --- a/drivers/gpu/drm/amd/display/include/dal_types.h +++ b/drivers/gpu/drm/amd/display/include/dal_types.h @@ -40,9 +40,7 @@ enum dce_version { DCE_VERSION_10_0, DCE_VERSION_11_0, DCE_VERSION_11_2, -#if defined(CONFIG_DRM_AMD_DC_VEGAM) DCE_VERSION_11_22, -#endif DCE_VERSION_12_0, DCE_VERSION_MAX, DCN_VERSION_1_0, diff --git a/drivers/gpu/drm/amd/display/include/fixed31_32.h b/drivers/gpu/drm/amd/display/include/fixed31_32.h index 16cbdb43d856..bb0d4ebba9f0 100644 --- a/drivers/gpu/drm/amd/display/include/fixed31_32.h +++ b/drivers/gpu/drm/amd/display/include/fixed31_32.h @@ -27,6 +27,12 @@ #define __DAL_FIXED31_32_H__ #define FIXED31_32_BITS_PER_FRACTIONAL_PART 32 +#ifndef LLONG_MIN +#define LLONG_MIN (1LL<<63) +#endif +#ifndef LLONG_MAX +#define LLONG_MAX (-1LL>>1) +#endif /* * @brief @@ -45,21 +51,22 @@ struct fixed31_32 { long long value; }; + /* * @brief * Useful constants */ -static const struct fixed31_32 dal_fixed31_32_zero = { 0 }; -static const struct fixed31_32 dal_fixed31_32_epsilon = { 1LL }; -static const struct fixed31_32 dal_fixed31_32_half = { 0x80000000LL }; -static const struct fixed31_32 dal_fixed31_32_one = { 0x100000000LL }; +static const struct fixed31_32 dc_fixpt_zero = { 0 }; +static const struct fixed31_32 dc_fixpt_epsilon = { 1LL }; +static const struct fixed31_32 dc_fixpt_half = { 0x80000000LL }; +static const struct fixed31_32 dc_fixpt_one = { 0x100000000LL }; -static const struct fixed31_32 dal_fixed31_32_pi = { 13493037705LL }; -static const struct fixed31_32 dal_fixed31_32_two_pi = { 26986075409LL }; -static const struct fixed31_32 dal_fixed31_32_e = { 11674931555LL }; -static const struct fixed31_32 dal_fixed31_32_ln2 = { 2977044471LL }; -static const struct fixed31_32 dal_fixed31_32_ln2_div_2 = { 1488522236LL }; +static const struct fixed31_32 dc_fixpt_pi = { 13493037705LL }; +static const struct fixed31_32 dc_fixpt_two_pi = { 26986075409LL }; +static const struct fixed31_32 dc_fixpt_e = { 11674931555LL }; +static const struct fixed31_32 dc_fixpt_ln2 = { 2977044471LL }; +static const struct fixed31_32 dc_fixpt_ln2_div_2 = { 1488522236LL }; /* * @brief @@ -70,24 +77,19 @@ static const struct fixed31_32 dal_fixed31_32_ln2_div_2 = { 1488522236LL }; * @brief * result = numerator / denominator */ -struct fixed31_32 dal_fixed31_32_from_fraction( - long long numerator, - long long denominator); +struct fixed31_32 dc_fixpt_from_fraction(long long numerator, long long denominator); /* * @brief * result = arg */ -struct fixed31_32 dal_fixed31_32_from_int_nonconst(long long arg); -static inline struct fixed31_32 dal_fixed31_32_from_int(long long arg) +static inline struct fixed31_32 dc_fixpt_from_int(int arg) { - if (__builtin_constant_p(arg)) { - struct fixed31_32 res; - BUILD_BUG_ON((LONG_MIN > arg) || (arg > LONG_MAX)); - res.value = arg << FIXED31_32_BITS_PER_FRACTIONAL_PART; - return res; - } else - return dal_fixed31_32_from_int_nonconst(arg); + struct fixed31_32 res; + + res.value = (long long) arg << FIXED31_32_BITS_PER_FRACTIONAL_PART; + + return res; } /* @@ -99,7 +101,7 @@ static inline struct fixed31_32 dal_fixed31_32_from_int(long long arg) * @brief * result = -arg */ -static inline struct fixed31_32 dal_fixed31_32_neg(struct fixed31_32 arg) +static inline struct fixed31_32 dc_fixpt_neg(struct fixed31_32 arg) { struct fixed31_32 res; @@ -112,10 +114,10 @@ static inline struct fixed31_32 dal_fixed31_32_neg(struct fixed31_32 arg) * @brief * result = abs(arg) := (arg >= 0) ? arg : -arg */ -static inline struct fixed31_32 dal_fixed31_32_abs(struct fixed31_32 arg) +static inline struct fixed31_32 dc_fixpt_abs(struct fixed31_32 arg) { if (arg.value < 0) - return dal_fixed31_32_neg(arg); + return dc_fixpt_neg(arg); else return arg; } @@ -129,8 +131,7 @@ static inline struct fixed31_32 dal_fixed31_32_abs(struct fixed31_32 arg) * @brief * result = arg1 < arg2 */ -static inline bool dal_fixed31_32_lt(struct fixed31_32 arg1, - struct fixed31_32 arg2) +static inline bool dc_fixpt_lt(struct fixed31_32 arg1, struct fixed31_32 arg2) { return arg1.value < arg2.value; } @@ -139,8 +140,7 @@ static inline bool dal_fixed31_32_lt(struct fixed31_32 arg1, * @brief * result = arg1 <= arg2 */ -static inline bool dal_fixed31_32_le(struct fixed31_32 arg1, - struct fixed31_32 arg2) +static inline bool dc_fixpt_le(struct fixed31_32 arg1, struct fixed31_32 arg2) { return arg1.value <= arg2.value; } @@ -149,8 +149,7 @@ static inline bool dal_fixed31_32_le(struct fixed31_32 arg1, * @brief * result = arg1 == arg2 */ -static inline bool dal_fixed31_32_eq(struct fixed31_32 arg1, - struct fixed31_32 arg2) +static inline bool dc_fixpt_eq(struct fixed31_32 arg1, struct fixed31_32 arg2) { return arg1.value == arg2.value; } @@ -159,8 +158,7 @@ static inline bool dal_fixed31_32_eq(struct fixed31_32 arg1, * @brief * result = min(arg1, arg2) := (arg1 <= arg2) ? arg1 : arg2 */ -static inline struct fixed31_32 dal_fixed31_32_min(struct fixed31_32 arg1, - struct fixed31_32 arg2) +static inline struct fixed31_32 dc_fixpt_min(struct fixed31_32 arg1, struct fixed31_32 arg2) { if (arg1.value <= arg2.value) return arg1; @@ -172,8 +170,7 @@ static inline struct fixed31_32 dal_fixed31_32_min(struct fixed31_32 arg1, * @brief * result = max(arg1, arg2) := (arg1 <= arg2) ? arg2 : arg1 */ -static inline struct fixed31_32 dal_fixed31_32_max(struct fixed31_32 arg1, - struct fixed31_32 arg2) +static inline struct fixed31_32 dc_fixpt_max(struct fixed31_32 arg1, struct fixed31_32 arg2) { if (arg1.value <= arg2.value) return arg2; @@ -187,14 +184,14 @@ static inline struct fixed31_32 dal_fixed31_32_max(struct fixed31_32 arg1, * result = | arg, when min_value < arg < max_value * | max_value, when arg >= max_value */ -static inline struct fixed31_32 dal_fixed31_32_clamp( +static inline struct fixed31_32 dc_fixpt_clamp( struct fixed31_32 arg, struct fixed31_32 min_value, struct fixed31_32 max_value) { - if (dal_fixed31_32_le(arg, min_value)) + if (dc_fixpt_le(arg, min_value)) return min_value; - else if (dal_fixed31_32_le(max_value, arg)) + else if (dc_fixpt_le(max_value, arg)) return max_value; else return arg; @@ -209,21 +206,30 @@ static inline struct fixed31_32 dal_fixed31_32_clamp( * @brief * result = arg << shift */ -struct fixed31_32 dal_fixed31_32_shl( - struct fixed31_32 arg, - unsigned char shift); +static inline struct fixed31_32 dc_fixpt_shl(struct fixed31_32 arg, unsigned char shift) +{ + ASSERT(((arg.value >= 0) && (arg.value <= LLONG_MAX >> shift)) || + ((arg.value < 0) && (arg.value >= ~(LLONG_MAX >> shift)))); + + arg.value = arg.value << shift; + + return arg; +} /* * @brief * result = arg >> shift */ -static inline struct fixed31_32 dal_fixed31_32_shr( - struct fixed31_32 arg, - unsigned char shift) +static inline struct fixed31_32 dc_fixpt_shr(struct fixed31_32 arg, unsigned char shift) { - struct fixed31_32 res; - res.value = arg.value >> shift; - return res; + bool negative = arg.value < 0; + + if (negative) + arg.value = -arg.value; + arg.value = arg.value >> shift; + if (negative) + arg.value = -arg.value; + return arg; } /* @@ -235,38 +241,50 @@ static inline struct fixed31_32 dal_fixed31_32_shr( * @brief * result = arg1 + arg2 */ -struct fixed31_32 dal_fixed31_32_add( - struct fixed31_32 arg1, - struct fixed31_32 arg2); +static inline struct fixed31_32 dc_fixpt_add(struct fixed31_32 arg1, struct fixed31_32 arg2) +{ + struct fixed31_32 res; + + ASSERT(((arg1.value >= 0) && (LLONG_MAX - arg1.value >= arg2.value)) || + ((arg1.value < 0) && (LLONG_MIN - arg1.value <= arg2.value))); + + res.value = arg1.value + arg2.value; + + return res; +} /* * @brief * result = arg1 + arg2 */ -static inline struct fixed31_32 dal_fixed31_32_add_int(struct fixed31_32 arg1, - int arg2) +static inline struct fixed31_32 dc_fixpt_add_int(struct fixed31_32 arg1, int arg2) { - return dal_fixed31_32_add(arg1, - dal_fixed31_32_from_int(arg2)); + return dc_fixpt_add(arg1, dc_fixpt_from_int(arg2)); } /* * @brief * result = arg1 - arg2 */ -struct fixed31_32 dal_fixed31_32_sub( - struct fixed31_32 arg1, - struct fixed31_32 arg2); +static inline struct fixed31_32 dc_fixpt_sub(struct fixed31_32 arg1, struct fixed31_32 arg2) +{ + struct fixed31_32 res; + + ASSERT(((arg2.value >= 0) && (LLONG_MIN + arg2.value <= arg1.value)) || + ((arg2.value < 0) && (LLONG_MAX + arg2.value >= arg1.value))); + + res.value = arg1.value - arg2.value; + + return res; +} /* * @brief * result = arg1 - arg2 */ -static inline struct fixed31_32 dal_fixed31_32_sub_int(struct fixed31_32 arg1, - int arg2) +static inline struct fixed31_32 dc_fixpt_sub_int(struct fixed31_32 arg1, int arg2) { - return dal_fixed31_32_sub(arg1, - dal_fixed31_32_from_int(arg2)); + return dc_fixpt_sub(arg1, dc_fixpt_from_int(arg2)); } @@ -279,49 +297,40 @@ static inline struct fixed31_32 dal_fixed31_32_sub_int(struct fixed31_32 arg1, * @brief * result = arg1 * arg2 */ -struct fixed31_32 dal_fixed31_32_mul( - struct fixed31_32 arg1, - struct fixed31_32 arg2); +struct fixed31_32 dc_fixpt_mul(struct fixed31_32 arg1, struct fixed31_32 arg2); /* * @brief * result = arg1 * arg2 */ -static inline struct fixed31_32 dal_fixed31_32_mul_int(struct fixed31_32 arg1, - int arg2) +static inline struct fixed31_32 dc_fixpt_mul_int(struct fixed31_32 arg1, int arg2) { - return dal_fixed31_32_mul(arg1, - dal_fixed31_32_from_int(arg2)); + return dc_fixpt_mul(arg1, dc_fixpt_from_int(arg2)); } /* * @brief * result = square(arg) := arg * arg */ -struct fixed31_32 dal_fixed31_32_sqr( - struct fixed31_32 arg); +struct fixed31_32 dc_fixpt_sqr(struct fixed31_32 arg); /* * @brief * result = arg1 / arg2 */ -static inline struct fixed31_32 dal_fixed31_32_div_int(struct fixed31_32 arg1, - long long arg2) +static inline struct fixed31_32 dc_fixpt_div_int(struct fixed31_32 arg1, long long arg2) { - return dal_fixed31_32_from_fraction(arg1.value, - dal_fixed31_32_from_int(arg2).value); + return dc_fixpt_from_fraction(arg1.value, dc_fixpt_from_int(arg2).value); } /* * @brief * result = arg1 / arg2 */ -static inline struct fixed31_32 dal_fixed31_32_div(struct fixed31_32 arg1, - struct fixed31_32 arg2) +static inline struct fixed31_32 dc_fixpt_div(struct fixed31_32 arg1, struct fixed31_32 arg2) { - return dal_fixed31_32_from_fraction(arg1.value, - arg2.value); + return dc_fixpt_from_fraction(arg1.value, arg2.value); } /* @@ -336,8 +345,7 @@ static inline struct fixed31_32 dal_fixed31_32_div(struct fixed31_32 arg1, * @note * No special actions taken in case argument is zero. */ -struct fixed31_32 dal_fixed31_32_recip( - struct fixed31_32 arg); +struct fixed31_32 dc_fixpt_recip(struct fixed31_32 arg); /* * @brief @@ -352,8 +360,7 @@ struct fixed31_32 dal_fixed31_32_recip( * Argument specified in radians, * internally it's normalized to [-2pi...2pi] range. */ -struct fixed31_32 dal_fixed31_32_sinc( - struct fixed31_32 arg); +struct fixed31_32 dc_fixpt_sinc(struct fixed31_32 arg); /* * @brief @@ -363,8 +370,7 @@ struct fixed31_32 dal_fixed31_32_sinc( * Argument specified in radians, * internally it's normalized to [-2pi...2pi] range. */ -struct fixed31_32 dal_fixed31_32_sin( - struct fixed31_32 arg); +struct fixed31_32 dc_fixpt_sin(struct fixed31_32 arg); /* * @brief @@ -376,8 +382,7 @@ struct fixed31_32 dal_fixed31_32_sin( * passing arguments outside that range * will cause incorrect result! */ -struct fixed31_32 dal_fixed31_32_cos( - struct fixed31_32 arg); +struct fixed31_32 dc_fixpt_cos(struct fixed31_32 arg); /* * @brief @@ -391,8 +396,7 @@ struct fixed31_32 dal_fixed31_32_cos( * @note * Currently, function is verified for abs(arg) <= 1. */ -struct fixed31_32 dal_fixed31_32_exp( - struct fixed31_32 arg); +struct fixed31_32 dc_fixpt_exp(struct fixed31_32 arg); /* * @brief @@ -404,8 +408,7 @@ struct fixed31_32 dal_fixed31_32_exp( * Currently, no special actions taken * in case of invalid argument(s). Take care! */ -struct fixed31_32 dal_fixed31_32_log( - struct fixed31_32 arg); +struct fixed31_32 dc_fixpt_log(struct fixed31_32 arg); /* * @brief @@ -419,9 +422,13 @@ struct fixed31_32 dal_fixed31_32_log( * @note * Currently, abs(arg1) should be less than 1. Take care! */ -struct fixed31_32 dal_fixed31_32_pow( - struct fixed31_32 arg1, - struct fixed31_32 arg2); +static inline struct fixed31_32 dc_fixpt_pow(struct fixed31_32 arg1, struct fixed31_32 arg2) +{ + return dc_fixpt_exp( + dc_fixpt_mul( + dc_fixpt_log(arg1), + arg2)); +} /* * @brief @@ -432,22 +439,56 @@ struct fixed31_32 dal_fixed31_32_pow( * @brief * result = floor(arg) := greatest integer lower than or equal to arg */ -int dal_fixed31_32_floor( - struct fixed31_32 arg); +static inline int dc_fixpt_floor(struct fixed31_32 arg) +{ + unsigned long long arg_value = arg.value > 0 ? arg.value : -arg.value; + + if (arg.value >= 0) + return (int)(arg_value >> FIXED31_32_BITS_PER_FRACTIONAL_PART); + else + return -(int)(arg_value >> FIXED31_32_BITS_PER_FRACTIONAL_PART); +} /* * @brief * result = round(arg) := integer nearest to arg */ -int dal_fixed31_32_round( - struct fixed31_32 arg); +static inline int dc_fixpt_round(struct fixed31_32 arg) +{ + unsigned long long arg_value = arg.value > 0 ? arg.value : -arg.value; + + const long long summand = dc_fixpt_half.value; + + ASSERT(LLONG_MAX - (long long)arg_value >= summand); + + arg_value += summand; + + if (arg.value >= 0) + return (int)(arg_value >> FIXED31_32_BITS_PER_FRACTIONAL_PART); + else + return -(int)(arg_value >> FIXED31_32_BITS_PER_FRACTIONAL_PART); +} /* * @brief * result = ceil(arg) := lowest integer greater than or equal to arg */ -int dal_fixed31_32_ceil( - struct fixed31_32 arg); +static inline int dc_fixpt_ceil(struct fixed31_32 arg) +{ + unsigned long long arg_value = arg.value > 0 ? arg.value : -arg.value; + + const long long summand = dc_fixpt_one.value - + dc_fixpt_epsilon.value; + + ASSERT(LLONG_MAX - (long long)arg_value >= summand); + + arg_value += summand; + + if (arg.value >= 0) + return (int)(arg_value >> FIXED31_32_BITS_PER_FRACTIONAL_PART); + else + return -(int)(arg_value >> FIXED31_32_BITS_PER_FRACTIONAL_PART); +} /* the following two function are used in scaler hw programming to convert fixed * point value to format 2 bits from integer part and 19 bits from fractional @@ -455,20 +496,31 @@ int dal_fixed31_32_ceil( * fractional */ -unsigned int dal_fixed31_32_u2d19( - struct fixed31_32 arg); +unsigned int dc_fixpt_u2d19(struct fixed31_32 arg); + +unsigned int dc_fixpt_u0d19(struct fixed31_32 arg); -unsigned int dal_fixed31_32_u0d19( - struct fixed31_32 arg); +unsigned int dc_fixpt_clamp_u0d14(struct fixed31_32 arg); +unsigned int dc_fixpt_clamp_u0d10(struct fixed31_32 arg); -unsigned int dal_fixed31_32_clamp_u0d14( - struct fixed31_32 arg); +int dc_fixpt_s4d19(struct fixed31_32 arg); -unsigned int dal_fixed31_32_clamp_u0d10( - struct fixed31_32 arg); +static inline struct fixed31_32 dc_fixpt_truncate(struct fixed31_32 arg, unsigned int frac_bits) +{ + bool negative = arg.value < 0; -int dal_fixed31_32_s4d19( - struct fixed31_32 arg); + if (frac_bits >= FIXED31_32_BITS_PER_FRACTIONAL_PART) { + ASSERT(frac_bits == FIXED31_32_BITS_PER_FRACTIONAL_PART); + return arg; + } + + if (negative) + arg.value = -arg.value; + arg.value &= (~0LL) << (FIXED31_32_BITS_PER_FRACTIONAL_PART - frac_bits); + if (negative) + arg.value = -arg.value; + return arg; +} #endif diff --git a/drivers/gpu/drm/amd/display/include/fixed32_32.h b/drivers/gpu/drm/amd/display/include/fixed32_32.h deleted file mode 100644 index 9c70341fe026..000000000000 --- a/drivers/gpu/drm/amd/display/include/fixed32_32.h +++ /dev/null @@ -1,129 +0,0 @@ -/* - * Copyright 2012-15 Advanced Micro Devices, Inc. - * - * Permission is hereby granted, free of charge, to any person obtaining a - * copy of this software and associated documentation files (the "Software"), - * to deal in the Software without restriction, including without limitation - * the rights to use, copy, modify, merge, publish, distribute, sublicense, - * and/or sell copies of the Software, and to permit persons to whom the - * Software is furnished to do so, subject to the following conditions: - * - * The above copyright notice and this permission notice shall be included in - * all copies or substantial portions of the Software. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL - * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR - * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, - * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR - * OTHER DEALINGS IN THE SOFTWARE. - * - * Authors: AMD - * - */ - - -#ifndef __DAL_FIXED32_32_H__ -#define __DAL_FIXED32_32_H__ - -#include "os_types.h" - -struct fixed32_32 { - uint64_t value; -}; - -static const struct fixed32_32 dal_fixed32_32_zero = { 0 }; -static const struct fixed32_32 dal_fixed32_32_one = { 0x100000000LL }; -static const struct fixed32_32 dal_fixed32_32_half = { 0x80000000LL }; - -struct fixed32_32 dal_fixed32_32_from_fraction(uint32_t n, uint32_t d); -static inline struct fixed32_32 dal_fixed32_32_from_int(uint32_t value) -{ - struct fixed32_32 fx; - - fx.value = (uint64_t)value<<32; - return fx; -} - -struct fixed32_32 dal_fixed32_32_add( - struct fixed32_32 lhs, - struct fixed32_32 rhs); -struct fixed32_32 dal_fixed32_32_add_int( - struct fixed32_32 lhs, - uint32_t rhs); -struct fixed32_32 dal_fixed32_32_sub( - struct fixed32_32 lhs, - struct fixed32_32 rhs); -struct fixed32_32 dal_fixed32_32_sub_int( - struct fixed32_32 lhs, - uint32_t rhs); -struct fixed32_32 dal_fixed32_32_mul( - struct fixed32_32 lhs, - struct fixed32_32 rhs); -struct fixed32_32 dal_fixed32_32_mul_int( - struct fixed32_32 lhs, - uint32_t rhs); -struct fixed32_32 dal_fixed32_32_div( - struct fixed32_32 lhs, - struct fixed32_32 rhs); -struct fixed32_32 dal_fixed32_32_div_int( - struct fixed32_32 lhs, - uint32_t rhs); - -static inline struct fixed32_32 dal_fixed32_32_min(struct fixed32_32 lhs, - struct fixed32_32 rhs) -{ - return (lhs.value < rhs.value) ? lhs : rhs; -} - -static inline struct fixed32_32 dal_fixed32_32_max(struct fixed32_32 lhs, - struct fixed32_32 rhs) -{ - return (lhs.value > rhs.value) ? lhs : rhs; -} - -static inline bool dal_fixed32_32_gt(struct fixed32_32 lhs, struct fixed32_32 rhs) -{ - return lhs.value > rhs.value; -} - -static inline bool dal_fixed32_32_gt_int(struct fixed32_32 lhs, uint32_t rhs) -{ - return lhs.value > ((uint64_t)rhs<<32); -} - -static inline bool dal_fixed32_32_lt(struct fixed32_32 lhs, struct fixed32_32 rhs) -{ - return lhs.value < rhs.value; -} - -static inline bool dal_fixed32_32_lt_int(struct fixed32_32 lhs, uint32_t rhs) -{ - return lhs.value < ((uint64_t)rhs<<32); -} - -static inline bool dal_fixed32_32_le(struct fixed32_32 lhs, struct fixed32_32 rhs) -{ - return lhs.value <= rhs.value; -} - -static inline bool dal_fixed32_32_le_int(struct fixed32_32 lhs, uint32_t rhs) -{ - return lhs.value <= ((uint64_t)rhs<<32); -} - -static inline bool dal_fixed32_32_eq(struct fixed32_32 lhs, struct fixed32_32 rhs) -{ - return lhs.value == rhs.value; -} - -uint32_t dal_fixed32_32_ceil(struct fixed32_32 value); -static inline uint32_t dal_fixed32_32_floor(struct fixed32_32 value) -{ - return value.value>>32; -} - -uint32_t dal_fixed32_32_round(struct fixed32_32 value); - -#endif diff --git a/drivers/gpu/drm/amd/display/include/logger_interface.h b/drivers/gpu/drm/amd/display/include/logger_interface.h index 28dee960d509..dc98d6d4b2bd 100644 --- a/drivers/gpu/drm/amd/display/include/logger_interface.h +++ b/drivers/gpu/drm/amd/display/include/logger_interface.h @@ -190,4 +190,13 @@ void context_clock_trace( } \ } while (0) +#define DISPLAY_STATS_BEGIN(entry) \ + dm_logger_open(dc->ctx->logger, &entry, LOG_DISPLAYSTATS) + +#define DISPLAY_STATS(msg, ...) \ + dm_logger_append(&log_entry, msg, ##__VA_ARGS__) + +#define DISPLAY_STATS_END(entry) \ + dm_logger_close(&entry) + #endif /* __DAL_LOGGER_INTERFACE_H__ */ diff --git a/drivers/gpu/drm/amd/display/include/logger_types.h b/drivers/gpu/drm/amd/display/include/logger_types.h index b608a0830801..0a540b9897a6 100644 --- a/drivers/gpu/drm/amd/display/include/logger_types.h +++ b/drivers/gpu/drm/amd/display/include/logger_types.h @@ -98,7 +98,7 @@ enum dc_log_type { LOG_EVENT_UNDERFLOW, LOG_IF_TRACE, LOG_PERF_TRACE, - LOG_PROFILING, + LOG_DISPLAYSTATS, LOG_SECTION_TOTAL_COUNT }; diff --git a/drivers/gpu/drm/amd/display/modules/color/color_gamma.c b/drivers/gpu/drm/amd/display/modules/color/color_gamma.c index 15e5b72e6e00..0cd111d59018 100644 --- a/drivers/gpu/drm/amd/display/modules/color/color_gamma.c +++ b/drivers/gpu/drm/amd/display/modules/color/color_gamma.c @@ -43,7 +43,7 @@ static bool de_pq_initialized; /* = false; */ /* one-time setup of X points */ void setup_x_points_distribution(void) { - struct fixed31_32 region_size = dal_fixed31_32_from_int(128); + struct fixed31_32 region_size = dc_fixpt_from_int(128); int32_t segment; uint32_t seg_offset; uint32_t index; @@ -53,8 +53,8 @@ void setup_x_points_distribution(void) coordinates_x[MAX_HW_POINTS + 1].x = region_size; for (segment = 6; segment > (6 - NUM_REGIONS); segment--) { - region_size = dal_fixed31_32_div_int(region_size, 2); - increment = dal_fixed31_32_div_int(region_size, + region_size = dc_fixpt_div_int(region_size, 2); + increment = dc_fixpt_div_int(region_size, NUM_PTS_IN_REGION); seg_offset = (segment + (NUM_REGIONS - 7)) * NUM_PTS_IN_REGION; coordinates_x[seg_offset].x = region_size; @@ -62,7 +62,7 @@ void setup_x_points_distribution(void) for (index = seg_offset + 1; index < seg_offset + NUM_PTS_IN_REGION; index++) { - coordinates_x[index].x = dal_fixed31_32_add + coordinates_x[index].x = dc_fixpt_add (coordinates_x[index-1].x, increment); } } @@ -72,63 +72,63 @@ static void compute_pq(struct fixed31_32 in_x, struct fixed31_32 *out_y) { /* consts for PQ gamma formula. */ const struct fixed31_32 m1 = - dal_fixed31_32_from_fraction(159301758, 1000000000); + dc_fixpt_from_fraction(159301758, 1000000000); const struct fixed31_32 m2 = - dal_fixed31_32_from_fraction(7884375, 100000); + dc_fixpt_from_fraction(7884375, 100000); const struct fixed31_32 c1 = - dal_fixed31_32_from_fraction(8359375, 10000000); + dc_fixpt_from_fraction(8359375, 10000000); const struct fixed31_32 c2 = - dal_fixed31_32_from_fraction(188515625, 10000000); + dc_fixpt_from_fraction(188515625, 10000000); const struct fixed31_32 c3 = - dal_fixed31_32_from_fraction(186875, 10000); + dc_fixpt_from_fraction(186875, 10000); struct fixed31_32 l_pow_m1; struct fixed31_32 base; - if (dal_fixed31_32_lt(in_x, dal_fixed31_32_zero)) - in_x = dal_fixed31_32_zero; + if (dc_fixpt_lt(in_x, dc_fixpt_zero)) + in_x = dc_fixpt_zero; - l_pow_m1 = dal_fixed31_32_pow(in_x, m1); - base = dal_fixed31_32_div( - dal_fixed31_32_add(c1, - (dal_fixed31_32_mul(c2, l_pow_m1))), - dal_fixed31_32_add(dal_fixed31_32_one, - (dal_fixed31_32_mul(c3, l_pow_m1)))); - *out_y = dal_fixed31_32_pow(base, m2); + l_pow_m1 = dc_fixpt_pow(in_x, m1); + base = dc_fixpt_div( + dc_fixpt_add(c1, + (dc_fixpt_mul(c2, l_pow_m1))), + dc_fixpt_add(dc_fixpt_one, + (dc_fixpt_mul(c3, l_pow_m1)))); + *out_y = dc_fixpt_pow(base, m2); } static void compute_de_pq(struct fixed31_32 in_x, struct fixed31_32 *out_y) { /* consts for dePQ gamma formula. */ const struct fixed31_32 m1 = - dal_fixed31_32_from_fraction(159301758, 1000000000); + dc_fixpt_from_fraction(159301758, 1000000000); const struct fixed31_32 m2 = - dal_fixed31_32_from_fraction(7884375, 100000); + dc_fixpt_from_fraction(7884375, 100000); const struct fixed31_32 c1 = - dal_fixed31_32_from_fraction(8359375, 10000000); + dc_fixpt_from_fraction(8359375, 10000000); const struct fixed31_32 c2 = - dal_fixed31_32_from_fraction(188515625, 10000000); + dc_fixpt_from_fraction(188515625, 10000000); const struct fixed31_32 c3 = - dal_fixed31_32_from_fraction(186875, 10000); + dc_fixpt_from_fraction(186875, 10000); struct fixed31_32 l_pow_m1; struct fixed31_32 base, div; - if (dal_fixed31_32_lt(in_x, dal_fixed31_32_zero)) - in_x = dal_fixed31_32_zero; + if (dc_fixpt_lt(in_x, dc_fixpt_zero)) + in_x = dc_fixpt_zero; - l_pow_m1 = dal_fixed31_32_pow(in_x, - dal_fixed31_32_div(dal_fixed31_32_one, m2)); - base = dal_fixed31_32_sub(l_pow_m1, c1); + l_pow_m1 = dc_fixpt_pow(in_x, + dc_fixpt_div(dc_fixpt_one, m2)); + base = dc_fixpt_sub(l_pow_m1, c1); - if (dal_fixed31_32_lt(base, dal_fixed31_32_zero)) - base = dal_fixed31_32_zero; + if (dc_fixpt_lt(base, dc_fixpt_zero)) + base = dc_fixpt_zero; - div = dal_fixed31_32_sub(c2, dal_fixed31_32_mul(c3, l_pow_m1)); + div = dc_fixpt_sub(c2, dc_fixpt_mul(c3, l_pow_m1)); - *out_y = dal_fixed31_32_pow(dal_fixed31_32_div(base, div), - dal_fixed31_32_div(dal_fixed31_32_one, m1)); + *out_y = dc_fixpt_pow(dc_fixpt_div(base, div), + dc_fixpt_div(dc_fixpt_one, m1)); } /* one-time pre-compute PQ values - only for sdr_white_level 80 */ @@ -138,14 +138,14 @@ void precompute_pq(void) struct fixed31_32 x; const struct hw_x_point *coord_x = coordinates_x + 32; struct fixed31_32 scaling_factor = - dal_fixed31_32_from_fraction(80, 10000); + dc_fixpt_from_fraction(80, 10000); /* pow function has problems with arguments too small */ for (i = 0; i < 32; i++) - pq_table[i] = dal_fixed31_32_zero; + pq_table[i] = dc_fixpt_zero; for (i = 32; i <= MAX_HW_POINTS; i++) { - x = dal_fixed31_32_mul(coord_x->x, scaling_factor); + x = dc_fixpt_mul(coord_x->x, scaling_factor); compute_pq(x, &pq_table[i]); ++coord_x; } @@ -158,7 +158,7 @@ void precompute_de_pq(void) struct fixed31_32 y; uint32_t begin_index, end_index; - struct fixed31_32 scaling_factor = dal_fixed31_32_from_int(125); + struct fixed31_32 scaling_factor = dc_fixpt_from_int(125); /* X points is 2^-25 to 2^7 * De-gamma X is 2^-12 to 2^0 – we are skipping first -12-(-25) = 13 regions @@ -167,11 +167,11 @@ void precompute_de_pq(void) end_index = begin_index + 12 * NUM_PTS_IN_REGION; for (i = 0; i <= begin_index; i++) - de_pq_table[i] = dal_fixed31_32_zero; + de_pq_table[i] = dc_fixpt_zero; for (; i <= end_index; i++) { compute_de_pq(coordinates_x[i].x, &y); - de_pq_table[i] = dal_fixed31_32_mul(y, scaling_factor); + de_pq_table[i] = dc_fixpt_mul(y, scaling_factor); } for (; i <= MAX_HW_POINTS; i++) @@ -195,15 +195,15 @@ static void build_coefficients(struct gamma_coefficients *coefficients, bool is_ uint32_t index = is_2_4 == true ? 0:1; do { - coefficients->a0[i] = dal_fixed31_32_from_fraction( + coefficients->a0[i] = dc_fixpt_from_fraction( numerator01[index], 10000000); - coefficients->a1[i] = dal_fixed31_32_from_fraction( + coefficients->a1[i] = dc_fixpt_from_fraction( numerator02[index], 1000); - coefficients->a2[i] = dal_fixed31_32_from_fraction( + coefficients->a2[i] = dc_fixpt_from_fraction( numerator03[index], 1000); - coefficients->a3[i] = dal_fixed31_32_from_fraction( + coefficients->a3[i] = dc_fixpt_from_fraction( numerator04[index], 1000); - coefficients->user_gamma[i] = dal_fixed31_32_from_fraction( + coefficients->user_gamma[i] = dc_fixpt_from_fraction( numerator05[index], 1000); ++i; @@ -218,33 +218,33 @@ static struct fixed31_32 translate_from_linear_space( struct fixed31_32 a3, struct fixed31_32 gamma) { - const struct fixed31_32 one = dal_fixed31_32_from_int(1); + const struct fixed31_32 one = dc_fixpt_from_int(1); - if (dal_fixed31_32_lt(one, arg)) + if (dc_fixpt_lt(one, arg)) return one; - if (dal_fixed31_32_le(arg, dal_fixed31_32_neg(a0))) - return dal_fixed31_32_sub( + if (dc_fixpt_le(arg, dc_fixpt_neg(a0))) + return dc_fixpt_sub( a2, - dal_fixed31_32_mul( - dal_fixed31_32_add( + dc_fixpt_mul( + dc_fixpt_add( one, a3), - dal_fixed31_32_pow( - dal_fixed31_32_neg(arg), - dal_fixed31_32_recip(gamma)))); - else if (dal_fixed31_32_le(a0, arg)) - return dal_fixed31_32_sub( - dal_fixed31_32_mul( - dal_fixed31_32_add( + dc_fixpt_pow( + dc_fixpt_neg(arg), + dc_fixpt_recip(gamma)))); + else if (dc_fixpt_le(a0, arg)) + return dc_fixpt_sub( + dc_fixpt_mul( + dc_fixpt_add( one, a3), - dal_fixed31_32_pow( + dc_fixpt_pow( arg, - dal_fixed31_32_recip(gamma))), + dc_fixpt_recip(gamma))), a2); else - return dal_fixed31_32_mul( + return dc_fixpt_mul( arg, a1); } @@ -259,25 +259,25 @@ static struct fixed31_32 translate_to_linear_space( { struct fixed31_32 linear; - a0 = dal_fixed31_32_mul(a0, a1); - if (dal_fixed31_32_le(arg, dal_fixed31_32_neg(a0))) + a0 = dc_fixpt_mul(a0, a1); + if (dc_fixpt_le(arg, dc_fixpt_neg(a0))) - linear = dal_fixed31_32_neg( - dal_fixed31_32_pow( - dal_fixed31_32_div( - dal_fixed31_32_sub(a2, arg), - dal_fixed31_32_add( - dal_fixed31_32_one, a3)), gamma)); + linear = dc_fixpt_neg( + dc_fixpt_pow( + dc_fixpt_div( + dc_fixpt_sub(a2, arg), + dc_fixpt_add( + dc_fixpt_one, a3)), gamma)); - else if (dal_fixed31_32_le(dal_fixed31_32_neg(a0), arg) && - dal_fixed31_32_le(arg, a0)) - linear = dal_fixed31_32_div(arg, a1); + else if (dc_fixpt_le(dc_fixpt_neg(a0), arg) && + dc_fixpt_le(arg, a0)) + linear = dc_fixpt_div(arg, a1); else - linear = dal_fixed31_32_pow( - dal_fixed31_32_div( - dal_fixed31_32_add(a2, arg), - dal_fixed31_32_add( - dal_fixed31_32_one, a3)), gamma); + linear = dc_fixpt_pow( + dc_fixpt_div( + dc_fixpt_add(a2, arg), + dc_fixpt_add( + dc_fixpt_one, a3)), gamma); return linear; } @@ -352,8 +352,8 @@ static bool find_software_points( right = axis_x[max_number - 1].b; } - if (dal_fixed31_32_le(left, hw_point) && - dal_fixed31_32_le(hw_point, right)) { + if (dc_fixpt_le(left, hw_point) && + dc_fixpt_le(hw_point, right)) { *index_to_start = i; *index_left = i; @@ -366,7 +366,7 @@ static bool find_software_points( return true; } else if ((i == *index_to_start) && - dal_fixed31_32_le(hw_point, left)) { + dc_fixpt_le(hw_point, left)) { *index_to_start = i; *index_left = i; *index_right = i; @@ -375,7 +375,7 @@ static bool find_software_points( return true; } else if ((i == max_number - 1) && - dal_fixed31_32_le(right, hw_point)) { + dc_fixpt_le(right, hw_point)) { *index_to_start = i; *index_left = i; *index_right = i; @@ -457,17 +457,17 @@ static bool build_custom_gamma_mapping_coefficients_worker( } if (hw_pos == HW_POINT_POSITION_MIDDLE) - point->coeff = dal_fixed31_32_div( - dal_fixed31_32_sub( + point->coeff = dc_fixpt_div( + dc_fixpt_sub( coord_x, left_pos), - dal_fixed31_32_sub( + dc_fixpt_sub( right_pos, left_pos)); else if (hw_pos == HW_POINT_POSITION_LEFT) - point->coeff = dal_fixed31_32_zero; + point->coeff = dc_fixpt_zero; else if (hw_pos == HW_POINT_POSITION_RIGHT) - point->coeff = dal_fixed31_32_from_int(2); + point->coeff = dc_fixpt_from_int(2); else { BREAK_TO_DEBUGGER(); return false; @@ -502,45 +502,45 @@ static struct fixed31_32 calculate_mapped_value( if ((point->left_index < 0) || (point->left_index > max_index)) { BREAK_TO_DEBUGGER(); - return dal_fixed31_32_zero; + return dc_fixpt_zero; } if ((point->right_index < 0) || (point->right_index > max_index)) { BREAK_TO_DEBUGGER(); - return dal_fixed31_32_zero; + return dc_fixpt_zero; } if (point->pos == HW_POINT_POSITION_MIDDLE) if (channel == CHANNEL_NAME_RED) - result = dal_fixed31_32_add( - dal_fixed31_32_mul( + result = dc_fixpt_add( + dc_fixpt_mul( point->coeff, - dal_fixed31_32_sub( + dc_fixpt_sub( rgb[point->right_index].r, rgb[point->left_index].r)), rgb[point->left_index].r); else if (channel == CHANNEL_NAME_GREEN) - result = dal_fixed31_32_add( - dal_fixed31_32_mul( + result = dc_fixpt_add( + dc_fixpt_mul( point->coeff, - dal_fixed31_32_sub( + dc_fixpt_sub( rgb[point->right_index].g, rgb[point->left_index].g)), rgb[point->left_index].g); else - result = dal_fixed31_32_add( - dal_fixed31_32_mul( + result = dc_fixpt_add( + dc_fixpt_mul( point->coeff, - dal_fixed31_32_sub( + dc_fixpt_sub( rgb[point->right_index].b, rgb[point->left_index].b)), rgb[point->left_index].b); else if (point->pos == HW_POINT_POSITION_LEFT) { BREAK_TO_DEBUGGER(); - result = dal_fixed31_32_zero; + result = dc_fixpt_zero; } else { BREAK_TO_DEBUGGER(); - result = dal_fixed31_32_one; + result = dc_fixpt_one; } return result; @@ -558,7 +558,7 @@ static void build_pq(struct pwl_float_data_ex *rgb_regamma, struct fixed31_32 x; struct fixed31_32 output; struct fixed31_32 scaling_factor = - dal_fixed31_32_from_fraction(sdr_white_level, 10000); + dc_fixpt_from_fraction(sdr_white_level, 10000); if (!pq_initialized && sdr_white_level == 80) { precompute_pq(); @@ -579,15 +579,15 @@ static void build_pq(struct pwl_float_data_ex *rgb_regamma, if (sdr_white_level == 80) { output = pq_table[i]; } else { - x = dal_fixed31_32_mul(coord_x->x, scaling_factor); + x = dc_fixpt_mul(coord_x->x, scaling_factor); compute_pq(x, &output); } /* should really not happen? */ - if (dal_fixed31_32_lt(output, dal_fixed31_32_zero)) - output = dal_fixed31_32_zero; - else if (dal_fixed31_32_lt(dal_fixed31_32_one, output)) - output = dal_fixed31_32_one; + if (dc_fixpt_lt(output, dc_fixpt_zero)) + output = dc_fixpt_zero; + else if (dc_fixpt_lt(dc_fixpt_one, output)) + output = dc_fixpt_one; rgb->r = output; rgb->g = output; @@ -605,7 +605,7 @@ static void build_de_pq(struct pwl_float_data_ex *de_pq, uint32_t i; struct fixed31_32 output; - struct fixed31_32 scaling_factor = dal_fixed31_32_from_int(125); + struct fixed31_32 scaling_factor = dc_fixpt_from_int(125); if (!de_pq_initialized) { precompute_de_pq(); @@ -616,9 +616,9 @@ static void build_de_pq(struct pwl_float_data_ex *de_pq, for (i = 0; i <= hw_points_num; i++) { output = de_pq_table[i]; /* should really not happen? */ - if (dal_fixed31_32_lt(output, dal_fixed31_32_zero)) - output = dal_fixed31_32_zero; - else if (dal_fixed31_32_lt(scaling_factor, output)) + if (dc_fixpt_lt(output, dc_fixpt_zero)) + output = dc_fixpt_zero; + else if (dc_fixpt_lt(scaling_factor, output)) output = scaling_factor; de_pq[i].r = output; de_pq[i].g = output; @@ -670,9 +670,9 @@ static void build_degamma(struct pwl_float_data_ex *curve, end_index = begin_index + 12 * NUM_PTS_IN_REGION; while (i != begin_index) { - curve[i].r = dal_fixed31_32_zero; - curve[i].g = dal_fixed31_32_zero; - curve[i].b = dal_fixed31_32_zero; + curve[i].r = dc_fixpt_zero; + curve[i].g = dc_fixpt_zero; + curve[i].b = dc_fixpt_zero; i++; } @@ -684,9 +684,9 @@ static void build_degamma(struct pwl_float_data_ex *curve, i++; } while (i != hw_points_num + 1) { - curve[i].r = dal_fixed31_32_one; - curve[i].g = dal_fixed31_32_one; - curve[i].b = dal_fixed31_32_one; + curve[i].r = dc_fixpt_one; + curve[i].g = dc_fixpt_one; + curve[i].b = dc_fixpt_one; i++; } } @@ -695,8 +695,8 @@ static void scale_gamma(struct pwl_float_data *pwl_rgb, const struct dc_gamma *ramp, struct dividers dividers) { - const struct fixed31_32 max_driver = dal_fixed31_32_from_int(0xFFFF); - const struct fixed31_32 max_os = dal_fixed31_32_from_int(0xFF00); + const struct fixed31_32 max_driver = dc_fixpt_from_int(0xFFFF); + const struct fixed31_32 max_os = dc_fixpt_from_int(0xFF00); struct fixed31_32 scaler = max_os; uint32_t i; struct pwl_float_data *rgb = pwl_rgb; @@ -705,9 +705,9 @@ static void scale_gamma(struct pwl_float_data *pwl_rgb, i = 0; do { - if (dal_fixed31_32_lt(max_os, ramp->entries.red[i]) || - dal_fixed31_32_lt(max_os, ramp->entries.green[i]) || - dal_fixed31_32_lt(max_os, ramp->entries.blue[i])) { + if (dc_fixpt_lt(max_os, ramp->entries.red[i]) || + dc_fixpt_lt(max_os, ramp->entries.green[i]) || + dc_fixpt_lt(max_os, ramp->entries.blue[i])) { scaler = max_driver; break; } @@ -717,40 +717,40 @@ static void scale_gamma(struct pwl_float_data *pwl_rgb, i = 0; do { - rgb->r = dal_fixed31_32_div( + rgb->r = dc_fixpt_div( ramp->entries.red[i], scaler); - rgb->g = dal_fixed31_32_div( + rgb->g = dc_fixpt_div( ramp->entries.green[i], scaler); - rgb->b = dal_fixed31_32_div( + rgb->b = dc_fixpt_div( ramp->entries.blue[i], scaler); ++rgb; ++i; } while (i != ramp->num_entries); - rgb->r = dal_fixed31_32_mul(rgb_last->r, + rgb->r = dc_fixpt_mul(rgb_last->r, dividers.divider1); - rgb->g = dal_fixed31_32_mul(rgb_last->g, + rgb->g = dc_fixpt_mul(rgb_last->g, dividers.divider1); - rgb->b = dal_fixed31_32_mul(rgb_last->b, + rgb->b = dc_fixpt_mul(rgb_last->b, dividers.divider1); ++rgb; - rgb->r = dal_fixed31_32_mul(rgb_last->r, + rgb->r = dc_fixpt_mul(rgb_last->r, dividers.divider2); - rgb->g = dal_fixed31_32_mul(rgb_last->g, + rgb->g = dc_fixpt_mul(rgb_last->g, dividers.divider2); - rgb->b = dal_fixed31_32_mul(rgb_last->b, + rgb->b = dc_fixpt_mul(rgb_last->b, dividers.divider2); ++rgb; - rgb->r = dal_fixed31_32_mul(rgb_last->r, + rgb->r = dc_fixpt_mul(rgb_last->r, dividers.divider3); - rgb->g = dal_fixed31_32_mul(rgb_last->g, + rgb->g = dc_fixpt_mul(rgb_last->g, dividers.divider3); - rgb->b = dal_fixed31_32_mul(rgb_last->b, + rgb->b = dc_fixpt_mul(rgb_last->b, dividers.divider3); } @@ -759,62 +759,62 @@ static void scale_gamma_dx(struct pwl_float_data *pwl_rgb, struct dividers dividers) { uint32_t i; - struct fixed31_32 min = dal_fixed31_32_zero; - struct fixed31_32 max = dal_fixed31_32_one; + struct fixed31_32 min = dc_fixpt_zero; + struct fixed31_32 max = dc_fixpt_one; - struct fixed31_32 delta = dal_fixed31_32_zero; - struct fixed31_32 offset = dal_fixed31_32_zero; + struct fixed31_32 delta = dc_fixpt_zero; + struct fixed31_32 offset = dc_fixpt_zero; for (i = 0 ; i < ramp->num_entries; i++) { - if (dal_fixed31_32_lt(ramp->entries.red[i], min)) + if (dc_fixpt_lt(ramp->entries.red[i], min)) min = ramp->entries.red[i]; - if (dal_fixed31_32_lt(ramp->entries.green[i], min)) + if (dc_fixpt_lt(ramp->entries.green[i], min)) min = ramp->entries.green[i]; - if (dal_fixed31_32_lt(ramp->entries.blue[i], min)) + if (dc_fixpt_lt(ramp->entries.blue[i], min)) min = ramp->entries.blue[i]; - if (dal_fixed31_32_lt(max, ramp->entries.red[i])) + if (dc_fixpt_lt(max, ramp->entries.red[i])) max = ramp->entries.red[i]; - if (dal_fixed31_32_lt(max, ramp->entries.green[i])) + if (dc_fixpt_lt(max, ramp->entries.green[i])) max = ramp->entries.green[i]; - if (dal_fixed31_32_lt(max, ramp->entries.blue[i])) + if (dc_fixpt_lt(max, ramp->entries.blue[i])) max = ramp->entries.blue[i]; } - if (dal_fixed31_32_lt(min, dal_fixed31_32_zero)) - delta = dal_fixed31_32_neg(min); + if (dc_fixpt_lt(min, dc_fixpt_zero)) + delta = dc_fixpt_neg(min); - offset = dal_fixed31_32_add(min, max); + offset = dc_fixpt_add(min, max); for (i = 0 ; i < ramp->num_entries; i++) { - pwl_rgb[i].r = dal_fixed31_32_div( - dal_fixed31_32_add( + pwl_rgb[i].r = dc_fixpt_div( + dc_fixpt_add( ramp->entries.red[i], delta), offset); - pwl_rgb[i].g = dal_fixed31_32_div( - dal_fixed31_32_add( + pwl_rgb[i].g = dc_fixpt_div( + dc_fixpt_add( ramp->entries.green[i], delta), offset); - pwl_rgb[i].b = dal_fixed31_32_div( - dal_fixed31_32_add( + pwl_rgb[i].b = dc_fixpt_div( + dc_fixpt_add( ramp->entries.blue[i], delta), offset); } - pwl_rgb[i].r = dal_fixed31_32_sub(dal_fixed31_32_mul_int( + pwl_rgb[i].r = dc_fixpt_sub(dc_fixpt_mul_int( pwl_rgb[i-1].r, 2), pwl_rgb[i-2].r); - pwl_rgb[i].g = dal_fixed31_32_sub(dal_fixed31_32_mul_int( + pwl_rgb[i].g = dc_fixpt_sub(dc_fixpt_mul_int( pwl_rgb[i-1].g, 2), pwl_rgb[i-2].g); - pwl_rgb[i].b = dal_fixed31_32_sub(dal_fixed31_32_mul_int( + pwl_rgb[i].b = dc_fixpt_sub(dc_fixpt_mul_int( pwl_rgb[i-1].b, 2), pwl_rgb[i-2].b); ++i; - pwl_rgb[i].r = dal_fixed31_32_sub(dal_fixed31_32_mul_int( + pwl_rgb[i].r = dc_fixpt_sub(dc_fixpt_mul_int( pwl_rgb[i-1].r, 2), pwl_rgb[i-2].r); - pwl_rgb[i].g = dal_fixed31_32_sub(dal_fixed31_32_mul_int( + pwl_rgb[i].g = dc_fixpt_sub(dc_fixpt_mul_int( pwl_rgb[i-1].g, 2), pwl_rgb[i-2].g); - pwl_rgb[i].b = dal_fixed31_32_sub(dal_fixed31_32_mul_int( + pwl_rgb[i].b = dc_fixpt_sub(dc_fixpt_mul_int( pwl_rgb[i-1].b, 2), pwl_rgb[i-2].b); } @@ -846,40 +846,40 @@ static void scale_user_regamma_ramp(struct pwl_float_data *pwl_rgb, i = 0; do { - rgb->r = dal_fixed31_32_from_fraction( + rgb->r = dc_fixpt_from_fraction( ramp->gamma[i], scaler); - rgb->g = dal_fixed31_32_from_fraction( + rgb->g = dc_fixpt_from_fraction( ramp->gamma[i + 256], scaler); - rgb->b = dal_fixed31_32_from_fraction( + rgb->b = dc_fixpt_from_fraction( ramp->gamma[i + 512], scaler); ++rgb; ++i; } while (i != GAMMA_RGB_256_ENTRIES); - rgb->r = dal_fixed31_32_mul(rgb_last->r, + rgb->r = dc_fixpt_mul(rgb_last->r, dividers.divider1); - rgb->g = dal_fixed31_32_mul(rgb_last->g, + rgb->g = dc_fixpt_mul(rgb_last->g, dividers.divider1); - rgb->b = dal_fixed31_32_mul(rgb_last->b, + rgb->b = dc_fixpt_mul(rgb_last->b, dividers.divider1); ++rgb; - rgb->r = dal_fixed31_32_mul(rgb_last->r, + rgb->r = dc_fixpt_mul(rgb_last->r, dividers.divider2); - rgb->g = dal_fixed31_32_mul(rgb_last->g, + rgb->g = dc_fixpt_mul(rgb_last->g, dividers.divider2); - rgb->b = dal_fixed31_32_mul(rgb_last->b, + rgb->b = dc_fixpt_mul(rgb_last->b, dividers.divider2); ++rgb; - rgb->r = dal_fixed31_32_mul(rgb_last->r, + rgb->r = dc_fixpt_mul(rgb_last->r, dividers.divider3); - rgb->g = dal_fixed31_32_mul(rgb_last->g, + rgb->g = dc_fixpt_mul(rgb_last->g, dividers.divider3); - rgb->b = dal_fixed31_32_mul(rgb_last->b, + rgb->b = dc_fixpt_mul(rgb_last->b, dividers.divider3); } @@ -913,7 +913,7 @@ static void apply_lut_1d( struct fixed31_32 lut2; const int max_lut_index = 4095; const struct fixed31_32 max_lut_index_f = - dal_fixed31_32_from_int_nonconst(max_lut_index); + dc_fixpt_from_int(max_lut_index); int32_t index = 0, index_next = 0; struct fixed31_32 index_f; struct fixed31_32 delta_lut; @@ -931,10 +931,10 @@ static void apply_lut_1d( else regamma_y = &tf_pts->blue[i]; - norm_y = dal_fixed31_32_mul(max_lut_index_f, + norm_y = dc_fixpt_mul(max_lut_index_f, *regamma_y); - index = dal_fixed31_32_floor(norm_y); - index_f = dal_fixed31_32_from_int_nonconst(index); + index = dc_fixpt_floor(norm_y); + index_f = dc_fixpt_from_int(index); if (index < 0 || index > max_lut_index) continue; @@ -953,11 +953,11 @@ static void apply_lut_1d( } // we have everything now, so interpolate - delta_lut = dal_fixed31_32_sub(lut2, lut1); - delta_index = dal_fixed31_32_sub(norm_y, index_f); + delta_lut = dc_fixpt_sub(lut2, lut1); + delta_index = dc_fixpt_sub(norm_y, index_f); - *regamma_y = dal_fixed31_32_add(lut1, - dal_fixed31_32_mul(delta_index, delta_lut)); + *regamma_y = dc_fixpt_add(lut1, + dc_fixpt_mul(delta_index, delta_lut)); } } } @@ -973,7 +973,7 @@ static void build_evenly_distributed_points( uint32_t i = 0; do { - struct fixed31_32 value = dal_fixed31_32_from_fraction(i, + struct fixed31_32 value = dc_fixpt_from_fraction(i, numberof_points - 1); p->r = value; @@ -984,21 +984,21 @@ static void build_evenly_distributed_points( ++i; } while (i != numberof_points); - p->r = dal_fixed31_32_div(p_last->r, dividers.divider1); - p->g = dal_fixed31_32_div(p_last->g, dividers.divider1); - p->b = dal_fixed31_32_div(p_last->b, dividers.divider1); + p->r = dc_fixpt_div(p_last->r, dividers.divider1); + p->g = dc_fixpt_div(p_last->g, dividers.divider1); + p->b = dc_fixpt_div(p_last->b, dividers.divider1); ++p; - p->r = dal_fixed31_32_div(p_last->r, dividers.divider2); - p->g = dal_fixed31_32_div(p_last->g, dividers.divider2); - p->b = dal_fixed31_32_div(p_last->b, dividers.divider2); + p->r = dc_fixpt_div(p_last->r, dividers.divider2); + p->g = dc_fixpt_div(p_last->g, dividers.divider2); + p->b = dc_fixpt_div(p_last->b, dividers.divider2); ++p; - p->r = dal_fixed31_32_div(p_last->r, dividers.divider3); - p->g = dal_fixed31_32_div(p_last->g, dividers.divider3); - p->b = dal_fixed31_32_div(p_last->b, dividers.divider3); + p->r = dc_fixpt_div(p_last->r, dividers.divider3); + p->g = dc_fixpt_div(p_last->g, dividers.divider3); + p->b = dc_fixpt_div(p_last->b, dividers.divider3); } static inline void copy_rgb_regamma_to_coordinates_x( @@ -1094,7 +1094,7 @@ static void interpolate_user_regamma(uint32_t hw_points_num, struct fixed31_32 *tf_point; struct fixed31_32 hw_x; struct fixed31_32 norm_factor = - dal_fixed31_32_from_int_nonconst(255); + dc_fixpt_from_int(255); struct fixed31_32 norm_x; struct fixed31_32 index_f; struct fixed31_32 lut1; @@ -1105,9 +1105,9 @@ static void interpolate_user_regamma(uint32_t hw_points_num, i = 0; /* fixed_pt library has problems handling too small values */ while (i != 32) { - tf_pts->red[i] = dal_fixed31_32_zero; - tf_pts->green[i] = dal_fixed31_32_zero; - tf_pts->blue[i] = dal_fixed31_32_zero; + tf_pts->red[i] = dc_fixpt_zero; + tf_pts->green[i] = dc_fixpt_zero; + tf_pts->blue[i] = dc_fixpt_zero; ++i; } while (i <= hw_points_num + 1) { @@ -1129,12 +1129,12 @@ static void interpolate_user_regamma(uint32_t hw_points_num, } else hw_x = coordinates_x[i].x; - norm_x = dal_fixed31_32_mul(norm_factor, hw_x); - index = dal_fixed31_32_floor(norm_x); + norm_x = dc_fixpt_mul(norm_factor, hw_x); + index = dc_fixpt_floor(norm_x); if (index < 0 || index > 255) continue; - index_f = dal_fixed31_32_from_int_nonconst(index); + index_f = dc_fixpt_from_int(index); index_next = (index == 255) ? index : index + 1; if (color == 0) { @@ -1149,11 +1149,11 @@ static void interpolate_user_regamma(uint32_t hw_points_num, } // we have everything now, so interpolate - delta_lut = dal_fixed31_32_sub(lut2, lut1); - delta_index = dal_fixed31_32_sub(norm_x, index_f); + delta_lut = dc_fixpt_sub(lut2, lut1); + delta_index = dc_fixpt_sub(norm_x, index_f); - *tf_point = dal_fixed31_32_add(lut1, - dal_fixed31_32_mul(delta_index, delta_lut)); + *tf_point = dc_fixpt_add(lut1, + dc_fixpt_mul(delta_index, delta_lut)); } ++i; } @@ -1168,15 +1168,15 @@ static void build_new_custom_resulted_curve( i = 0; while (i != hw_points_num + 1) { - tf_pts->red[i] = dal_fixed31_32_clamp( - tf_pts->red[i], dal_fixed31_32_zero, - dal_fixed31_32_one); - tf_pts->green[i] = dal_fixed31_32_clamp( - tf_pts->green[i], dal_fixed31_32_zero, - dal_fixed31_32_one); - tf_pts->blue[i] = dal_fixed31_32_clamp( - tf_pts->blue[i], dal_fixed31_32_zero, - dal_fixed31_32_one); + tf_pts->red[i] = dc_fixpt_clamp( + tf_pts->red[i], dc_fixpt_zero, + dc_fixpt_one); + tf_pts->green[i] = dc_fixpt_clamp( + tf_pts->green[i], dc_fixpt_zero, + dc_fixpt_one); + tf_pts->blue[i] = dc_fixpt_clamp( + tf_pts->blue[i], dc_fixpt_zero, + dc_fixpt_one); ++i; } @@ -1290,9 +1290,9 @@ bool mod_color_calculate_regamma_params(struct dc_transfer_func *output_tf, if (!coeff) goto coeff_alloc_fail; - dividers.divider1 = dal_fixed31_32_from_fraction(3, 2); - dividers.divider2 = dal_fixed31_32_from_int(2); - dividers.divider3 = dal_fixed31_32_from_fraction(5, 2); + dividers.divider1 = dc_fixpt_from_fraction(3, 2); + dividers.divider2 = dc_fixpt_from_int(2); + dividers.divider3 = dc_fixpt_from_fraction(5, 2); tf = output_tf->tf; @@ -1357,15 +1357,15 @@ bool calculate_user_regamma_coeff(struct dc_transfer_func *output_tf, uint32_t i = 0; do { - coeff.a0[i] = dal_fixed31_32_from_fraction( + coeff.a0[i] = dc_fixpt_from_fraction( regamma->coeff.A0[i], 10000000); - coeff.a1[i] = dal_fixed31_32_from_fraction( + coeff.a1[i] = dc_fixpt_from_fraction( regamma->coeff.A1[i], 1000); - coeff.a2[i] = dal_fixed31_32_from_fraction( + coeff.a2[i] = dc_fixpt_from_fraction( regamma->coeff.A2[i], 1000); - coeff.a3[i] = dal_fixed31_32_from_fraction( + coeff.a3[i] = dc_fixpt_from_fraction( regamma->coeff.A3[i], 1000); - coeff.user_gamma[i] = dal_fixed31_32_from_fraction( + coeff.user_gamma[i] = dc_fixpt_from_fraction( regamma->coeff.gamma[i], 1000); ++i; @@ -1374,9 +1374,9 @@ bool calculate_user_regamma_coeff(struct dc_transfer_func *output_tf, i = 0; /* fixed_pt library has problems handling too small values */ while (i != 32) { - output_tf->tf_pts.red[i] = dal_fixed31_32_zero; - output_tf->tf_pts.green[i] = dal_fixed31_32_zero; - output_tf->tf_pts.blue[i] = dal_fixed31_32_zero; + output_tf->tf_pts.red[i] = dc_fixpt_zero; + output_tf->tf_pts.green[i] = dc_fixpt_zero; + output_tf->tf_pts.blue[i] = dc_fixpt_zero; ++coord_x; ++i; } @@ -1423,9 +1423,9 @@ bool calculate_user_regamma_ramp(struct dc_transfer_func *output_tf, if (!rgb_regamma) goto rgb_regamma_alloc_fail; - dividers.divider1 = dal_fixed31_32_from_fraction(3, 2); - dividers.divider2 = dal_fixed31_32_from_int(2); - dividers.divider3 = dal_fixed31_32_from_fraction(5, 2); + dividers.divider1 = dc_fixpt_from_fraction(3, 2); + dividers.divider2 = dc_fixpt_from_int(2); + dividers.divider3 = dc_fixpt_from_fraction(5, 2); scale_user_regamma_ramp(rgb_user, ®amma->ramp, dividers); @@ -1451,7 +1451,7 @@ bool calculate_user_regamma_ramp(struct dc_transfer_func *output_tf, kfree(rgb_regamma); rgb_regamma_alloc_fail: - kfree(rgb_user); + kvfree(rgb_user); rgb_user_alloc_fail: return ret; } @@ -1496,9 +1496,9 @@ bool mod_color_calculate_degamma_params(struct dc_transfer_func *input_tf, if (!coeff) goto coeff_alloc_fail; - dividers.divider1 = dal_fixed31_32_from_fraction(3, 2); - dividers.divider2 = dal_fixed31_32_from_int(2); - dividers.divider3 = dal_fixed31_32_from_fraction(5, 2); + dividers.divider1 = dc_fixpt_from_fraction(3, 2); + dividers.divider2 = dc_fixpt_from_int(2); + dividers.divider3 = dc_fixpt_from_fraction(5, 2); tf = input_tf->tf; diff --git a/drivers/gpu/drm/amd/display/modules/inc/mod_stats.h b/drivers/gpu/drm/amd/display/modules/inc/mod_stats.h index 3230e2adb870..3812094b52e8 100644 --- a/drivers/gpu/drm/amd/display/modules/inc/mod_stats.h +++ b/drivers/gpu/drm/amd/display/modules/inc/mod_stats.h @@ -46,6 +46,10 @@ void mod_stats_dump(struct mod_stats *mod_stats); void mod_stats_reset_data(struct mod_stats *mod_stats); +void mod_stats_update_event(struct mod_stats *mod_stats, + char *event_string, + unsigned int length); + void mod_stats_update_flip(struct mod_stats *mod_stats, unsigned long timestamp_in_ns); diff --git a/drivers/gpu/drm/amd/display/modules/stats/stats.c b/drivers/gpu/drm/amd/display/modules/stats/stats.c index 48e02197919f..3f7d47fdc367 100644 --- a/drivers/gpu/drm/amd/display/modules/stats/stats.c +++ b/drivers/gpu/drm/amd/display/modules/stats/stats.c @@ -36,9 +36,14 @@ #define DAL_STATS_ENTRIES_REGKEY_DEFAULT 0x00350000 #define DAL_STATS_ENTRIES_REGKEY_MAX 0x01000000 +#define DAL_STATS_EVENT_ENTRIES_DEFAULT 0x00000100 + #define MOD_STATS_NUM_VSYNCS 5 +#define MOD_STATS_EVENT_STRING_MAX 512 struct stats_time_cache { + unsigned int entry_id; + unsigned long flip_timestamp_in_ns; unsigned long vupdate_timestamp_in_ns; @@ -63,15 +68,26 @@ struct stats_time_cache { unsigned int flags; }; +struct stats_event_cache { + unsigned int entry_id; + char event_string[MOD_STATS_EVENT_STRING_MAX]; +}; + struct core_stats { struct mod_stats public; struct dc *dc; + bool enabled; + unsigned int entries; + unsigned int event_entries; + unsigned int entry_id; + struct stats_time_cache *time; unsigned int index; - bool enabled; - unsigned int entries; + struct stats_event_cache *events; + unsigned int event_index; + }; #define MOD_STATS_TO_CORE(mod_stats)\ @@ -99,12 +115,12 @@ struct mod_stats *mod_stats_create(struct dc *dc) unsigned int reg_data; int i = 0; + if (dc == NULL) + goto fail_construct; + core_stats = kzalloc(sizeof(struct core_stats), GFP_KERNEL); if (core_stats == NULL) - goto fail_alloc_context; - - if (dc == NULL) goto fail_construct; core_stats->dc = dc; @@ -125,27 +141,45 @@ struct mod_stats *mod_stats_create(struct dc *dc) else core_stats->entries = reg_data; } + core_stats->time = kzalloc( + sizeof(struct stats_time_cache) * + core_stats->entries, + GFP_KERNEL); + + if (core_stats->time == NULL) + goto fail_construct_time; - core_stats->time = kzalloc(sizeof(struct stats_time_cache) * core_stats->entries, + core_stats->event_entries = DAL_STATS_EVENT_ENTRIES_DEFAULT; + core_stats->events = kzalloc( + sizeof(struct stats_event_cache) * + core_stats->event_entries, GFP_KERNEL); + + if (core_stats->events == NULL) + goto fail_construct_events; + } else { core_stats->entries = 0; } - if (core_stats->time == NULL) - goto fail_construct; - /* Purposely leave index 0 unused so we don't need special logic to * handle calculation cases that depend on previous flip data. */ core_stats->index = 1; + core_stats->event_index = 0; + + // Keeps track of ordering within the different stats structures + core_stats->entry_id = 0; return &core_stats->public; -fail_construct: +fail_construct_events: + kfree(core_stats->time); + +fail_construct_time: kfree(core_stats); -fail_alloc_context: +fail_construct: return NULL; } @@ -157,6 +191,9 @@ void mod_stats_destroy(struct mod_stats *mod_stats) if (core_stats->time != NULL) kfree(core_stats->time); + if (core_stats->events != NULL) + kfree(core_stats->events); + kfree(core_stats); } } @@ -167,7 +204,11 @@ void mod_stats_dump(struct mod_stats *mod_stats) struct dal_logger *logger = NULL; struct core_stats *core_stats = NULL; struct stats_time_cache *time = NULL; + struct stats_event_cache *events = NULL; + unsigned int time_index = 1; + unsigned int event_index = 0; unsigned int index = 0; + struct log_entry log_entry; if (mod_stats == NULL) return; @@ -176,45 +217,62 @@ void mod_stats_dump(struct mod_stats *mod_stats) dc = core_stats->dc; logger = dc->ctx->logger; time = core_stats->time; - - //LogEntry* pLog = GetLog()->Open(LogMajor_ISR, LogMinor_ISR_FreeSyncSW); - - //if (!pLog->IsDummyEntry()) - { - dm_logger_write(logger, LOG_PROFILING, "==Display Caps==\n"); - dm_logger_write(logger, LOG_PROFILING, "\n"); - dm_logger_write(logger, LOG_PROFILING, "\n"); - - dm_logger_write(logger, LOG_PROFILING, "==Stats==\n"); - dm_logger_write(logger, LOG_PROFILING, - "render avgRender minWindow midPoint maxWindow vsyncToFlip flipToVsync #vsyncBetweenFlip #frame insertDuration vTotalMin vTotalMax eventTrigs vSyncTime1 vSyncTime2 vSyncTime3 vSyncTime4 vSyncTime5 flags\n"); - - for (int i = 0; i < core_stats->index && i < core_stats->entries; i++) { - dm_logger_write(logger, LOG_PROFILING, - "%u %u %u %u %u %u %u %u %u %u %u %u %u %u %u %u %u %u %u", - time[i].render_time_in_us, - time[i].avg_render_time_in_us_last_ten, - time[i].min_window, - time[i].lfc_mid_point_in_us, - time[i].max_window, - time[i].vsync_to_flip_time_in_us, - time[i].flip_to_vsync_time_in_us, - time[i].num_vsync_between_flips, - time[i].num_frames_inserted, - time[i].inserted_duration_in_us, - time[i].v_total_min, - time[i].v_total_max, - time[i].event_triggers, - time[i].v_sync_time_in_us[0], - time[i].v_sync_time_in_us[1], - time[i].v_sync_time_in_us[2], - time[i].v_sync_time_in_us[3], - time[i].v_sync_time_in_us[4], - time[i].flags); + events = core_stats->events; + + DISPLAY_STATS_BEGIN(log_entry); + + DISPLAY_STATS("==Display Caps==\n"); + + DISPLAY_STATS("==Display Stats==\n"); + + DISPLAY_STATS("%10s %10s %10s %10s %10s" + " %11s %11s %17s %10s %14s" + " %10s %10s %10s %10s %10s" + " %10s %10s %10s %10s\n", + "render", "avgRender", + "minWindow", "midPoint", "maxWindow", + "vsyncToFlip", "flipToVsync", "vsyncsBetweenFlip", + "numFrame", "insertDuration", + "vTotalMin", "vTotalMax", "eventTrigs", + "vSyncTime1", "vSyncTime2", "vSyncTime3", + "vSyncTime4", "vSyncTime5", "flags"); + + for (int i = 0; i < core_stats->entry_id; i++) { + if (event_index < core_stats->event_index && + i == events[event_index].entry_id) { + DISPLAY_STATS("%s\n", events[event_index].event_string); + event_index++; + } else if (time_index < core_stats->index && + i == time[time_index].entry_id) { + DISPLAY_STATS("%10u %10u %10u %10u %10u" + " %11u %11u %17u %10u %14u" + " %10u %10u %10u %10u %10u" + " %10u %10u %10u %10u\n", + time[time_index].render_time_in_us, + time[time_index].avg_render_time_in_us_last_ten, + time[time_index].min_window, + time[time_index].lfc_mid_point_in_us, + time[time_index].max_window, + time[time_index].vsync_to_flip_time_in_us, + time[time_index].flip_to_vsync_time_in_us, + time[time_index].num_vsync_between_flips, + time[time_index].num_frames_inserted, + time[time_index].inserted_duration_in_us, + time[time_index].v_total_min, + time[time_index].v_total_max, + time[time_index].event_triggers, + time[time_index].v_sync_time_in_us[0], + time[time_index].v_sync_time_in_us[1], + time[time_index].v_sync_time_in_us[2], + time[time_index].v_sync_time_in_us[3], + time[time_index].v_sync_time_in_us[4], + time[time_index].flags); + + time_index++; } } - //GetLog()->Close(pLog); - //GetLog()->UnSetLogMask(LogMajor_ISR, LogMinor_ISR_FreeSyncSW); + + DISPLAY_STATS_END(log_entry); } void mod_stats_reset_data(struct mod_stats *mod_stats) @@ -231,7 +289,46 @@ void mod_stats_reset_data(struct mod_stats *mod_stats) memset(core_stats->time, 0, sizeof(struct stats_time_cache) * core_stats->entries); + memset(core_stats->events, 0, + sizeof(struct stats_event_cache) * core_stats->event_entries); + core_stats->index = 1; + core_stats->event_index = 0; + + // Keeps track of ordering within the different stats structures + core_stats->entry_id = 0; +} + +void mod_stats_update_event(struct mod_stats *mod_stats, + char *event_string, + unsigned int length) +{ + struct core_stats *core_stats = NULL; + struct stats_event_cache *events = NULL; + unsigned int index = 0; + unsigned int copy_length = 0; + + if (mod_stats == NULL) + return; + + core_stats = MOD_STATS_TO_CORE(mod_stats); + + if (core_stats->event_index >= core_stats->event_entries) + return; + + events = core_stats->events; + index = core_stats->event_index; + + copy_length = length; + if (length > MOD_STATS_EVENT_STRING_MAX) + copy_length = MOD_STATS_EVENT_STRING_MAX; + + memcpy(&events[index].event_string, event_string, copy_length); + events[index].event_string[copy_length - 1] = '\0'; + + events[index].entry_id = core_stats->entry_id; + core_stats->event_index++; + core_stats->entry_id++; } void mod_stats_update_flip(struct mod_stats *mod_stats, @@ -272,7 +369,9 @@ void mod_stats_update_flip(struct mod_stats *mod_stats, (timestamp_in_ns - time[index - 1].vupdate_timestamp_in_ns) / 1000; + time[index].entry_id = core_stats->entry_id; core_stats->index++; + core_stats->entry_id++; } void mod_stats_update_vupdate(struct mod_stats *mod_stats, diff --git a/drivers/gpu/drm/amd/include/amd_shared.h b/drivers/gpu/drm/amd/include/amd_shared.h index 33de33016bda..b178176b72ac 100644 --- a/drivers/gpu/drm/amd/include/amd_shared.h +++ b/drivers/gpu/drm/amd/include/amd_shared.h @@ -92,7 +92,7 @@ enum amd_powergating_state { #define AMD_CG_SUPPORT_GFX_3D_CGLS (1 << 21) #define AMD_CG_SUPPORT_DRM_MGCG (1 << 22) #define AMD_CG_SUPPORT_DF_MGCG (1 << 23) - +#define AMD_CG_SUPPORT_VCN_MGCG (1 << 24) /* PG flags */ #define AMD_PG_SUPPORT_GFX_PG (1 << 0) #define AMD_PG_SUPPORT_GFX_SMG (1 << 1) @@ -108,6 +108,7 @@ enum amd_powergating_state { #define AMD_PG_SUPPORT_GFX_QUICK_MG (1 << 11) #define AMD_PG_SUPPORT_GFX_PIPELINE (1 << 12) #define AMD_PG_SUPPORT_MMHUB (1 << 13) +#define AMD_PG_SUPPORT_VCN (1 << 14) enum PP_FEATURE_MASK { PP_SCLK_DPM_MASK = 0x1, diff --git a/drivers/gpu/drm/amd/include/asic_reg/dce/dce_12_0_offset.h b/drivers/gpu/drm/amd/include/asic_reg/dce/dce_12_0_offset.h index f730d0629020..b6f74bf4af02 100644 --- a/drivers/gpu/drm/amd/include/asic_reg/dce/dce_12_0_offset.h +++ b/drivers/gpu/drm/amd/include/asic_reg/dce/dce_12_0_offset.h @@ -2095,6 +2095,18 @@ #define mmDC_GPIO_AUX_CTRL_2_BASE_IDX 2 #define mmDC_GPIO_RXEN 0x212f #define mmDC_GPIO_RXEN_BASE_IDX 2 +#define mmDC_GPIO_AUX_CTRL_3 0x2130 +#define mmDC_GPIO_AUX_CTRL_3_BASE_IDX 2 +#define mmDC_GPIO_AUX_CTRL_4 0x2131 +#define mmDC_GPIO_AUX_CTRL_4_BASE_IDX 2 +#define mmDC_GPIO_AUX_CTRL_5 0x2132 +#define mmDC_GPIO_AUX_CTRL_5_BASE_IDX 2 +#define mmAUXI2C_PAD_ALL_PWR_OK 0x2133 +#define mmAUXI2C_PAD_ALL_PWR_OK_BASE_IDX 2 +#define mmDC_GPIO_PULLUPEN 0x2134 +#define mmDC_GPIO_PULLUPEN_BASE_IDX 2 +#define mmDC_GPIO_AUX_CTRL_6 0x2135 +#define mmDC_GPIO_AUX_CTRL_6_BASE_IDX 2 #define mmBPHYC_DAC_MACRO_CNTL 0x2136 #define mmBPHYC_DAC_MACRO_CNTL_BASE_IDX 2 #define mmDAC_MACRO_CNTL_RESERVED0 0x2136 diff --git a/drivers/gpu/drm/amd/include/asic_reg/dce/dce_12_0_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/dce/dce_12_0_sh_mask.h index 6d3162c42957..bcd190a3fcdd 100644 --- a/drivers/gpu/drm/amd/include/asic_reg/dce/dce_12_0_sh_mask.h +++ b/drivers/gpu/drm/amd/include/asic_reg/dce/dce_12_0_sh_mask.h @@ -10971,6 +10971,158 @@ #define DC_GPIO_RXEN__DC_GPIO_BLON_RXEN_MASK 0x00100000L #define DC_GPIO_RXEN__DC_GPIO_DIGON_RXEN_MASK 0x00200000L #define DC_GPIO_RXEN__DC_GPIO_ENA_BL_RXEN_MASK 0x00400000L +//DC_GPIO_AUX_CTRL_3 +#define DC_GPIO_AUX_CTRL_3__AUX1_NEN_RTERM__SHIFT 0x0 +#define DC_GPIO_AUX_CTRL_3__AUX2_NEN_RTERM__SHIFT 0x1 +#define DC_GPIO_AUX_CTRL_3__AUX3_NEN_RTERM__SHIFT 0x2 +#define DC_GPIO_AUX_CTRL_3__AUX4_NEN_RTERM__SHIFT 0x3 +#define DC_GPIO_AUX_CTRL_3__AUX5_NEN_RTERM__SHIFT 0x4 +#define DC_GPIO_AUX_CTRL_3__AUX6_NEN_RTERM__SHIFT 0x5 +#define DC_GPIO_AUX_CTRL_3__AUX1_DP_DN_SWAP__SHIFT 0x8 +#define DC_GPIO_AUX_CTRL_3__AUX2_DP_DN_SWAP__SHIFT 0x9 +#define DC_GPIO_AUX_CTRL_3__AUX3_DP_DN_SWAP__SHIFT 0xa +#define DC_GPIO_AUX_CTRL_3__AUX4_DP_DN_SWAP__SHIFT 0xb +#define DC_GPIO_AUX_CTRL_3__AUX5_DP_DN_SWAP__SHIFT 0xc +#define DC_GPIO_AUX_CTRL_3__AUX6_DP_DN_SWAP__SHIFT 0xd +#define DC_GPIO_AUX_CTRL_3__AUX1_HYS_TUNE__SHIFT 0x10 +#define DC_GPIO_AUX_CTRL_3__AUX2_HYS_TUNE__SHIFT 0x12 +#define DC_GPIO_AUX_CTRL_3__AUX3_HYS_TUNE__SHIFT 0x14 +#define DC_GPIO_AUX_CTRL_3__AUX4_HYS_TUNE__SHIFT 0x16 +#define DC_GPIO_AUX_CTRL_3__AUX5_HYS_TUNE__SHIFT 0x18 +#define DC_GPIO_AUX_CTRL_3__AUX6_HYS_TUNE__SHIFT 0x1a +#define DC_GPIO_AUX_CTRL_3__AUX1_NEN_RTERM_MASK 0x00000001L +#define DC_GPIO_AUX_CTRL_3__AUX2_NEN_RTERM_MASK 0x00000002L +#define DC_GPIO_AUX_CTRL_3__AUX3_NEN_RTERM_MASK 0x00000004L +#define DC_GPIO_AUX_CTRL_3__AUX4_NEN_RTERM_MASK 0x00000008L +#define DC_GPIO_AUX_CTRL_3__AUX5_NEN_RTERM_MASK 0x00000010L +#define DC_GPIO_AUX_CTRL_3__AUX6_NEN_RTERM_MASK 0x00000020L +#define DC_GPIO_AUX_CTRL_3__AUX1_DP_DN_SWAP_MASK 0x00000100L +#define DC_GPIO_AUX_CTRL_3__AUX2_DP_DN_SWAP_MASK 0x00000200L +#define DC_GPIO_AUX_CTRL_3__AUX3_DP_DN_SWAP_MASK 0x00000400L +#define DC_GPIO_AUX_CTRL_3__AUX4_DP_DN_SWAP_MASK 0x00000800L +#define DC_GPIO_AUX_CTRL_3__AUX5_DP_DN_SWAP_MASK 0x00001000L +#define DC_GPIO_AUX_CTRL_3__AUX6_DP_DN_SWAP_MASK 0x00002000L +#define DC_GPIO_AUX_CTRL_3__AUX1_HYS_TUNE_MASK 0x00030000L +#define DC_GPIO_AUX_CTRL_3__AUX2_HYS_TUNE_MASK 0x000C0000L +#define DC_GPIO_AUX_CTRL_3__AUX3_HYS_TUNE_MASK 0x00300000L +#define DC_GPIO_AUX_CTRL_3__AUX4_HYS_TUNE_MASK 0x00C00000L +#define DC_GPIO_AUX_CTRL_3__AUX5_HYS_TUNE_MASK 0x03000000L +#define DC_GPIO_AUX_CTRL_3__AUX6_HYS_TUNE_MASK 0x0C000000L +//DC_GPIO_AUX_CTRL_4 +#define DC_GPIO_AUX_CTRL_4__AUX1_AUX_CTRL__SHIFT 0x0 +#define DC_GPIO_AUX_CTRL_4__AUX2_AUX_CTRL__SHIFT 0x4 +#define DC_GPIO_AUX_CTRL_4__AUX3_AUX_CTRL__SHIFT 0x8 +#define DC_GPIO_AUX_CTRL_4__AUX4_AUX_CTRL__SHIFT 0xc +#define DC_GPIO_AUX_CTRL_4__AUX5_AUX_CTRL__SHIFT 0x10 +#define DC_GPIO_AUX_CTRL_4__AUX6_AUX_CTRL__SHIFT 0x14 +#define DC_GPIO_AUX_CTRL_4__AUX1_AUX_CTRL_MASK 0x0000000FL +#define DC_GPIO_AUX_CTRL_4__AUX2_AUX_CTRL_MASK 0x000000F0L +#define DC_GPIO_AUX_CTRL_4__AUX3_AUX_CTRL_MASK 0x00000F00L +#define DC_GPIO_AUX_CTRL_4__AUX4_AUX_CTRL_MASK 0x0000F000L +#define DC_GPIO_AUX_CTRL_4__AUX5_AUX_CTRL_MASK 0x000F0000L +#define DC_GPIO_AUX_CTRL_4__AUX6_AUX_CTRL_MASK 0x00F00000L +//DC_GPIO_AUX_CTRL_5 +#define DC_GPIO_AUX_CTRL_5__AUX1_VOD_TUNE__SHIFT 0x0 +#define DC_GPIO_AUX_CTRL_5__AUX2_VOD_TUNE__SHIFT 0x2 +#define DC_GPIO_AUX_CTRL_5__AUX3_VOD_TUNE__SHIFT 0x4 +#define DC_GPIO_AUX_CTRL_5__AUX4_VOD_TUNE__SHIFT 0x6 +#define DC_GPIO_AUX_CTRL_5__AUX5_VOD_TUNE__SHIFT 0x8 +#define DC_GPIO_AUX_CTRL_5__AUX6_VOD_TUNE__SHIFT 0xa +#define DC_GPIO_AUX_CTRL_5__DDC_PAD1_I2CMODE__SHIFT 0xc +#define DC_GPIO_AUX_CTRL_5__DDC_PAD2_I2CMODE__SHIFT 0xd +#define DC_GPIO_AUX_CTRL_5__DDC_PAD3_I2CMODE__SHIFT 0xe +#define DC_GPIO_AUX_CTRL_5__DDC_PAD4_I2CMODE__SHIFT 0xf +#define DC_GPIO_AUX_CTRL_5__DDC_PAD5_I2CMODE__SHIFT 0x10 +#define DC_GPIO_AUX_CTRL_5__DDC_PAD6_I2CMODE__SHIFT 0x11 +#define DC_GPIO_AUX_CTRL_5__DDC1_I2C_VPH_1V2_EN__SHIFT 0x12 +#define DC_GPIO_AUX_CTRL_5__DDC2_I2C_VPH_1V2_EN__SHIFT 0x13 +#define DC_GPIO_AUX_CTRL_5__DDC3_I2C_VPH_1V2_EN__SHIFT 0x14 +#define DC_GPIO_AUX_CTRL_5__DDC4_I2C_VPH_1V2_EN__SHIFT 0x15 +#define DC_GPIO_AUX_CTRL_5__DDC5_I2C_VPH_1V2_EN__SHIFT 0x16 +#define DC_GPIO_AUX_CTRL_5__DDC6_I2C_VPH_1V2_EN__SHIFT 0x17 +#define DC_GPIO_AUX_CTRL_5__DDC1_PAD_I2C_CTRL__SHIFT 0x18 +#define DC_GPIO_AUX_CTRL_5__DDC2_PAD_I2C_CTRL__SHIFT 0x19 +#define DC_GPIO_AUX_CTRL_5__DDC3_PAD_I2C_CTRL__SHIFT 0x1a +#define DC_GPIO_AUX_CTRL_5__DDC4_PAD_I2C_CTRL__SHIFT 0x1b +#define DC_GPIO_AUX_CTRL_5__DDC5_PAD_I2C_CTRL__SHIFT 0x1c +#define DC_GPIO_AUX_CTRL_5__DDC6_PAD_I2C_CTRL__SHIFT 0x1d +#define DC_GPIO_AUX_CTRL_5__AUX1_VOD_TUNE_MASK 0x00000003L +#define DC_GPIO_AUX_CTRL_5__AUX2_VOD_TUNE_MASK 0x0000000CL +#define DC_GPIO_AUX_CTRL_5__AUX3_VOD_TUNE_MASK 0x00000030L +#define DC_GPIO_AUX_CTRL_5__AUX4_VOD_TUNE_MASK 0x000000C0L +#define DC_GPIO_AUX_CTRL_5__AUX5_VOD_TUNE_MASK 0x00000300L +#define DC_GPIO_AUX_CTRL_5__AUX6_VOD_TUNE_MASK 0x00000C00L +#define DC_GPIO_AUX_CTRL_5__DDC_PAD1_I2CMODE_MASK 0x00001000L +#define DC_GPIO_AUX_CTRL_5__DDC_PAD2_I2CMODE_MASK 0x00002000L +#define DC_GPIO_AUX_CTRL_5__DDC_PAD3_I2CMODE_MASK 0x00004000L +#define DC_GPIO_AUX_CTRL_5__DDC_PAD4_I2CMODE_MASK 0x00008000L +#define DC_GPIO_AUX_CTRL_5__DDC_PAD5_I2CMODE_MASK 0x00010000L +#define DC_GPIO_AUX_CTRL_5__DDC_PAD6_I2CMODE_MASK 0x00020000L +#define DC_GPIO_AUX_CTRL_5__DDC1_I2C_VPH_1V2_EN_MASK 0x00040000L +#define DC_GPIO_AUX_CTRL_5__DDC2_I2C_VPH_1V2_EN_MASK 0x00080000L +#define DC_GPIO_AUX_CTRL_5__DDC3_I2C_VPH_1V2_EN_MASK 0x00100000L +#define DC_GPIO_AUX_CTRL_5__DDC4_I2C_VPH_1V2_EN_MASK 0x00200000L +#define DC_GPIO_AUX_CTRL_5__DDC5_I2C_VPH_1V2_EN_MASK 0x00400000L +#define DC_GPIO_AUX_CTRL_5__DDC6_I2C_VPH_1V2_EN_MASK 0x00800000L +#define DC_GPIO_AUX_CTRL_5__DDC1_PAD_I2C_CTRL_MASK 0x01000000L +#define DC_GPIO_AUX_CTRL_5__DDC2_PAD_I2C_CTRL_MASK 0x02000000L +#define DC_GPIO_AUX_CTRL_5__DDC3_PAD_I2C_CTRL_MASK 0x04000000L +#define DC_GPIO_AUX_CTRL_5__DDC4_PAD_I2C_CTRL_MASK 0x08000000L +#define DC_GPIO_AUX_CTRL_5__DDC5_PAD_I2C_CTRL_MASK 0x10000000L +#define DC_GPIO_AUX_CTRL_5__DDC6_PAD_I2C_CTRL_MASK 0x20000000L +//AUXI2C_PAD_ALL_PWR_OK +#define AUXI2C_PAD_ALL_PWR_OK__AUXI2C_PHY1_ALL_PWR_OK__SHIFT 0x0 +#define AUXI2C_PAD_ALL_PWR_OK__AUXI2C_PHY2_ALL_PWR_OK__SHIFT 0x1 +#define AUXI2C_PAD_ALL_PWR_OK__AUXI2C_PHY3_ALL_PWR_OK__SHIFT 0x2 +#define AUXI2C_PAD_ALL_PWR_OK__AUXI2C_PHY4_ALL_PWR_OK__SHIFT 0x3 +#define AUXI2C_PAD_ALL_PWR_OK__AUXI2C_PHY5_ALL_PWR_OK__SHIFT 0x4 +#define AUXI2C_PAD_ALL_PWR_OK__AUXI2C_PHY6_ALL_PWR_OK__SHIFT 0x5 +#define AUXI2C_PAD_ALL_PWR_OK__AUXI2C_PHY1_ALL_PWR_OK_MASK 0x00000001L +#define AUXI2C_PAD_ALL_PWR_OK__AUXI2C_PHY2_ALL_PWR_OK_MASK 0x00000002L +#define AUXI2C_PAD_ALL_PWR_OK__AUXI2C_PHY3_ALL_PWR_OK_MASK 0x00000004L +#define AUXI2C_PAD_ALL_PWR_OK__AUXI2C_PHY4_ALL_PWR_OK_MASK 0x00000008L +#define AUXI2C_PAD_ALL_PWR_OK__AUXI2C_PHY5_ALL_PWR_OK_MASK 0x00000010L +#define AUXI2C_PAD_ALL_PWR_OK__AUXI2C_PHY6_ALL_PWR_OK_MASK 0x00000020L +//DC_GPIO_PULLUPEN +#define DC_GPIO_PULLUPEN__DC_GPIO_GENERICA_PU_EN__SHIFT 0x0 +#define DC_GPIO_PULLUPEN__DC_GPIO_GENERICB_PU_EN__SHIFT 0x1 +#define DC_GPIO_PULLUPEN__DC_GPIO_GENERICC_PU_EN__SHIFT 0x2 +#define DC_GPIO_PULLUPEN__DC_GPIO_GENERICD_PU_EN__SHIFT 0x3 +#define DC_GPIO_PULLUPEN__DC_GPIO_GENERICE_PU_EN__SHIFT 0x4 +#define DC_GPIO_PULLUPEN__DC_GPIO_GENERICF_PU_EN__SHIFT 0x5 +#define DC_GPIO_PULLUPEN__DC_GPIO_GENERICG_PU_EN__SHIFT 0x6 +#define DC_GPIO_PULLUPEN__DC_GPIO_HSYNCA_PU_EN__SHIFT 0x8 +#define DC_GPIO_PULLUPEN__DC_GPIO_VSYNCA_PU_EN__SHIFT 0x9 +#define DC_GPIO_PULLUPEN__DC_GPIO_HPD1_PU_EN__SHIFT 0xe +#define DC_GPIO_PULLUPEN__DC_GPIO_BLON_PU_EN__SHIFT 0x14 +#define DC_GPIO_PULLUPEN__DC_GPIO_DIGON_PU_EN__SHIFT 0x15 +#define DC_GPIO_PULLUPEN__DC_GPIO_ENA_BL_PU_EN__SHIFT 0x16 +#define DC_GPIO_PULLUPEN__DC_GPIO_GENERICA_PU_EN_MASK 0x00000001L +#define DC_GPIO_PULLUPEN__DC_GPIO_GENERICB_PU_EN_MASK 0x00000002L +#define DC_GPIO_PULLUPEN__DC_GPIO_GENERICC_PU_EN_MASK 0x00000004L +#define DC_GPIO_PULLUPEN__DC_GPIO_GENERICD_PU_EN_MASK 0x00000008L +#define DC_GPIO_PULLUPEN__DC_GPIO_GENERICE_PU_EN_MASK 0x00000010L +#define DC_GPIO_PULLUPEN__DC_GPIO_GENERICF_PU_EN_MASK 0x00000020L +#define DC_GPIO_PULLUPEN__DC_GPIO_GENERICG_PU_EN_MASK 0x00000040L +#define DC_GPIO_PULLUPEN__DC_GPIO_HSYNCA_PU_EN_MASK 0x00000100L +#define DC_GPIO_PULLUPEN__DC_GPIO_VSYNCA_PU_EN_MASK 0x00000200L +#define DC_GPIO_PULLUPEN__DC_GPIO_HPD1_PU_EN_MASK 0x00004000L +#define DC_GPIO_PULLUPEN__DC_GPIO_BLON_PU_EN_MASK 0x00100000L +#define DC_GPIO_PULLUPEN__DC_GPIO_DIGON_PU_EN_MASK 0x00200000L +#define DC_GPIO_PULLUPEN__DC_GPIO_ENA_BL_PU_EN_MASK 0x00400000L +//DC_GPIO_AUX_CTRL_6 +#define DC_GPIO_AUX_CTRL_6__AUX1_PAD_RXSEL__SHIFT 0x0 +#define DC_GPIO_AUX_CTRL_6__AUX2_PAD_RXSEL__SHIFT 0x2 +#define DC_GPIO_AUX_CTRL_6__AUX3_PAD_RXSEL__SHIFT 0x4 +#define DC_GPIO_AUX_CTRL_6__AUX4_PAD_RXSEL__SHIFT 0x6 +#define DC_GPIO_AUX_CTRL_6__AUX5_PAD_RXSEL__SHIFT 0x8 +#define DC_GPIO_AUX_CTRL_6__AUX6_PAD_RXSEL__SHIFT 0xa +#define DC_GPIO_AUX_CTRL_6__AUX1_PAD_RXSEL_MASK 0x00000003L +#define DC_GPIO_AUX_CTRL_6__AUX2_PAD_RXSEL_MASK 0x0000000CL +#define DC_GPIO_AUX_CTRL_6__AUX3_PAD_RXSEL_MASK 0x00000030L +#define DC_GPIO_AUX_CTRL_6__AUX4_PAD_RXSEL_MASK 0x000000C0L +#define DC_GPIO_AUX_CTRL_6__AUX5_PAD_RXSEL_MASK 0x00000300L +#define DC_GPIO_AUX_CTRL_6__AUX6_PAD_RXSEL_MASK 0x00000C00L //BPHYC_DAC_MACRO_CNTL #define BPHYC_DAC_MACRO_CNTL__BPHYC_DAC_WHITE_LEVEL__SHIFT 0x0 #define BPHYC_DAC_MACRO_CNTL__BPHYC_DAC_WHITE_FINE_CONTROL__SHIFT 0x8 diff --git a/drivers/gpu/drm/amd/include/asic_reg/df/df_1_7_offset.h b/drivers/gpu/drm/amd/include/asic_reg/df/df_1_7_offset.h index 2b305dd021e8..e6044e27a913 100644 --- a/drivers/gpu/drm/amd/include/asic_reg/df/df_1_7_offset.h +++ b/drivers/gpu/drm/amd/include/asic_reg/df/df_1_7_offset.h @@ -30,4 +30,8 @@ #define mmDF_CS_AON0_DramBaseAddress0 0x0044 #define mmDF_CS_AON0_DramBaseAddress0_BASE_IDX 0 +#define mmDF_CS_AON0_CoherentSlaveModeCtrlA0 0x0214 +#define mmDF_CS_AON0_CoherentSlaveModeCtrlA0_BASE_IDX 0 + + #endif diff --git a/drivers/gpu/drm/amd/include/asic_reg/df/df_1_7_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/df/df_1_7_sh_mask.h index 2ba849798924..a78c99480e2d 100644 --- a/drivers/gpu/drm/amd/include/asic_reg/df/df_1_7_sh_mask.h +++ b/drivers/gpu/drm/amd/include/asic_reg/df/df_1_7_sh_mask.h @@ -45,4 +45,8 @@ #define DF_CS_AON0_DramBaseAddress0__IntLvAddrSel_MASK 0x00000700L #define DF_CS_AON0_DramBaseAddress0__DramBaseAddr_MASK 0xFFFFF000L +//DF_CS_AON0_CoherentSlaveModeCtrlA0 +#define DF_CS_AON0_CoherentSlaveModeCtrlA0__ForceParWrRMW__SHIFT 0x3 +#define DF_CS_AON0_CoherentSlaveModeCtrlA0__ForceParWrRMW_MASK 0x00000008L + #endif diff --git a/drivers/gpu/drm/amd/include/asic_reg/df/df_3_6_default.h b/drivers/gpu/drm/amd/include/asic_reg/df/df_3_6_default.h new file mode 100644 index 000000000000..e58c207ac980 --- /dev/null +++ b/drivers/gpu/drm/amd/include/asic_reg/df/df_3_6_default.h @@ -0,0 +1,26 @@ +/* + * Copyright (C) 2018 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included + * in all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS + * OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN + * AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. + */ +#ifndef _df_3_6_DEFAULT_HEADER +#define _df_3_6_DEFAULT_HEADER + +#define mmFabricConfigAccessControl_DEFAULT 0x00000000 + +#endif diff --git a/drivers/gpu/drm/amd/include/asic_reg/df/df_3_6_offset.h b/drivers/gpu/drm/amd/include/asic_reg/df/df_3_6_offset.h new file mode 100644 index 000000000000..a9575db8d7aa --- /dev/null +++ b/drivers/gpu/drm/amd/include/asic_reg/df/df_3_6_offset.h @@ -0,0 +1,33 @@ +/* + * Copyright (C) 2018 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included + * in all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS + * OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN + * AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. + */ +#ifndef _df_3_6_OFFSET_HEADER +#define _df_3_6_OFFSET_HEADER + +#define mmFabricConfigAccessControl 0x0410 +#define mmFabricConfigAccessControl_BASE_IDX 0 + +#define mmDF_PIE_AON0_DfGlobalClkGater 0x00fc +#define mmDF_PIE_AON0_DfGlobalClkGater_BASE_IDX 0 + +#define mmDF_CS_UMC_AON0_DramBaseAddress0 0x0044 +#define mmDF_CS_UMC_AON0_DramBaseAddress0_BASE_IDX 0 + +#endif diff --git a/drivers/gpu/drm/amd/include/asic_reg/df/df_3_6_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/df/df_3_6_sh_mask.h new file mode 100644 index 000000000000..88f7c69df6b9 --- /dev/null +++ b/drivers/gpu/drm/amd/include/asic_reg/df/df_3_6_sh_mask.h @@ -0,0 +1,48 @@ +/* + * Copyright (C) 2018 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included + * in all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS + * OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN + * AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. + */ +#ifndef _df_3_6_SH_MASK_HEADER +#define _df_3_6_SH_MASK_HEADER + +/* FabricConfigAccessControl */ +#define FabricConfigAccessControl__CfgRegInstAccEn__SHIFT 0x0 +#define FabricConfigAccessControl__CfgRegInstAccRegLock__SHIFT 0x1 +#define FabricConfigAccessControl__CfgRegInstID__SHIFT 0x10 +#define FabricConfigAccessControl__CfgRegInstAccEn_MASK 0x00000001L +#define FabricConfigAccessControl__CfgRegInstAccRegLock_MASK 0x00000002L +#define FabricConfigAccessControl__CfgRegInstID_MASK 0x00FF0000L + +/* DF_PIE_AON0_DfGlobalClkGater */ +#define DF_PIE_AON0_DfGlobalClkGater__MGCGMode__SHIFT 0x0 +#define DF_PIE_AON0_DfGlobalClkGater__MGCGMode_MASK 0x0000000FL + +/* DF_CS_AON0_DramBaseAddress0 */ +#define DF_CS_UMC_AON0_DramBaseAddress0__AddrRngVal__SHIFT 0x0 +#define DF_CS_UMC_AON0_DramBaseAddress0__LgcyMmioHoleEn__SHIFT 0x1 +#define DF_CS_UMC_AON0_DramBaseAddress0__IntLvNumChan__SHIFT 0x4 +#define DF_CS_UMC_AON0_DramBaseAddress0__IntLvAddrSel__SHIFT 0x8 +#define DF_CS_UMC_AON0_DramBaseAddress0__DramBaseAddr__SHIFT 0xc +#define DF_CS_UMC_AON0_DramBaseAddress0__AddrRngVal_MASK 0x00000001L +#define DF_CS_UMC_AON0_DramBaseAddress0__LgcyMmioHoleEn_MASK 0x00000002L +#define DF_CS_UMC_AON0_DramBaseAddress0__IntLvNumChan_MASK 0x000000F0L +#define DF_CS_UMC_AON0_DramBaseAddress0__IntLvAddrSel_MASK 0x00000700L +#define DF_CS_UMC_AON0_DramBaseAddress0__DramBaseAddr_MASK 0xFFFFF000L + +#endif diff --git a/drivers/gpu/drm/amd/include/atomfirmware.h b/drivers/gpu/drm/amd/include/atomfirmware.h index de177ce8ca80..c6c1666ac120 100644 --- a/drivers/gpu/drm/amd/include/atomfirmware.h +++ b/drivers/gpu/drm/amd/include/atomfirmware.h @@ -1219,6 +1219,41 @@ struct atom_gfx_info_v2_3 { uint32_t rm21_sram_vmin_value; }; +struct atom_gfx_info_v2_4 { + struct atom_common_table_header table_header; + uint8_t gfxip_min_ver; + uint8_t gfxip_max_ver; + uint8_t gc_num_se; + uint8_t max_tile_pipes; + uint8_t gc_num_cu_per_sh; + uint8_t gc_num_sh_per_se; + uint8_t gc_num_rb_per_se; + uint8_t gc_num_tccs; + uint32_t regaddr_cp_dma_src_addr; + uint32_t regaddr_cp_dma_src_addr_hi; + uint32_t regaddr_cp_dma_dst_addr; + uint32_t regaddr_cp_dma_dst_addr_hi; + uint32_t regaddr_cp_dma_command; + uint32_t regaddr_cp_status; + uint32_t regaddr_rlc_gpu_clock_32; + uint32_t rlc_gpu_timer_refclk; + uint8_t active_cu_per_sh; + uint8_t active_rb_per_se; + uint16_t gcgoldenoffset; + uint16_t gc_num_gprs; + uint16_t gc_gsprim_buff_depth; + uint16_t gc_parameter_cache_depth; + uint16_t gc_wave_size; + uint16_t gc_max_waves_per_simd; + uint16_t gc_lds_size; + uint8_t gc_num_max_gs_thds; + uint8_t gc_gs_table_depth; + uint8_t gc_double_offchip_lds_buffer; + uint8_t gc_max_scratch_slots_per_cu; + uint32_t sram_rm_fuses_val; + uint32_t sram_custom_rm_fuses_val; +}; + /* *************************************************************************** Data Table smu_info structure diff --git a/drivers/gpu/drm/amd/include/soc15_ih_clientid.h b/drivers/gpu/drm/amd/include/soc15_ih_clientid.h index a12d4f27cfa4..12e196c15bbe 100644 --- a/drivers/gpu/drm/amd/include/soc15_ih_clientid.h +++ b/drivers/gpu/drm/amd/include/soc15_ih_clientid.h @@ -43,6 +43,7 @@ enum soc15_ih_clientid { SOC15_IH_CLIENTID_SE2SH = 0x0c, SOC15_IH_CLIENTID_SE3SH = 0x0d, SOC15_IH_CLIENTID_SYSHUB = 0x0e, + SOC15_IH_CLIENTID_UVD1 = 0x0e, SOC15_IH_CLIENTID_THM = 0x0f, SOC15_IH_CLIENTID_UVD = 0x10, SOC15_IH_CLIENTID_VCE0 = 0x11, diff --git a/drivers/gpu/drm/amd/include/vega20_ip_offset.h b/drivers/gpu/drm/amd/include/vega20_ip_offset.h new file mode 100644 index 000000000000..2a2a9cc8bedb --- /dev/null +++ b/drivers/gpu/drm/amd/include/vega20_ip_offset.h @@ -0,0 +1,1051 @@ +/* + * Copyright (C) 2018 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included + * in all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS + * OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN + * AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. + */ +#ifndef _vega20_ip_offset_HEADER +#define _vega20_ip_offset_HEADER + +#define MAX_INSTANCE 6 +#define MAX_SEGMENT 6 + + +struct IP_BASE_INSTANCE +{ + unsigned int segment[MAX_SEGMENT]; +}; + +struct IP_BASE +{ + struct IP_BASE_INSTANCE instance[MAX_INSTANCE]; +}; + + +static const struct IP_BASE ATHUB_BASE ={ { { { 0x00000C20, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } } } }; +static const struct IP_BASE CLK_BASE ={ { { { 0x00016C00, 0x00016E00, 0x00017000, 0x00017200, 0x0001B000, 0x0001B200 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } } } }; +static const struct IP_BASE DCE_BASE ={ { { { 0x00000012, 0x000000C0, 0x000034C0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } } } }; +static const struct IP_BASE DF_BASE ={ { { { 0x00007000, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } } } }; +static const struct IP_BASE FUSE_BASE ={ { { { 0x00017400, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } } } }; +static const struct IP_BASE GC_BASE ={ { { { 0x00002000, 0x0000A000, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } } } }; +static const struct IP_BASE HDP_BASE ={ { { { 0x00000F20, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } } } }; +static const struct IP_BASE MMHUB_BASE ={ { { { 0x0001A000, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } } } }; +static const struct IP_BASE MP0_BASE ={ { { { 0x00016000, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } } } }; +static const struct IP_BASE MP1_BASE ={ { { { 0x00016000, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } } } }; +static const struct IP_BASE NBIO_BASE ={ { { { 0x00000000, 0x00000014, 0x00000D20, 0x00010400, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } } } }; +static const struct IP_BASE OSSSYS_BASE ={ { { { 0x000010A0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } } } }; +static const struct IP_BASE SDMA0_BASE ={ { { { 0x00001260, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } } } }; +static const struct IP_BASE SDMA1_BASE ={ { { { 0x00001860, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } } } }; +static const struct IP_BASE SMUIO_BASE ={ { { { 0x00016800, 0x00016A00, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } } } }; +static const struct IP_BASE THM_BASE ={ { { { 0x00016600, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } } } }; +static const struct IP_BASE UMC_BASE ={ { { { 0x00014000, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } } } }; +static const struct IP_BASE UVD_BASE ={ { { { 0x00007800, 0x00007E00, 0, 0, 0, 0 } }, + { { 0, 0x00009000, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } } } }; +/* Adjust VCE_BASE to make vce_4_1 use vce_4_0 offset header files*/ +static const struct IP_BASE VCE_BASE ={ { { { 0x00007E00/* 0x00008800 */, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } } } }; +static const struct IP_BASE XDMA_BASE ={ { { { 0x00003400, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } } } }; +static const struct IP_BASE RSMU_BASE ={ { { { 0x00012000, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } } } }; + + +#define ATHUB_BASE__INST0_SEG0 0x00000C20 +#define ATHUB_BASE__INST0_SEG1 0 +#define ATHUB_BASE__INST0_SEG2 0 +#define ATHUB_BASE__INST0_SEG3 0 +#define ATHUB_BASE__INST0_SEG4 0 +#define ATHUB_BASE__INST0_SEG5 0 + +#define ATHUB_BASE__INST1_SEG0 0 +#define ATHUB_BASE__INST1_SEG1 0 +#define ATHUB_BASE__INST1_SEG2 0 +#define ATHUB_BASE__INST1_SEG3 0 +#define ATHUB_BASE__INST1_SEG4 0 +#define ATHUB_BASE__INST1_SEG5 0 + +#define ATHUB_BASE__INST2_SEG0 0 +#define ATHUB_BASE__INST2_SEG1 0 +#define ATHUB_BASE__INST2_SEG2 0 +#define ATHUB_BASE__INST2_SEG3 0 +#define ATHUB_BASE__INST2_SEG4 0 +#define ATHUB_BASE__INST2_SEG5 0 + +#define ATHUB_BASE__INST3_SEG0 0 +#define ATHUB_BASE__INST3_SEG1 0 +#define ATHUB_BASE__INST3_SEG2 0 +#define ATHUB_BASE__INST3_SEG3 0 +#define ATHUB_BASE__INST3_SEG4 0 +#define ATHUB_BASE__INST3_SEG5 0 + +#define ATHUB_BASE__INST4_SEG0 0 +#define ATHUB_BASE__INST4_SEG1 0 +#define ATHUB_BASE__INST4_SEG2 0 +#define ATHUB_BASE__INST4_SEG3 0 +#define ATHUB_BASE__INST4_SEG4 0 +#define ATHUB_BASE__INST4_SEG5 0 + +#define ATHUB_BASE__INST5_SEG0 0 +#define ATHUB_BASE__INST5_SEG1 0 +#define ATHUB_BASE__INST5_SEG2 0 +#define ATHUB_BASE__INST5_SEG3 0 +#define ATHUB_BASE__INST5_SEG4 0 +#define ATHUB_BASE__INST5_SEG5 0 + +#define CLK_BASE__INST0_SEG0 0x00016C00 +#define CLK_BASE__INST0_SEG1 0x00016E00 +#define CLK_BASE__INST0_SEG2 0x00017000 +#define CLK_BASE__INST0_SEG3 0x00017200 +#define CLK_BASE__INST0_SEG4 0x0001B000 +#define CLK_BASE__INST0_SEG5 0x0001B200 + +#define CLK_BASE__INST1_SEG0 0 +#define CLK_BASE__INST1_SEG1 0 +#define CLK_BASE__INST1_SEG2 0 +#define CLK_BASE__INST1_SEG3 0 +#define CLK_BASE__INST1_SEG4 0 +#define CLK_BASE__INST1_SEG5 0 + +#define CLK_BASE__INST2_SEG0 0 +#define CLK_BASE__INST2_SEG1 0 +#define CLK_BASE__INST2_SEG2 0 +#define CLK_BASE__INST2_SEG3 0 +#define CLK_BASE__INST2_SEG4 0 +#define CLK_BASE__INST2_SEG5 0 + +#define CLK_BASE__INST3_SEG0 0 +#define CLK_BASE__INST3_SEG1 0 +#define CLK_BASE__INST3_SEG2 0 +#define CLK_BASE__INST3_SEG3 0 +#define CLK_BASE__INST3_SEG4 0 +#define CLK_BASE__INST3_SEG5 0 + +#define CLK_BASE__INST4_SEG0 0 +#define CLK_BASE__INST4_SEG1 0 +#define CLK_BASE__INST4_SEG2 0 +#define CLK_BASE__INST4_SEG3 0 +#define CLK_BASE__INST4_SEG4 0 +#define CLK_BASE__INST4_SEG5 0 + +#define CLK_BASE__INST5_SEG0 0 +#define CLK_BASE__INST5_SEG1 0 +#define CLK_BASE__INST5_SEG2 0 +#define CLK_BASE__INST5_SEG3 0 +#define CLK_BASE__INST5_SEG4 0 +#define CLK_BASE__INST5_SEG5 0 + +#define DCE_BASE__INST0_SEG0 0x00000012 +#define DCE_BASE__INST0_SEG1 0x000000C0 +#define DCE_BASE__INST0_SEG2 0x000034C0 +#define DCE_BASE__INST0_SEG3 0 +#define DCE_BASE__INST0_SEG4 0 +#define DCE_BASE__INST0_SEG5 0 + +#define DCE_BASE__INST1_SEG0 0 +#define DCE_BASE__INST1_SEG1 0 +#define DCE_BASE__INST1_SEG2 0 +#define DCE_BASE__INST1_SEG3 0 +#define DCE_BASE__INST1_SEG4 0 +#define DCE_BASE__INST1_SEG5 0 + +#define DCE_BASE__INST2_SEG0 0 +#define DCE_BASE__INST2_SEG1 0 +#define DCE_BASE__INST2_SEG2 0 +#define DCE_BASE__INST2_SEG3 0 +#define DCE_BASE__INST2_SEG4 0 +#define DCE_BASE__INST2_SEG5 0 + +#define DCE_BASE__INST3_SEG0 0 +#define DCE_BASE__INST3_SEG1 0 +#define DCE_BASE__INST3_SEG2 0 +#define DCE_BASE__INST3_SEG3 0 +#define DCE_BASE__INST3_SEG4 0 +#define DCE_BASE__INST3_SEG5 0 + +#define DCE_BASE__INST4_SEG0 0 +#define DCE_BASE__INST4_SEG1 0 +#define DCE_BASE__INST4_SEG2 0 +#define DCE_BASE__INST4_SEG3 0 +#define DCE_BASE__INST4_SEG4 0 +#define DCE_BASE__INST4_SEG5 0 + +#define DCE_BASE__INST5_SEG0 0 +#define DCE_BASE__INST5_SEG1 0 +#define DCE_BASE__INST5_SEG2 0 +#define DCE_BASE__INST5_SEG3 0 +#define DCE_BASE__INST5_SEG4 0 +#define DCE_BASE__INST5_SEG5 0 + +#define DF_BASE__INST0_SEG0 0x00007000 +#define DF_BASE__INST0_SEG1 0 +#define DF_BASE__INST0_SEG2 0 +#define DF_BASE__INST0_SEG3 0 +#define DF_BASE__INST0_SEG4 0 +#define DF_BASE__INST0_SEG5 0 + +#define DF_BASE__INST1_SEG0 0 +#define DF_BASE__INST1_SEG1 0 +#define DF_BASE__INST1_SEG2 0 +#define DF_BASE__INST1_SEG3 0 +#define DF_BASE__INST1_SEG4 0 +#define DF_BASE__INST1_SEG5 0 + +#define DF_BASE__INST2_SEG0 0 +#define DF_BASE__INST2_SEG1 0 +#define DF_BASE__INST2_SEG2 0 +#define DF_BASE__INST2_SEG3 0 +#define DF_BASE__INST2_SEG4 0 +#define DF_BASE__INST2_SEG5 0 + +#define DF_BASE__INST3_SEG0 0 +#define DF_BASE__INST3_SEG1 0 +#define DF_BASE__INST3_SEG2 0 +#define DF_BASE__INST3_SEG3 0 +#define DF_BASE__INST3_SEG4 0 +#define DF_BASE__INST3_SEG5 0 + +#define DF_BASE__INST4_SEG0 0 +#define DF_BASE__INST4_SEG1 0 +#define DF_BASE__INST4_SEG2 0 +#define DF_BASE__INST4_SEG3 0 +#define DF_BASE__INST4_SEG4 0 +#define DF_BASE__INST4_SEG5 0 + +#define DF_BASE__INST5_SEG0 0 +#define DF_BASE__INST5_SEG1 0 +#define DF_BASE__INST5_SEG2 0 +#define DF_BASE__INST5_SEG3 0 +#define DF_BASE__INST5_SEG4 0 +#define DF_BASE__INST5_SEG5 0 + +#define FUSE_BASE__INST0_SEG0 0x00017400 +#define FUSE_BASE__INST0_SEG1 0 +#define FUSE_BASE__INST0_SEG2 0 +#define FUSE_BASE__INST0_SEG3 0 +#define FUSE_BASE__INST0_SEG4 0 +#define FUSE_BASE__INST0_SEG5 0 + +#define FUSE_BASE__INST1_SEG0 0 +#define FUSE_BASE__INST1_SEG1 0 +#define FUSE_BASE__INST1_SEG2 0 +#define FUSE_BASE__INST1_SEG3 0 +#define FUSE_BASE__INST1_SEG4 0 +#define FUSE_BASE__INST1_SEG5 0 + +#define FUSE_BASE__INST2_SEG0 0 +#define FUSE_BASE__INST2_SEG1 0 +#define FUSE_BASE__INST2_SEG2 0 +#define FUSE_BASE__INST2_SEG3 0 +#define FUSE_BASE__INST2_SEG4 0 +#define FUSE_BASE__INST2_SEG5 0 + +#define FUSE_BASE__INST3_SEG0 0 +#define FUSE_BASE__INST3_SEG1 0 +#define FUSE_BASE__INST3_SEG2 0 +#define FUSE_BASE__INST3_SEG3 0 +#define FUSE_BASE__INST3_SEG4 0 +#define FUSE_BASE__INST3_SEG5 0 + +#define FUSE_BASE__INST4_SEG0 0 +#define FUSE_BASE__INST4_SEG1 0 +#define FUSE_BASE__INST4_SEG2 0 +#define FUSE_BASE__INST4_SEG3 0 +#define FUSE_BASE__INST4_SEG4 0 +#define FUSE_BASE__INST4_SEG5 0 + +#define FUSE_BASE__INST5_SEG0 0 +#define FUSE_BASE__INST5_SEG1 0 +#define FUSE_BASE__INST5_SEG2 0 +#define FUSE_BASE__INST5_SEG3 0 +#define FUSE_BASE__INST5_SEG4 0 +#define FUSE_BASE__INST5_SEG5 0 + +#define GC_BASE__INST0_SEG0 0x00002000 +#define GC_BASE__INST0_SEG1 0x0000A000 +#define GC_BASE__INST0_SEG2 0 +#define GC_BASE__INST0_SEG3 0 +#define GC_BASE__INST0_SEG4 0 +#define GC_BASE__INST0_SEG5 0 + +#define GC_BASE__INST1_SEG0 0 +#define GC_BASE__INST1_SEG1 0 +#define GC_BASE__INST1_SEG2 0 +#define GC_BASE__INST1_SEG3 0 +#define GC_BASE__INST1_SEG4 0 +#define GC_BASE__INST1_SEG5 0 + +#define GC_BASE__INST2_SEG0 0 +#define GC_BASE__INST2_SEG1 0 +#define GC_BASE__INST2_SEG2 0 +#define GC_BASE__INST2_SEG3 0 +#define GC_BASE__INST2_SEG4 0 +#define GC_BASE__INST2_SEG5 0 + +#define GC_BASE__INST3_SEG0 0 +#define GC_BASE__INST3_SEG1 0 +#define GC_BASE__INST3_SEG2 0 +#define GC_BASE__INST3_SEG3 0 +#define GC_BASE__INST3_SEG4 0 +#define GC_BASE__INST3_SEG5 0 + +#define GC_BASE__INST4_SEG0 0 +#define GC_BASE__INST4_SEG1 0 +#define GC_BASE__INST4_SEG2 0 +#define GC_BASE__INST4_SEG3 0 +#define GC_BASE__INST4_SEG4 0 +#define GC_BASE__INST4_SEG5 0 + +#define GC_BASE__INST5_SEG0 0 +#define GC_BASE__INST5_SEG1 0 +#define GC_BASE__INST5_SEG2 0 +#define GC_BASE__INST5_SEG3 0 +#define GC_BASE__INST5_SEG4 0 +#define GC_BASE__INST5_SEG5 0 + +#define HDP_BASE__INST0_SEG0 0x00000F20 +#define HDP_BASE__INST0_SEG1 0 +#define HDP_BASE__INST0_SEG2 0 +#define HDP_BASE__INST0_SEG3 0 +#define HDP_BASE__INST0_SEG4 0 +#define HDP_BASE__INST0_SEG5 0 + +#define HDP_BASE__INST1_SEG0 0 +#define HDP_BASE__INST1_SEG1 0 +#define HDP_BASE__INST1_SEG2 0 +#define HDP_BASE__INST1_SEG3 0 +#define HDP_BASE__INST1_SEG4 0 +#define HDP_BASE__INST1_SEG5 0 + +#define HDP_BASE__INST2_SEG0 0 +#define HDP_BASE__INST2_SEG1 0 +#define HDP_BASE__INST2_SEG2 0 +#define HDP_BASE__INST2_SEG3 0 +#define HDP_BASE__INST2_SEG4 0 +#define HDP_BASE__INST2_SEG5 0 + +#define HDP_BASE__INST3_SEG0 0 +#define HDP_BASE__INST3_SEG1 0 +#define HDP_BASE__INST3_SEG2 0 +#define HDP_BASE__INST3_SEG3 0 +#define HDP_BASE__INST3_SEG4 0 +#define HDP_BASE__INST3_SEG5 0 + +#define HDP_BASE__INST4_SEG0 0 +#define HDP_BASE__INST4_SEG1 0 +#define HDP_BASE__INST4_SEG2 0 +#define HDP_BASE__INST4_SEG3 0 +#define HDP_BASE__INST4_SEG4 0 +#define HDP_BASE__INST4_SEG5 0 + +#define HDP_BASE__INST5_SEG0 0 +#define HDP_BASE__INST5_SEG1 0 +#define HDP_BASE__INST5_SEG2 0 +#define HDP_BASE__INST5_SEG3 0 +#define HDP_BASE__INST5_SEG4 0 +#define HDP_BASE__INST5_SEG5 0 + +#define MMHUB_BASE__INST0_SEG0 0x0001A000 +#define MMHUB_BASE__INST0_SEG1 0 +#define MMHUB_BASE__INST0_SEG2 0 +#define MMHUB_BASE__INST0_SEG3 0 +#define MMHUB_BASE__INST0_SEG4 0 +#define MMHUB_BASE__INST0_SEG5 0 + +#define MMHUB_BASE__INST1_SEG0 0 +#define MMHUB_BASE__INST1_SEG1 0 +#define MMHUB_BASE__INST1_SEG2 0 +#define MMHUB_BASE__INST1_SEG3 0 +#define MMHUB_BASE__INST1_SEG4 0 +#define MMHUB_BASE__INST1_SEG5 0 + +#define MMHUB_BASE__INST2_SEG0 0 +#define MMHUB_BASE__INST2_SEG1 0 +#define MMHUB_BASE__INST2_SEG2 0 +#define MMHUB_BASE__INST2_SEG3 0 +#define MMHUB_BASE__INST2_SEG4 0 +#define MMHUB_BASE__INST2_SEG5 0 + +#define MMHUB_BASE__INST3_SEG0 0 +#define MMHUB_BASE__INST3_SEG1 0 +#define MMHUB_BASE__INST3_SEG2 0 +#define MMHUB_BASE__INST3_SEG3 0 +#define MMHUB_BASE__INST3_SEG4 0 +#define MMHUB_BASE__INST3_SEG5 0 + +#define MMHUB_BASE__INST4_SEG0 0 +#define MMHUB_BASE__INST4_SEG1 0 +#define MMHUB_BASE__INST4_SEG2 0 +#define MMHUB_BASE__INST4_SEG3 0 +#define MMHUB_BASE__INST4_SEG4 0 +#define MMHUB_BASE__INST4_SEG5 0 + +#define MMHUB_BASE__INST5_SEG0 0 +#define MMHUB_BASE__INST5_SEG1 0 +#define MMHUB_BASE__INST5_SEG2 0 +#define MMHUB_BASE__INST5_SEG3 0 +#define MMHUB_BASE__INST5_SEG4 0 +#define MMHUB_BASE__INST5_SEG5 0 + +#define MP0_BASE__INST0_SEG0 0x00016000 +#define MP0_BASE__INST0_SEG1 0 +#define MP0_BASE__INST0_SEG2 0 +#define MP0_BASE__INST0_SEG3 0 +#define MP0_BASE__INST0_SEG4 0 +#define MP0_BASE__INST0_SEG5 0 + +#define MP0_BASE__INST1_SEG0 0 +#define MP0_BASE__INST1_SEG1 0 +#define MP0_BASE__INST1_SEG2 0 +#define MP0_BASE__INST1_SEG3 0 +#define MP0_BASE__INST1_SEG4 0 +#define MP0_BASE__INST1_SEG5 0 + +#define MP0_BASE__INST2_SEG0 0 +#define MP0_BASE__INST2_SEG1 0 +#define MP0_BASE__INST2_SEG2 0 +#define MP0_BASE__INST2_SEG3 0 +#define MP0_BASE__INST2_SEG4 0 +#define MP0_BASE__INST2_SEG5 0 + +#define MP0_BASE__INST3_SEG0 0 +#define MP0_BASE__INST3_SEG1 0 +#define MP0_BASE__INST3_SEG2 0 +#define MP0_BASE__INST3_SEG3 0 +#define MP0_BASE__INST3_SEG4 0 +#define MP0_BASE__INST3_SEG5 0 + +#define MP0_BASE__INST4_SEG0 0 +#define MP0_BASE__INST4_SEG1 0 +#define MP0_BASE__INST4_SEG2 0 +#define MP0_BASE__INST4_SEG3 0 +#define MP0_BASE__INST4_SEG4 0 +#define MP0_BASE__INST4_SEG5 0 + +#define MP0_BASE__INST5_SEG0 0 +#define MP0_BASE__INST5_SEG1 0 +#define MP0_BASE__INST5_SEG2 0 +#define MP0_BASE__INST5_SEG3 0 +#define MP0_BASE__INST5_SEG4 0 +#define MP0_BASE__INST5_SEG5 0 + +#define MP1_BASE__INST0_SEG0 0x00016000 +#define MP1_BASE__INST0_SEG1 0 +#define MP1_BASE__INST0_SEG2 0 +#define MP1_BASE__INST0_SEG3 0 +#define MP1_BASE__INST0_SEG4 0 +#define MP1_BASE__INST0_SEG5 0 + +#define MP1_BASE__INST1_SEG0 0 +#define MP1_BASE__INST1_SEG1 0 +#define MP1_BASE__INST1_SEG2 0 +#define MP1_BASE__INST1_SEG3 0 +#define MP1_BASE__INST1_SEG4 0 +#define MP1_BASE__INST1_SEG5 0 + +#define MP1_BASE__INST2_SEG0 0 +#define MP1_BASE__INST2_SEG1 0 +#define MP1_BASE__INST2_SEG2 0 +#define MP1_BASE__INST2_SEG3 0 +#define MP1_BASE__INST2_SEG4 0 +#define MP1_BASE__INST2_SEG5 0 + +#define MP1_BASE__INST3_SEG0 0 +#define MP1_BASE__INST3_SEG1 0 +#define MP1_BASE__INST3_SEG2 0 +#define MP1_BASE__INST3_SEG3 0 +#define MP1_BASE__INST3_SEG4 0 +#define MP1_BASE__INST3_SEG5 0 + +#define MP1_BASE__INST4_SEG0 0 +#define MP1_BASE__INST4_SEG1 0 +#define MP1_BASE__INST4_SEG2 0 +#define MP1_BASE__INST4_SEG3 0 +#define MP1_BASE__INST4_SEG4 0 +#define MP1_BASE__INST4_SEG5 0 + +#define MP1_BASE__INST5_SEG0 0 +#define MP1_BASE__INST5_SEG1 0 +#define MP1_BASE__INST5_SEG2 0 +#define MP1_BASE__INST5_SEG3 0 +#define MP1_BASE__INST5_SEG4 0 +#define MP1_BASE__INST5_SEG5 0 + +#define NBIO_BASE__INST0_SEG0 0x00000000 +#define NBIO_BASE__INST0_SEG1 0x00000014 +#define NBIO_BASE__INST0_SEG2 0x00000D20 +#define NBIO_BASE__INST0_SEG3 0x00010400 +#define NBIO_BASE__INST0_SEG4 0 +#define NBIO_BASE__INST0_SEG5 0 + +#define NBIO_BASE__INST1_SEG0 0 +#define NBIO_BASE__INST1_SEG1 0 +#define NBIO_BASE__INST1_SEG2 0 +#define NBIO_BASE__INST1_SEG3 0 +#define NBIO_BASE__INST1_SEG4 0 +#define NBIO_BASE__INST1_SEG5 0 + +#define NBIO_BASE__INST2_SEG0 0 +#define NBIO_BASE__INST2_SEG1 0 +#define NBIO_BASE__INST2_SEG2 0 +#define NBIO_BASE__INST2_SEG3 0 +#define NBIO_BASE__INST2_SEG4 0 +#define NBIO_BASE__INST2_SEG5 0 + +#define NBIO_BASE__INST3_SEG0 0 +#define NBIO_BASE__INST3_SEG1 0 +#define NBIO_BASE__INST3_SEG2 0 +#define NBIO_BASE__INST3_SEG3 0 +#define NBIO_BASE__INST3_SEG4 0 +#define NBIO_BASE__INST3_SEG5 0 + +#define NBIO_BASE__INST4_SEG0 0 +#define NBIO_BASE__INST4_SEG1 0 +#define NBIO_BASE__INST4_SEG2 0 +#define NBIO_BASE__INST4_SEG3 0 +#define NBIO_BASE__INST4_SEG4 0 +#define NBIO_BASE__INST4_SEG5 0 + +#define NBIO_BASE__INST5_SEG0 0 +#define NBIO_BASE__INST5_SEG1 0 +#define NBIO_BASE__INST5_SEG2 0 +#define NBIO_BASE__INST5_SEG3 0 +#define NBIO_BASE__INST5_SEG4 0 +#define NBIO_BASE__INST5_SEG5 0 + +#define OSSSYS_BASE__INST0_SEG0 0x000010A0 +#define OSSSYS_BASE__INST0_SEG1 0 +#define OSSSYS_BASE__INST0_SEG2 0 +#define OSSSYS_BASE__INST0_SEG3 0 +#define OSSSYS_BASE__INST0_SEG4 0 +#define OSSSYS_BASE__INST0_SEG5 0 + +#define OSSSYS_BASE__INST1_SEG0 0 +#define OSSSYS_BASE__INST1_SEG1 0 +#define OSSSYS_BASE__INST1_SEG2 0 +#define OSSSYS_BASE__INST1_SEG3 0 +#define OSSSYS_BASE__INST1_SEG4 0 +#define OSSSYS_BASE__INST1_SEG5 0 + +#define OSSSYS_BASE__INST2_SEG0 0 +#define OSSSYS_BASE__INST2_SEG1 0 +#define OSSSYS_BASE__INST2_SEG2 0 +#define OSSSYS_BASE__INST2_SEG3 0 +#define OSSSYS_BASE__INST2_SEG4 0 +#define OSSSYS_BASE__INST2_SEG5 0 + +#define OSSSYS_BASE__INST3_SEG0 0 +#define OSSSYS_BASE__INST3_SEG1 0 +#define OSSSYS_BASE__INST3_SEG2 0 +#define OSSSYS_BASE__INST3_SEG3 0 +#define OSSSYS_BASE__INST3_SEG4 0 +#define OSSSYS_BASE__INST3_SEG5 0 + +#define OSSSYS_BASE__INST4_SEG0 0 +#define OSSSYS_BASE__INST4_SEG1 0 +#define OSSSYS_BASE__INST4_SEG2 0 +#define OSSSYS_BASE__INST4_SEG3 0 +#define OSSSYS_BASE__INST4_SEG4 0 +#define OSSSYS_BASE__INST4_SEG5 0 + +#define OSSSYS_BASE__INST5_SEG0 0 +#define OSSSYS_BASE__INST5_SEG1 0 +#define OSSSYS_BASE__INST5_SEG2 0 +#define OSSSYS_BASE__INST5_SEG3 0 +#define OSSSYS_BASE__INST5_SEG4 0 +#define OSSSYS_BASE__INST5_SEG5 0 + +#define SDMA0_BASE__INST0_SEG0 0x00001260 +#define SDMA0_BASE__INST0_SEG1 0 +#define SDMA0_BASE__INST0_SEG2 0 +#define SDMA0_BASE__INST0_SEG3 0 +#define SDMA0_BASE__INST0_SEG4 0 +#define SDMA0_BASE__INST0_SEG5 0 + +#define SDMA0_BASE__INST1_SEG0 0 +#define SDMA0_BASE__INST1_SEG1 0 +#define SDMA0_BASE__INST1_SEG2 0 +#define SDMA0_BASE__INST1_SEG3 0 +#define SDMA0_BASE__INST1_SEG4 0 +#define SDMA0_BASE__INST1_SEG5 0 + +#define SDMA0_BASE__INST2_SEG0 0 +#define SDMA0_BASE__INST2_SEG1 0 +#define SDMA0_BASE__INST2_SEG2 0 +#define SDMA0_BASE__INST2_SEG3 0 +#define SDMA0_BASE__INST2_SEG4 0 +#define SDMA0_BASE__INST2_SEG5 0 + +#define SDMA0_BASE__INST3_SEG0 0 +#define SDMA0_BASE__INST3_SEG1 0 +#define SDMA0_BASE__INST3_SEG2 0 +#define SDMA0_BASE__INST3_SEG3 0 +#define SDMA0_BASE__INST3_SEG4 0 +#define SDMA0_BASE__INST3_SEG5 0 + +#define SDMA0_BASE__INST4_SEG0 0 +#define SDMA0_BASE__INST4_SEG1 0 +#define SDMA0_BASE__INST4_SEG2 0 +#define SDMA0_BASE__INST4_SEG3 0 +#define SDMA0_BASE__INST4_SEG4 0 +#define SDMA0_BASE__INST4_SEG5 0 + +#define SDMA0_BASE__INST5_SEG0 0 +#define SDMA0_BASE__INST5_SEG1 0 +#define SDMA0_BASE__INST5_SEG2 0 +#define SDMA0_BASE__INST5_SEG3 0 +#define SDMA0_BASE__INST5_SEG4 0 +#define SDMA0_BASE__INST5_SEG5 0 + +#define SDMA1_BASE__INST0_SEG0 0x00001860 +#define SDMA1_BASE__INST0_SEG1 0 +#define SDMA1_BASE__INST0_SEG2 0 +#define SDMA1_BASE__INST0_SEG3 0 +#define SDMA1_BASE__INST0_SEG4 0 +#define SDMA1_BASE__INST0_SEG5 0 + +#define SDMA1_BASE__INST1_SEG0 0 +#define SDMA1_BASE__INST1_SEG1 0 +#define SDMA1_BASE__INST1_SEG2 0 +#define SDMA1_BASE__INST1_SEG3 0 +#define SDMA1_BASE__INST1_SEG4 0 +#define SDMA1_BASE__INST1_SEG5 0 + +#define SDMA1_BASE__INST2_SEG0 0 +#define SDMA1_BASE__INST2_SEG1 0 +#define SDMA1_BASE__INST2_SEG2 0 +#define SDMA1_BASE__INST2_SEG3 0 +#define SDMA1_BASE__INST2_SEG4 0 +#define SDMA1_BASE__INST2_SEG5 0 + +#define SDMA1_BASE__INST3_SEG0 0 +#define SDMA1_BASE__INST3_SEG1 0 +#define SDMA1_BASE__INST3_SEG2 0 +#define SDMA1_BASE__INST3_SEG3 0 +#define SDMA1_BASE__INST3_SEG4 0 +#define SDMA1_BASE__INST3_SEG5 0 + +#define SDMA1_BASE__INST4_SEG0 0 +#define SDMA1_BASE__INST4_SEG1 0 +#define SDMA1_BASE__INST4_SEG2 0 +#define SDMA1_BASE__INST4_SEG3 0 +#define SDMA1_BASE__INST4_SEG4 0 +#define SDMA1_BASE__INST4_SEG5 0 + +#define SDMA1_BASE__INST5_SEG0 0 +#define SDMA1_BASE__INST5_SEG1 0 +#define SDMA1_BASE__INST5_SEG2 0 +#define SDMA1_BASE__INST5_SEG3 0 +#define SDMA1_BASE__INST5_SEG4 0 +#define SDMA1_BASE__INST5_SEG5 0 + +#define SMUIO_BASE__INST0_SEG0 0x00016800 +#define SMUIO_BASE__INST0_SEG1 0x00016A00 +#define SMUIO_BASE__INST0_SEG2 0 +#define SMUIO_BASE__INST0_SEG3 0 +#define SMUIO_BASE__INST0_SEG4 0 +#define SMUIO_BASE__INST0_SEG5 0 + +#define SMUIO_BASE__INST1_SEG0 0 +#define SMUIO_BASE__INST1_SEG1 0 +#define SMUIO_BASE__INST1_SEG2 0 +#define SMUIO_BASE__INST1_SEG3 0 +#define SMUIO_BASE__INST1_SEG4 0 +#define SMUIO_BASE__INST1_SEG5 0 + +#define SMUIO_BASE__INST2_SEG0 0 +#define SMUIO_BASE__INST2_SEG1 0 +#define SMUIO_BASE__INST2_SEG2 0 +#define SMUIO_BASE__INST2_SEG3 0 +#define SMUIO_BASE__INST2_SEG4 0 +#define SMUIO_BASE__INST2_SEG5 0 + +#define SMUIO_BASE__INST3_SEG0 0 +#define SMUIO_BASE__INST3_SEG1 0 +#define SMUIO_BASE__INST3_SEG2 0 +#define SMUIO_BASE__INST3_SEG3 0 +#define SMUIO_BASE__INST3_SEG4 0 +#define SMUIO_BASE__INST3_SEG5 0 + +#define SMUIO_BASE__INST4_SEG0 0 +#define SMUIO_BASE__INST4_SEG1 0 +#define SMUIO_BASE__INST4_SEG2 0 +#define SMUIO_BASE__INST4_SEG3 0 +#define SMUIO_BASE__INST4_SEG4 0 +#define SMUIO_BASE__INST4_SEG5 0 + +#define SMUIO_BASE__INST5_SEG0 0 +#define SMUIO_BASE__INST5_SEG1 0 +#define SMUIO_BASE__INST5_SEG2 0 +#define SMUIO_BASE__INST5_SEG3 0 +#define SMUIO_BASE__INST5_SEG4 0 +#define SMUIO_BASE__INST5_SEG5 0 + +#define THM_BASE__INST0_SEG0 0x00016600 +#define THM_BASE__INST0_SEG1 0 +#define THM_BASE__INST0_SEG2 0 +#define THM_BASE__INST0_SEG3 0 +#define THM_BASE__INST0_SEG4 0 +#define THM_BASE__INST0_SEG5 0 + +#define THM_BASE__INST1_SEG0 0 +#define THM_BASE__INST1_SEG1 0 +#define THM_BASE__INST1_SEG2 0 +#define THM_BASE__INST1_SEG3 0 +#define THM_BASE__INST1_SEG4 0 +#define THM_BASE__INST1_SEG5 0 + +#define THM_BASE__INST2_SEG0 0 +#define THM_BASE__INST2_SEG1 0 +#define THM_BASE__INST2_SEG2 0 +#define THM_BASE__INST2_SEG3 0 +#define THM_BASE__INST2_SEG4 0 +#define THM_BASE__INST2_SEG5 0 + +#define THM_BASE__INST3_SEG0 0 +#define THM_BASE__INST3_SEG1 0 +#define THM_BASE__INST3_SEG2 0 +#define THM_BASE__INST3_SEG3 0 +#define THM_BASE__INST3_SEG4 0 +#define THM_BASE__INST3_SEG5 0 + +#define THM_BASE__INST4_SEG0 0 +#define THM_BASE__INST4_SEG1 0 +#define THM_BASE__INST4_SEG2 0 +#define THM_BASE__INST4_SEG3 0 +#define THM_BASE__INST4_SEG4 0 +#define THM_BASE__INST4_SEG5 0 + +#define THM_BASE__INST5_SEG0 0 +#define THM_BASE__INST5_SEG1 0 +#define THM_BASE__INST5_SEG2 0 +#define THM_BASE__INST5_SEG3 0 +#define THM_BASE__INST5_SEG4 0 +#define THM_BASE__INST5_SEG5 0 + +#define UMC_BASE__INST0_SEG0 0x00014000 +#define UMC_BASE__INST0_SEG1 0 +#define UMC_BASE__INST0_SEG2 0 +#define UMC_BASE__INST0_SEG3 0 +#define UMC_BASE__INST0_SEG4 0 +#define UMC_BASE__INST0_SEG5 0 + +#define UMC_BASE__INST1_SEG0 0 +#define UMC_BASE__INST1_SEG1 0 +#define UMC_BASE__INST1_SEG2 0 +#define UMC_BASE__INST1_SEG3 0 +#define UMC_BASE__INST1_SEG4 0 +#define UMC_BASE__INST1_SEG5 0 + +#define UMC_BASE__INST2_SEG0 0 +#define UMC_BASE__INST2_SEG1 0 +#define UMC_BASE__INST2_SEG2 0 +#define UMC_BASE__INST2_SEG3 0 +#define UMC_BASE__INST2_SEG4 0 +#define UMC_BASE__INST2_SEG5 0 + +#define UMC_BASE__INST3_SEG0 0 +#define UMC_BASE__INST3_SEG1 0 +#define UMC_BASE__INST3_SEG2 0 +#define UMC_BASE__INST3_SEG3 0 +#define UMC_BASE__INST3_SEG4 0 +#define UMC_BASE__INST3_SEG5 0 + +#define UMC_BASE__INST4_SEG0 0 +#define UMC_BASE__INST4_SEG1 0 +#define UMC_BASE__INST4_SEG2 0 +#define UMC_BASE__INST4_SEG3 0 +#define UMC_BASE__INST4_SEG4 0 +#define UMC_BASE__INST4_SEG5 0 + +#define UMC_BASE__INST5_SEG0 0 +#define UMC_BASE__INST5_SEG1 0 +#define UMC_BASE__INST5_SEG2 0 +#define UMC_BASE__INST5_SEG3 0 +#define UMC_BASE__INST5_SEG4 0 +#define UMC_BASE__INST5_SEG5 0 + +#define UVD_BASE__INST0_SEG0 0x00007800 +#define UVD_BASE__INST0_SEG1 0x00007E00 +#define UVD_BASE__INST0_SEG2 0 +#define UVD_BASE__INST0_SEG3 0 +#define UVD_BASE__INST0_SEG4 0 +#define UVD_BASE__INST0_SEG5 0 + +#define UVD_BASE__INST1_SEG0 0 +#define UVD_BASE__INST1_SEG1 0x00009000 +#define UVD_BASE__INST1_SEG2 0 +#define UVD_BASE__INST1_SEG3 0 +#define UVD_BASE__INST1_SEG4 0 +#define UVD_BASE__INST1_SEG5 0 + +#define UVD_BASE__INST2_SEG0 0 +#define UVD_BASE__INST2_SEG1 0 +#define UVD_BASE__INST2_SEG2 0 +#define UVD_BASE__INST2_SEG3 0 +#define UVD_BASE__INST2_SEG4 0 +#define UVD_BASE__INST2_SEG5 0 + +#define UVD_BASE__INST3_SEG0 0 +#define UVD_BASE__INST3_SEG1 0 +#define UVD_BASE__INST3_SEG2 0 +#define UVD_BASE__INST3_SEG3 0 +#define UVD_BASE__INST3_SEG4 0 +#define UVD_BASE__INST3_SEG5 0 + +#define UVD_BASE__INST4_SEG0 0 +#define UVD_BASE__INST4_SEG1 0 +#define UVD_BASE__INST4_SEG2 0 +#define UVD_BASE__INST4_SEG3 0 +#define UVD_BASE__INST4_SEG4 0 +#define UVD_BASE__INST4_SEG5 0 + +#define UVD_BASE__INST5_SEG0 0 +#define UVD_BASE__INST5_SEG1 0 +#define UVD_BASE__INST5_SEG2 0 +#define UVD_BASE__INST5_SEG3 0 +#define UVD_BASE__INST5_SEG4 0 +#define UVD_BASE__INST5_SEG5 0 + +#define VCE_BASE__INST0_SEG0 0x00008800 +#define VCE_BASE__INST0_SEG1 0 +#define VCE_BASE__INST0_SEG2 0 +#define VCE_BASE__INST0_SEG3 0 +#define VCE_BASE__INST0_SEG4 0 +#define VCE_BASE__INST0_SEG5 0 + +#define VCE_BASE__INST1_SEG0 0 +#define VCE_BASE__INST1_SEG1 0 +#define VCE_BASE__INST1_SEG2 0 +#define VCE_BASE__INST1_SEG3 0 +#define VCE_BASE__INST1_SEG4 0 +#define VCE_BASE__INST1_SEG5 0 + +#define VCE_BASE__INST2_SEG0 0 +#define VCE_BASE__INST2_SEG1 0 +#define VCE_BASE__INST2_SEG2 0 +#define VCE_BASE__INST2_SEG3 0 +#define VCE_BASE__INST2_SEG4 0 +#define VCE_BASE__INST2_SEG5 0 + +#define VCE_BASE__INST3_SEG0 0 +#define VCE_BASE__INST3_SEG1 0 +#define VCE_BASE__INST3_SEG2 0 +#define VCE_BASE__INST3_SEG3 0 +#define VCE_BASE__INST3_SEG4 0 +#define VCE_BASE__INST3_SEG5 0 + +#define VCE_BASE__INST4_SEG0 0 +#define VCE_BASE__INST4_SEG1 0 +#define VCE_BASE__INST4_SEG2 0 +#define VCE_BASE__INST4_SEG3 0 +#define VCE_BASE__INST4_SEG4 0 +#define VCE_BASE__INST4_SEG5 0 + +#define VCE_BASE__INST5_SEG0 0 +#define VCE_BASE__INST5_SEG1 0 +#define VCE_BASE__INST5_SEG2 0 +#define VCE_BASE__INST5_SEG3 0 +#define VCE_BASE__INST5_SEG4 0 +#define VCE_BASE__INST5_SEG5 0 + +#define XDMA_BASE__INST0_SEG0 0x00003400 +#define XDMA_BASE__INST0_SEG1 0 +#define XDMA_BASE__INST0_SEG2 0 +#define XDMA_BASE__INST0_SEG3 0 +#define XDMA_BASE__INST0_SEG4 0 +#define XDMA_BASE__INST0_SEG5 0 + +#define XDMA_BASE__INST1_SEG0 0 +#define XDMA_BASE__INST1_SEG1 0 +#define XDMA_BASE__INST1_SEG2 0 +#define XDMA_BASE__INST1_SEG3 0 +#define XDMA_BASE__INST1_SEG4 0 +#define XDMA_BASE__INST1_SEG5 0 + +#define XDMA_BASE__INST2_SEG0 0 +#define XDMA_BASE__INST2_SEG1 0 +#define XDMA_BASE__INST2_SEG2 0 +#define XDMA_BASE__INST2_SEG3 0 +#define XDMA_BASE__INST2_SEG4 0 +#define XDMA_BASE__INST2_SEG5 0 + +#define XDMA_BASE__INST3_SEG0 0 +#define XDMA_BASE__INST3_SEG1 0 +#define XDMA_BASE__INST3_SEG2 0 +#define XDMA_BASE__INST3_SEG3 0 +#define XDMA_BASE__INST3_SEG4 0 +#define XDMA_BASE__INST3_SEG5 0 + +#define XDMA_BASE__INST4_SEG0 0 +#define XDMA_BASE__INST4_SEG1 0 +#define XDMA_BASE__INST4_SEG2 0 +#define XDMA_BASE__INST4_SEG3 0 +#define XDMA_BASE__INST4_SEG4 0 +#define XDMA_BASE__INST4_SEG5 0 + +#define XDMA_BASE__INST5_SEG0 0 +#define XDMA_BASE__INST5_SEG1 0 +#define XDMA_BASE__INST5_SEG2 0 +#define XDMA_BASE__INST5_SEG3 0 +#define XDMA_BASE__INST5_SEG4 0 +#define XDMA_BASE__INST5_SEG5 0 + +#define RSMU_BASE__INST0_SEG0 0x00012000 +#define RSMU_BASE__INST0_SEG1 0 +#define RSMU_BASE__INST0_SEG2 0 +#define RSMU_BASE__INST0_SEG3 0 +#define RSMU_BASE__INST0_SEG4 0 +#define RSMU_BASE__INST0_SEG5 0 + +#define RSMU_BASE__INST1_SEG0 0 +#define RSMU_BASE__INST1_SEG1 0 +#define RSMU_BASE__INST1_SEG2 0 +#define RSMU_BASE__INST1_SEG3 0 +#define RSMU_BASE__INST1_SEG4 0 +#define RSMU_BASE__INST1_SEG5 0 + +#define RSMU_BASE__INST2_SEG0 0 +#define RSMU_BASE__INST2_SEG1 0 +#define RSMU_BASE__INST2_SEG2 0 +#define RSMU_BASE__INST2_SEG3 0 +#define RSMU_BASE__INST2_SEG4 0 +#define RSMU_BASE__INST2_SEG5 0 + +#define RSMU_BASE__INST3_SEG0 0 +#define RSMU_BASE__INST3_SEG1 0 +#define RSMU_BASE__INST3_SEG2 0 +#define RSMU_BASE__INST3_SEG3 0 +#define RSMU_BASE__INST3_SEG4 0 +#define RSMU_BASE__INST3_SEG5 0 + +#define RSMU_BASE__INST4_SEG0 0 +#define RSMU_BASE__INST4_SEG1 0 +#define RSMU_BASE__INST4_SEG2 0 +#define RSMU_BASE__INST4_SEG3 0 +#define RSMU_BASE__INST4_SEG4 0 +#define RSMU_BASE__INST4_SEG5 0 + +#define RSMU_BASE__INST5_SEG0 0 +#define RSMU_BASE__INST5_SEG1 0 +#define RSMU_BASE__INST5_SEG2 0 +#define RSMU_BASE__INST5_SEG3 0 +#define RSMU_BASE__INST5_SEG4 0 +#define RSMU_BASE__INST5_SEG5 0 + +#endif + diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/hardwaremanager.c b/drivers/gpu/drm/amd/powerplay/hwmgr/hardwaremanager.c index e411012b3dcb..a0bb921fac22 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/hardwaremanager.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/hardwaremanager.c @@ -132,6 +132,15 @@ int phm_apply_state_adjust_rules(struct pp_hwmgr *hwmgr, return 0; } +int phm_apply_clock_adjust_rules(struct pp_hwmgr *hwmgr) +{ + PHM_FUNC_CHECK(hwmgr); + + if (hwmgr->hwmgr_func->apply_clocks_adjust_rules != NULL) + return hwmgr->hwmgr_func->apply_clocks_adjust_rules(hwmgr); + return 0; +} + int phm_powerdown_uvd(struct pp_hwmgr *hwmgr) { PHM_FUNC_CHECK(hwmgr); @@ -161,6 +170,16 @@ int phm_disable_clock_power_gatings(struct pp_hwmgr *hwmgr) return 0; } +int phm_pre_display_configuration_changed(struct pp_hwmgr *hwmgr) +{ + PHM_FUNC_CHECK(hwmgr); + + if (NULL != hwmgr->hwmgr_func->pre_display_config_changed) + hwmgr->hwmgr_func->pre_display_config_changed(hwmgr); + + return 0; + +} int phm_display_configuration_changed(struct pp_hwmgr *hwmgr) { diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/hwmgr.c index 71b42331f185..e63bc47dc715 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/hwmgr.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/hwmgr.c @@ -151,6 +151,7 @@ int hwmgr_early_init(struct pp_hwmgr *hwmgr) hwmgr->feature_mask &= ~PP_GFXOFF_MASK; switch (hwmgr->chip_id) { case CHIP_VEGA10: + case CHIP_VEGA20: hwmgr->smumgr_funcs = &vega10_smu_funcs; vega10_hwmgr_init(hwmgr); break; diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/pp_psm.c b/drivers/gpu/drm/amd/powerplay/hwmgr/pp_psm.c index 308bff2b5d1d..0af13c154328 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/pp_psm.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/pp_psm.c @@ -265,6 +265,15 @@ int psm_adjust_power_state_dynamic(struct pp_hwmgr *hwmgr, bool skip, if (skip) return 0; + if (!hwmgr->ps) + /* + * for vega12/vega20 which does not support power state manager + * DAL clock limits should also be honoured + */ + phm_apply_clock_adjust_rules(hwmgr); + + phm_pre_display_configuration_changed(hwmgr); + phm_display_configuration_changed(hwmgr); if (hwmgr->ps) diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/ppatomctrl.c b/drivers/gpu/drm/amd/powerplay/hwmgr/ppatomctrl.c index cf99c5eaf080..7047e29755c3 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/ppatomctrl.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/ppatomctrl.c @@ -320,7 +320,7 @@ int atomctrl_get_memory_pll_dividers_ai(struct pp_hwmgr *hwmgr, pp_atomctrl_memory_clock_param_ai *mpll_param) { struct amdgpu_device *adev = hwmgr->adev; - COMPUTE_MEMORY_CLOCK_PARAM_PARAMETERS_V2_3 mpll_parameters = {0}; + COMPUTE_MEMORY_CLOCK_PARAM_PARAMETERS_V2_3 mpll_parameters = {{0}, 0, 0}; int result; mpll_parameters.ulClock.ulClock = cpu_to_le32(clock_value); @@ -1104,10 +1104,8 @@ int atomctrl_get_voltage_evv_on_sclk( GetIndexIntoMasterTable(COMMAND, GetVoltageInfo), (uint32_t *)&get_voltage_info_param_space); - if (0 != result) - return result; - - *voltage = le16_to_cpu(((GET_EVV_VOLTAGE_INFO_OUTPUT_PARAMETER_V1_2 *) + *voltage = result ? 0 : + le16_to_cpu(((GET_EVV_VOLTAGE_INFO_OUTPUT_PARAMETER_V1_2 *) (&get_voltage_info_param_space))->usVoltageLevel); return result; @@ -1312,8 +1310,7 @@ int atomctrl_read_efuse(struct pp_hwmgr *hwmgr, uint16_t start_index, result = amdgpu_atom_execute_table(adev->mode_info.atom_context, GetIndexIntoMasterTable(COMMAND, ReadEfuseValue), (uint32_t *)&efuse_param); - if (!result) - *efuse = le32_to_cpu(efuse_param.ulEfuseValue) & mask; + *efuse = result ? 0 : le32_to_cpu(efuse_param.ulEfuseValue) & mask; return result; } @@ -1354,11 +1351,8 @@ int atomctrl_get_voltage_evv_on_sclk_ai(struct pp_hwmgr *hwmgr, uint8_t voltage_ GetIndexIntoMasterTable(COMMAND, GetVoltageInfo), (uint32_t *)&get_voltage_info_param_space); - if (0 != result) - return result; - - *voltage = le32_to_cpu(((GET_EVV_VOLTAGE_INFO_OUTPUT_PARAMETER_V1_3 *) - (&get_voltage_info_param_space))->ulVoltageLevel); + *voltage = result ? 0 : + le32_to_cpu(((GET_EVV_VOLTAGE_INFO_OUTPUT_PARAMETER_V1_3 *)(&get_voltage_info_param_space))->ulVoltageLevel); return result; } @@ -1552,15 +1546,17 @@ void atomctrl_get_voltage_range(struct pp_hwmgr *hwmgr, uint32_t *max_vddc, case CHIP_FIJI: *max_vddc = le32_to_cpu(((ATOM_ASIC_PROFILING_INFO_V3_3 *)profile)->ulMaxVddc/4); *min_vddc = le32_to_cpu(((ATOM_ASIC_PROFILING_INFO_V3_3 *)profile)->ulMinVddc/4); - break; + return; case CHIP_POLARIS11: case CHIP_POLARIS10: case CHIP_POLARIS12: *max_vddc = le32_to_cpu(((ATOM_ASIC_PROFILING_INFO_V3_6 *)profile)->ulMaxVddc/100); *min_vddc = le32_to_cpu(((ATOM_ASIC_PROFILING_INFO_V3_6 *)profile)->ulMinVddc/100); - break; - default: return; + default: + break; } } + *max_vddc = 0; + *min_vddc = 0; } diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu10_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/smu10_hwmgr.c index 2f69bfa478a7..85f84f4d8be5 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/smu10_hwmgr.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/smu10_hwmgr.c @@ -600,7 +600,10 @@ static int smu10_dpm_force_dpm_level(struct pp_hwmgr *hwmgr, data->gfx_min_freq_limit/100); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetHardMinFclkByFreq, + hwmgr->display_config->num_display > 3 ? + SMU10_UMD_PSTATE_PEAK_FCLK : SMU10_UMD_PSTATE_MIN_FCLK); + smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetHardMinSocclkByFreq, SMU10_UMD_PSTATE_MIN_SOCCLK); @@ -1125,6 +1128,23 @@ static int smu10_set_mmhub_powergating_by_smu(struct pp_hwmgr *hwmgr) return smum_send_msg_to_smc(hwmgr, PPSMC_MSG_PowerGateMmHub); } +static void smu10_powergate_vcn(struct pp_hwmgr *hwmgr, bool bgate) +{ + if (bgate) { + amdgpu_device_ip_set_powergating_state(hwmgr->adev, + AMD_IP_BLOCK_TYPE_VCN, + AMD_PG_STATE_GATE); + smum_send_msg_to_smc_with_parameter(hwmgr, + PPSMC_MSG_PowerDownVcn, 0); + } else { + smum_send_msg_to_smc_with_parameter(hwmgr, + PPSMC_MSG_PowerUpVcn, 0); + amdgpu_device_ip_set_powergating_state(hwmgr->adev, + AMD_IP_BLOCK_TYPE_VCN, + AMD_PG_STATE_UNGATE); + } +} + static const struct pp_hwmgr_func smu10_hwmgr_funcs = { .backend_init = smu10_hwmgr_backend_init, .backend_fini = smu10_hwmgr_backend_fini, @@ -1133,7 +1153,7 @@ static const struct pp_hwmgr_func smu10_hwmgr_funcs = { .force_dpm_level = smu10_dpm_force_dpm_level, .get_power_state_size = smu10_get_power_state_size, .powerdown_uvd = NULL, - .powergate_uvd = NULL, + .powergate_uvd = smu10_powergate_vcn, .powergate_vce = NULL, .get_mclk = smu10_dpm_get_mclk, .get_sclk = smu10_dpm_get_sclk, diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c index 8eb3f5176646..45e9b8cb169d 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c @@ -860,7 +860,8 @@ static void smu7_setup_voltage_range_from_vbios(struct pp_hwmgr *hwmgr) struct phm_ppt_v1_clock_voltage_dependency_table *dep_sclk_table; struct phm_ppt_v1_information *table_info = (struct phm_ppt_v1_information *)(hwmgr->pptable); - uint32_t min_vddc, max_vddc; + uint32_t min_vddc = 0; + uint32_t max_vddc = 0; if (!table_info) return; @@ -1018,7 +1019,7 @@ static int smu7_enable_sclk_mclk_dpm(struct pp_hwmgr *hwmgr) struct smu7_hwmgr *data = (struct smu7_hwmgr *)(hwmgr->backend); /* enable SCLK dpm */ - if (!data->sclk_dpm_key_disabled) + if (!data->sclk_dpm_key_disabled) { if (hwmgr->chip_id == CHIP_VEGAM) smu7_disable_sclk_vce_handshake(hwmgr); @@ -1026,6 +1027,7 @@ static int smu7_enable_sclk_mclk_dpm(struct pp_hwmgr *hwmgr) (0 == smum_send_msg_to_smc(hwmgr, PPSMC_MSG_DPM_Enable)), "Failed to enable SCLK DPM during DPM Start Function!", return -EINVAL); + } /* enable MCLK dpm */ if (0 == data->mclk_dpm_key_disabled) { diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_powertune.c b/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_powertune.c index 99b29ff45d91..c952845833d7 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_powertune.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_powertune.c @@ -936,45 +936,49 @@ int smu7_enable_didt_config(struct pp_hwmgr *hwmgr) if (hwmgr->chip_id == CHIP_POLARIS10) { result = smu7_program_pt_config_registers(hwmgr, GCCACConfig_Polaris10); - PP_ASSERT_WITH_CODE((result == 0), "DIDT Config failed.", return result); + PP_ASSERT_WITH_CODE((result == 0), "DIDT Config failed.", goto error); result = smu7_program_pt_config_registers(hwmgr, DIDTConfig_Polaris10); - PP_ASSERT_WITH_CODE((result == 0), "DIDT Config failed.", return result); + PP_ASSERT_WITH_CODE((result == 0), "DIDT Config failed.", goto error); } else if (hwmgr->chip_id == CHIP_POLARIS11) { result = smu7_program_pt_config_registers(hwmgr, GCCACConfig_Polaris11); - PP_ASSERT_WITH_CODE((result == 0), "DIDT Config failed.", return result); + PP_ASSERT_WITH_CODE((result == 0), "DIDT Config failed.", goto error); if (hwmgr->is_kicker) result = smu7_program_pt_config_registers(hwmgr, DIDTConfig_Polaris11_Kicker); else result = smu7_program_pt_config_registers(hwmgr, DIDTConfig_Polaris11); - PP_ASSERT_WITH_CODE((result == 0), "DIDT Config failed.", return result); + PP_ASSERT_WITH_CODE((result == 0), "DIDT Config failed.", goto error); } else if (hwmgr->chip_id == CHIP_POLARIS12) { result = smu7_program_pt_config_registers(hwmgr, GCCACConfig_Polaris11); - PP_ASSERT_WITH_CODE((result == 0), "DIDT Config failed.", return result); + PP_ASSERT_WITH_CODE((result == 0), "DIDT Config failed.", goto error); result = smu7_program_pt_config_registers(hwmgr, DIDTConfig_Polaris12); - PP_ASSERT_WITH_CODE((result == 0), "DIDT Config failed.", return result); + PP_ASSERT_WITH_CODE((result == 0), "DIDT Config failed.", goto error); } else if (hwmgr->chip_id == CHIP_VEGAM) { result = smu7_program_pt_config_registers(hwmgr, GCCACConfig_VegaM); - PP_ASSERT_WITH_CODE((result == 0), "DIDT Config failed.", return result); + PP_ASSERT_WITH_CODE((result == 0), "DIDT Config failed.", goto error); result = smu7_program_pt_config_registers(hwmgr, DIDTConfig_VegaM); - PP_ASSERT_WITH_CODE((result == 0), "DIDT Config failed.", return result); + PP_ASSERT_WITH_CODE((result == 0), "DIDT Config failed.", goto error); } } cgs_write_register(hwmgr->device, mmGRBM_GFX_INDEX, value2); result = smu7_enable_didt(hwmgr, true); - PP_ASSERT_WITH_CODE((result == 0), "EnableDiDt failed.", return result); + PP_ASSERT_WITH_CODE((result == 0), "EnableDiDt failed.", goto error); if (hwmgr->chip_id == CHIP_POLARIS11) { result = smum_send_msg_to_smc(hwmgr, (uint16_t)(PPSMC_MSG_EnableDpmDidt)); PP_ASSERT_WITH_CODE((0 == result), - "Failed to enable DPM DIDT.", return result); + "Failed to enable DPM DIDT.", goto error); } mutex_unlock(&adev->grbm_idx_mutex); adev->gfx.rlc.funcs->exit_safe_mode(adev); } return 0; +error: + mutex_unlock(&adev->grbm_idx_mutex); + adev->gfx.rlc.funcs->exit_safe_mode(adev); + return result; } int smu7_disable_didt_config(struct pp_hwmgr *hwmgr) @@ -992,17 +996,20 @@ int smu7_disable_didt_config(struct pp_hwmgr *hwmgr) result = smu7_enable_didt(hwmgr, false); PP_ASSERT_WITH_CODE((result == 0), "Post DIDT enable clock gating failed.", - return result); + goto error); if (hwmgr->chip_id == CHIP_POLARIS11) { result = smum_send_msg_to_smc(hwmgr, (uint16_t)(PPSMC_MSG_DisableDpmDidt)); PP_ASSERT_WITH_CODE((0 == result), - "Failed to disable DPM DIDT.", return result); + "Failed to disable DPM DIDT.", goto error); } adev->gfx.rlc.funcs->exit_safe_mode(adev); } return 0; +error: + adev->gfx.rlc.funcs->exit_safe_mode(adev); + return result; } int smu7_enable_smc_cac(struct pp_hwmgr *hwmgr) diff --git a/drivers/gpu/drm/amd/powerplay/inc/hardwaremanager.h b/drivers/gpu/drm/amd/powerplay/inc/hardwaremanager.h index 9bb87857a20f..a202247c9894 100644 --- a/drivers/gpu/drm/amd/powerplay/inc/hardwaremanager.h +++ b/drivers/gpu/drm/amd/powerplay/inc/hardwaremanager.h @@ -410,7 +410,10 @@ extern int phm_apply_state_adjust_rules(struct pp_hwmgr *hwmgr, struct pp_power_state *adjusted_ps, const struct pp_power_state *current_ps); +extern int phm_apply_clock_adjust_rules(struct pp_hwmgr *hwmgr); + extern int phm_force_dpm_levels(struct pp_hwmgr *hwmgr, enum amd_dpm_forced_level level); +extern int phm_pre_display_configuration_changed(struct pp_hwmgr *hwmgr); extern int phm_display_configuration_changed(struct pp_hwmgr *hwmgr); extern int phm_notify_smc_display_config_after_ps_adjustment(struct pp_hwmgr *hwmgr); extern int phm_register_irq_handlers(struct pp_hwmgr *hwmgr); diff --git a/drivers/gpu/drm/amd/powerplay/inc/hwmgr.h b/drivers/gpu/drm/amd/powerplay/inc/hwmgr.h index 3c321c7d9626..b99fb8ac822c 100644 --- a/drivers/gpu/drm/amd/powerplay/inc/hwmgr.h +++ b/drivers/gpu/drm/amd/powerplay/inc/hwmgr.h @@ -229,6 +229,8 @@ struct pp_hwmgr_func { struct pp_power_state *prequest_ps, const struct pp_power_state *pcurrent_ps); + int (*apply_clocks_adjust_rules)(struct pp_hwmgr *hwmgr); + int (*force_dpm_level)(struct pp_hwmgr *hw_mgr, enum amd_dpm_forced_level level); @@ -252,6 +254,7 @@ struct pp_hwmgr_func { const void *state); int (*enable_clock_power_gating)(struct pp_hwmgr *hwmgr); int (*notify_smc_display_config_after_ps_adjustment)(struct pp_hwmgr *hwmgr); + int (*pre_display_config_changed)(struct pp_hwmgr *hwmgr); int (*display_config_changed)(struct pp_hwmgr *hwmgr); int (*disable_clock_power_gating)(struct pp_hwmgr *hwmgr); int (*update_clock_gatings)(struct pp_hwmgr *hwmgr, diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/smu7_smumgr.c b/drivers/gpu/drm/amd/powerplay/smumgr/smu7_smumgr.c index 64d33b775906..d644a9bb9078 100644 --- a/drivers/gpu/drm/amd/powerplay/smumgr/smu7_smumgr.c +++ b/drivers/gpu/drm/amd/powerplay/smumgr/smu7_smumgr.c @@ -283,11 +283,9 @@ int smu7_read_smc_sram_dword(struct pp_hwmgr *hwmgr, uint32_t smc_addr, uint32_t result = smu7_set_smc_sram_address(hwmgr, smc_addr, limit); - if (result) - return result; + *value = result ? 0 : cgs_read_register(hwmgr->device, mmSMC_IND_DATA_11); - *value = cgs_read_register(hwmgr->device, mmSMC_IND_DATA_11); - return 0; + return result; } int smu7_write_smc_sram_dword(struct pp_hwmgr *hwmgr, uint32_t smc_addr, uint32_t value, uint32_t limit) diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/smumgr.c b/drivers/gpu/drm/amd/powerplay/smumgr/smumgr.c index ee236dfbf1d6..c9837935f0f5 100644 --- a/drivers/gpu/drm/amd/powerplay/smumgr/smumgr.c +++ b/drivers/gpu/drm/amd/powerplay/smumgr/smumgr.c @@ -45,6 +45,7 @@ MODULE_FIRMWARE("amdgpu/vegam_smc.bin"); MODULE_FIRMWARE("amdgpu/vega10_smc.bin"); MODULE_FIRMWARE("amdgpu/vega10_acg_smc.bin"); MODULE_FIRMWARE("amdgpu/vega12_smc.bin"); +MODULE_FIRMWARE("amdgpu/vega20_smc.bin"); int smum_thermal_avfs_enable(struct pp_hwmgr *hwmgr) { diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/vegam_smumgr.c b/drivers/gpu/drm/amd/powerplay/smumgr/vegam_smumgr.c index c9a563399330..2de48959ac93 100644 --- a/drivers/gpu/drm/amd/powerplay/smumgr/vegam_smumgr.c +++ b/drivers/gpu/drm/amd/powerplay/smumgr/vegam_smumgr.c @@ -1366,10 +1366,12 @@ static int vegam_program_memory_timing_parameters(struct pp_hwmgr *hwmgr) { struct smu7_hwmgr *hw_data = (struct smu7_hwmgr *)(hwmgr->backend); struct vegam_smumgr *smu_data = (struct vegam_smumgr *)(hwmgr->smu_backend); - struct SMU75_Discrete_MCArbDramTimingTable arb_regs = {0}; + struct SMU75_Discrete_MCArbDramTimingTable arb_regs; uint32_t i, j; int result = 0; + memset(&arb_regs, 0, sizeof(SMU75_Discrete_MCArbDramTimingTable)); + for (i = 0; i < hw_data->dpm_table.sclk_table.count; i++) { for (j = 0; j < hw_data->dpm_table.mclk_table.count; j++) { result = vegam_populate_memory_timing_parameters(hwmgr, @@ -2377,6 +2379,5 @@ const struct pp_smumgr_func vegam_smu_funcs = { .update_sclk_threshold = vegam_update_sclk_threshold, .is_hw_avfs_present = vegam_is_hw_avfs_present, .thermal_avfs_enable = vegam_thermal_avfs_enable, - .is_dpm_running = vegam_is_dpm_running, .thermal_setup_fan_table = vegam_thermal_setup_fan_table, }; diff --git a/drivers/gpu/drm/bridge/Kconfig b/drivers/gpu/drm/bridge/Kconfig index 1d75d3a1f951..fa2c7997e2fd 100644 --- a/drivers/gpu/drm/bridge/Kconfig +++ b/drivers/gpu/drm/bridge/Kconfig @@ -84,6 +84,7 @@ config DRM_SIL_SII8620 tristate "Silicon Image SII8620 HDMI/MHL bridge" depends on OF && RC_CORE select DRM_KMS_HELPER + imply EXTCON help Silicon Image SII8620 HDMI/MHL bridge chip driver. diff --git a/drivers/gpu/drm/drm_atomic.c b/drivers/gpu/drm/drm_atomic.c index 26a22f529e9b..895741e9cd7d 100644 --- a/drivers/gpu/drm/drm_atomic.c +++ b/drivers/gpu/drm/drm_atomic.c @@ -155,6 +155,8 @@ void drm_atomic_state_default_clear(struct drm_atomic_state *state) state->connectors[i].state); state->connectors[i].ptr = NULL; state->connectors[i].state = NULL; + state->connectors[i].old_state = NULL; + state->connectors[i].new_state = NULL; drm_connector_put(connector); } @@ -169,6 +171,8 @@ void drm_atomic_state_default_clear(struct drm_atomic_state *state) state->crtcs[i].ptr = NULL; state->crtcs[i].state = NULL; + state->crtcs[i].old_state = NULL; + state->crtcs[i].new_state = NULL; } for (i = 0; i < config->num_total_plane; i++) { @@ -181,6 +185,8 @@ void drm_atomic_state_default_clear(struct drm_atomic_state *state) state->planes[i].state); state->planes[i].ptr = NULL; state->planes[i].state = NULL; + state->planes[i].old_state = NULL; + state->planes[i].new_state = NULL; } for (i = 0; i < state->num_private_objs; i++) { @@ -190,6 +196,8 @@ void drm_atomic_state_default_clear(struct drm_atomic_state *state) state->private_objs[i].state); state->private_objs[i].ptr = NULL; state->private_objs[i].state = NULL; + state->private_objs[i].old_state = NULL; + state->private_objs[i].new_state = NULL; } state->num_private_objs = 0; diff --git a/drivers/gpu/drm/drm_file.c b/drivers/gpu/drm/drm_file.c index e394799979a6..6d9b9453707c 100644 --- a/drivers/gpu/drm/drm_file.c +++ b/drivers/gpu/drm/drm_file.c @@ -212,6 +212,7 @@ static int drm_open_helper(struct file *filp, struct drm_minor *minor) return -ENOMEM; filp->private_data = priv; + filp->f_mode |= FMODE_UNSIGNED_OFFSET; priv->filp = filp; priv->pid = get_pid(task_pid(current)); priv->minor = minor; diff --git a/drivers/gpu/drm/etnaviv/Kconfig b/drivers/gpu/drm/etnaviv/Kconfig index e5bfeca361bd..041a77e400d4 100644 --- a/drivers/gpu/drm/etnaviv/Kconfig +++ b/drivers/gpu/drm/etnaviv/Kconfig @@ -22,11 +22,3 @@ config DRM_ETNAVIV_THERMAL help Compile in support for thermal throttling. Say Y unless you want to risk burning your SoC. - -config DRM_ETNAVIV_REGISTER_LOGGING - bool "enable ETNAVIV register logging" - depends on DRM_ETNAVIV - help - Compile in support for logging register reads/writes in a format - that can be parsed by envytools demsm tool. If enabled, register - logging can be switched on via etnaviv.reglog=y module param. diff --git a/drivers/gpu/drm/etnaviv/etnaviv_buffer.c b/drivers/gpu/drm/etnaviv/etnaviv_buffer.c index bfc6d4aa3b7c..7fea74861a87 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_buffer.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_buffer.c @@ -1,18 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* - * Copyright (C) 2014 Etnaviv Project - * Author: Christian Gmeiner <christian.gmeiner@gmail.com> - * - * This program is free software; you can redistribute it and/or modify it - * under the terms of the GNU General Public License version 2 as published by - * the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License along with - * this program. If not, see <http://www.gnu.org/licenses/>. + * Copyright (C) 2014-2018 Etnaviv Project */ #include "etnaviv_cmdbuf.h" diff --git a/drivers/gpu/drm/etnaviv/etnaviv_cmd_parser.c b/drivers/gpu/drm/etnaviv/etnaviv_cmd_parser.c index 68e6d3772ad8..b106e8b288ad 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_cmd_parser.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_cmd_parser.c @@ -1,17 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* - * Copyright (C) 2015 Etnaviv Project - * - * This program is free software; you can redistribute it and/or modify it - * under the terms of the GNU General Public License version 2 as published by - * the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License along with - * this program. If not, see <http://www.gnu.org/licenses/>. + * Copyright (C) 2015-2018 Etnaviv Project */ #include <linux/kernel.h> diff --git a/drivers/gpu/drm/etnaviv/etnaviv_cmdbuf.c b/drivers/gpu/drm/etnaviv/etnaviv_cmdbuf.c index 3746827f45eb..a3c44f145c1d 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_cmdbuf.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_cmdbuf.c @@ -1,17 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* - * Copyright (C) 2017 Etnaviv Project - * - * This program is free software; you can redistribute it and/or modify it - * under the terms of the GNU General Public License version 2 as published by - * the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License along with - * this program. If not, see <http://www.gnu.org/licenses/>. + * Copyright (C) 2017-2018 Etnaviv Project */ #include <drm/drm_mm.h> diff --git a/drivers/gpu/drm/etnaviv/etnaviv_cmdbuf.h b/drivers/gpu/drm/etnaviv/etnaviv_cmdbuf.h index ddc3f7ea169c..acb68c698363 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_cmdbuf.h +++ b/drivers/gpu/drm/etnaviv/etnaviv_cmdbuf.h @@ -1,17 +1,6 @@ +/* SPDX-License-Identifier: GPL-2.0 */ /* * Copyright (C) 2017 Etnaviv Project - * - * This program is free software; you can redistribute it and/or modify it - * under the terms of the GNU General Public License version 2 as published by - * the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License along with - * this program. If not, see <http://www.gnu.org/licenses/>. */ #ifndef __ETNAVIV_CMDBUF_H__ diff --git a/drivers/gpu/drm/etnaviv/etnaviv_drv.c b/drivers/gpu/drm/etnaviv/etnaviv_drv.c index 23e73c2a19f4..e5013a999147 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_drv.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_drv.c @@ -1,17 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* - * Copyright (C) 2015 Etnaviv Project - * - * This program is free software; you can redistribute it and/or modify it - * under the terms of the GNU General Public License version 2 as published by - * the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License along with - * this program. If not, see <http://www.gnu.org/licenses/>. + * Copyright (C) 2015-2018 Etnaviv Project */ #include <linux/component.h> @@ -25,57 +14,6 @@ #include "etnaviv_mmu.h" #include "etnaviv_perfmon.h" -#ifdef CONFIG_DRM_ETNAVIV_REGISTER_LOGGING -static bool reglog; -MODULE_PARM_DESC(reglog, "Enable register read/write logging"); -module_param(reglog, bool, 0600); -#else -#define reglog 0 -#endif - -void __iomem *etnaviv_ioremap(struct platform_device *pdev, const char *name, - const char *dbgname) -{ - struct resource *res; - void __iomem *ptr; - - if (name) - res = platform_get_resource_byname(pdev, IORESOURCE_MEM, name); - else - res = platform_get_resource(pdev, IORESOURCE_MEM, 0); - - ptr = devm_ioremap_resource(&pdev->dev, res); - if (IS_ERR(ptr)) { - dev_err(&pdev->dev, "failed to ioremap %s: %ld\n", name, - PTR_ERR(ptr)); - return ptr; - } - - if (reglog) - dev_printk(KERN_DEBUG, &pdev->dev, "IO:region %s 0x%p %08zx\n", - dbgname, ptr, (size_t)resource_size(res)); - - return ptr; -} - -void etnaviv_writel(u32 data, void __iomem *addr) -{ - if (reglog) - printk(KERN_DEBUG "IO:W %p %08x\n", addr, data); - - writel(data, addr); -} - -u32 etnaviv_readl(const void __iomem *addr) -{ - u32 val = readl(addr); - - if (reglog) - printk(KERN_DEBUG "IO:R %p %08x\n", addr, val); - - return val; -} - /* * DRM operations: */ diff --git a/drivers/gpu/drm/etnaviv/etnaviv_drv.h b/drivers/gpu/drm/etnaviv/etnaviv_drv.h index ddb17ee565e9..d36c7bbe66db 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_drv.h +++ b/drivers/gpu/drm/etnaviv/etnaviv_drv.h @@ -1,17 +1,6 @@ +/* SPDX-License-Identifier: GPL-2.0 */ /* - * Copyright (C) 2015 Etnaviv Project - * - * This program is free software; you can redistribute it and/or modify it - * under the terms of the GNU General Public License version 2 as published by - * the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License along with - * this program. If not, see <http://www.gnu.org/licenses/>. + * Copyright (C) 2015-2018 Etnaviv Project */ #ifndef __ETNAVIV_DRV_H__ @@ -26,6 +15,7 @@ #include <linux/pm_runtime.h> #include <linux/slab.h> #include <linux/list.h> +#include <linux/time64.h> #include <linux/types.h> #include <linux/sizes.h> @@ -101,11 +91,6 @@ void etnaviv_gem_describe_objects(struct etnaviv_drm_private *priv, struct seq_file *m); #endif -void __iomem *etnaviv_ioremap(struct platform_device *pdev, const char *name, - const char *dbgname); -void etnaviv_writel(u32 data, void __iomem *addr); -u32 etnaviv_readl(const void __iomem *addr); - #define DBG(fmt, ...) DRM_DEBUG(fmt"\n", ##__VA_ARGS__) #define VERB(fmt, ...) if (0) DRM_DEBUG(fmt"\n", ##__VA_ARGS__) @@ -132,19 +117,27 @@ static inline bool fence_after_eq(u32 a, u32 b) return (s32)(a - b) >= 0; } +/* + * Etnaviv timeouts are specified wrt CLOCK_MONOTONIC, not jiffies. + * We need to calculate the timeout in terms of number of jiffies + * between the specified timeout and the current CLOCK_MONOTONIC time. + */ static inline unsigned long etnaviv_timeout_to_jiffies( const struct timespec *timeout) { - unsigned long timeout_jiffies = timespec_to_jiffies(timeout); - unsigned long start_jiffies = jiffies; - unsigned long remaining_jiffies; + struct timespec64 ts, to; + + to = timespec_to_timespec64(*timeout); + + ktime_get_ts64(&ts); + + /* timeouts before "now" have already expired */ + if (timespec64_compare(&to, &ts) <= 0) + return 0; - if (time_after(start_jiffies, timeout_jiffies)) - remaining_jiffies = 0; - else - remaining_jiffies = timeout_jiffies - start_jiffies; + ts = timespec64_sub(to, ts); - return remaining_jiffies; + return timespec64_to_jiffies(&ts); } #endif /* __ETNAVIV_DRV_H__ */ diff --git a/drivers/gpu/drm/etnaviv/etnaviv_dump.c b/drivers/gpu/drm/etnaviv/etnaviv_dump.c index 48aef6cf6a42..9146e30e24a6 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_dump.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_dump.c @@ -1,17 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* - * Copyright (C) 2015 Etnaviv Project - * - * This program is free software; you can redistribute it and/or modify it - * under the terms of the GNU General Public License version 2 as published by - * the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License along with - * this program. If not, see <http://www.gnu.org/licenses/>. + * Copyright (C) 2015-2018 Etnaviv Project */ #include <linux/devcoredump.h> diff --git a/drivers/gpu/drm/etnaviv/etnaviv_dump.h b/drivers/gpu/drm/etnaviv/etnaviv_dump.h index 97f2f8db9133..2d916c2667ee 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_dump.h +++ b/drivers/gpu/drm/etnaviv/etnaviv_dump.h @@ -1,20 +1,8 @@ +/* SPDX-License-Identifier: GPL-2.0 */ /* * Copyright (C) 2015 Etnaviv Project - * - * This program is free software; you can redistribute it and/or modify it - * under the terms of the GNU General Public License version 2 as published by - * the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License along with - * this program. If not, see <http://www.gnu.org/licenses/>. - * - * Etnaviv devcoredump file definitions */ + #ifndef ETNAVIV_DUMP_H #define ETNAVIV_DUMP_H diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gem.c b/drivers/gpu/drm/etnaviv/etnaviv_gem.c index fcc969fa0e69..209ef1274b80 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_gem.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_gem.c @@ -1,17 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* - * Copyright (C) 2015 Etnaviv Project - * - * This program is free software; you can redistribute it and/or modify it - * under the terms of the GNU General Public License version 2 as published by - * the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License along with - * this program. If not, see <http://www.gnu.org/licenses/>. + * Copyright (C) 2015-2018 Etnaviv Project */ #include <linux/spinlock.h> diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gem.h b/drivers/gpu/drm/etnaviv/etnaviv_gem.h index 93e696fcc14f..76079c2291f8 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_gem.h +++ b/drivers/gpu/drm/etnaviv/etnaviv_gem.h @@ -1,17 +1,6 @@ +/* SPDX-License-Identifier: GPL-2.0 */ /* - * Copyright (C) 2015 Etnaviv Project - * - * This program is free software; you can redistribute it and/or modify it - * under the terms of the GNU General Public License version 2 as published by - * the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License along with - * this program. If not, see <http://www.gnu.org/licenses/>. + * Copyright (C) 2015-2018 Etnaviv Project */ #ifndef __ETNAVIV_GEM_H__ diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gem_prime.c b/drivers/gpu/drm/etnaviv/etnaviv_gem_prime.c index 5704305d41e6..0566171f8df2 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_gem_prime.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_gem_prime.c @@ -1,18 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* - * Copyright (C) 2013 Red Hat - * Author: Rob Clark <robdclark@gmail.com> - * - * This program is free software; you can redistribute it and/or modify it - * under the terms of the GNU General Public License version 2 as published by - * the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License along with - * this program. If not, see <http://www.gnu.org/licenses/>. + * Copyright (C) 2014-2018 Etnaviv Project */ #include <linux/dma-buf.h> diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gpu.c b/drivers/gpu/drm/etnaviv/etnaviv_gpu.c index 8a88799bf79b..686f6552db48 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_gpu.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_gpu.c @@ -1,17 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* - * Copyright (C) 2015 Etnaviv Project - * - * This program is free software; you can redistribute it and/or modify it - * under the terms of the GNU General Public License version 2 as published by - * the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License along with - * this program. If not, see <http://www.gnu.org/licenses/>. + * Copyright (C) 2015-2018 Etnaviv Project */ #include <linux/component.h> @@ -1735,6 +1724,7 @@ static int etnaviv_gpu_platform_probe(struct platform_device *pdev) { struct device *dev = &pdev->dev; struct etnaviv_gpu *gpu; + struct resource *res; int err; gpu = devm_kzalloc(dev, sizeof(*gpu), GFP_KERNEL); @@ -1746,7 +1736,8 @@ static int etnaviv_gpu_platform_probe(struct platform_device *pdev) mutex_init(&gpu->fence_idr_lock); /* Map registers: */ - gpu->mmio = etnaviv_ioremap(pdev, NULL, dev_name(gpu->dev)); + res = platform_get_resource(pdev, IORESOURCE_MEM, 0); + gpu->mmio = devm_ioremap_resource(&pdev->dev, res); if (IS_ERR(gpu->mmio)) return PTR_ERR(gpu->mmio); diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gpu.h b/drivers/gpu/drm/etnaviv/etnaviv_gpu.h index 3c3005501846..dd430f0f8ff5 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_gpu.h +++ b/drivers/gpu/drm/etnaviv/etnaviv_gpu.h @@ -1,17 +1,6 @@ +/* SPDX-License-Identifier: GPL-2.0 */ /* - * Copyright (C) 2015 Etnaviv Project - * - * This program is free software; you can redistribute it and/or modify it - * under the terms of the GNU General Public License version 2 as published by - * the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License along with - * this program. If not, see <http://www.gnu.org/licenses/>. + * Copyright (C) 2015-2018 Etnaviv Project */ #ifndef __ETNAVIV_GPU_H__ @@ -161,12 +150,12 @@ struct etnaviv_gpu { static inline void gpu_write(struct etnaviv_gpu *gpu, u32 reg, u32 data) { - etnaviv_writel(data, gpu->mmio + reg); + writel(data, gpu->mmio + reg); } static inline u32 gpu_read(struct etnaviv_gpu *gpu, u32 reg) { - return etnaviv_readl(gpu->mmio + reg); + return readl(gpu->mmio + reg); } static inline bool fence_completed(struct etnaviv_gpu *gpu, u32 fence) diff --git a/drivers/gpu/drm/etnaviv/etnaviv_hwdb.c b/drivers/gpu/drm/etnaviv/etnaviv_hwdb.c index ea08bb38caaf..39b463db76c9 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_hwdb.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_hwdb.c @@ -1,17 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2018 Etnaviv Project - * - * This program is free software; you can redistribute it and/or modify it - * under the terms of the GNU General Public License version 2 as published by - * the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License along with - * this program. If not, see <http://www.gnu.org/licenses/>. */ #include "etnaviv_gpu.h" diff --git a/drivers/gpu/drm/etnaviv/etnaviv_iommu.c b/drivers/gpu/drm/etnaviv/etnaviv_iommu.c index 4b9b11ca6f03..b163bdbcb880 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_iommu.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_iommu.c @@ -1,17 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* - * Copyright (C) 2014 Christian Gmeiner <christian.gmeiner@gmail.com> - * - * This program is free software; you can redistribute it and/or modify it - * under the terms of the GNU General Public License version 2 as published by - * the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License along with - * this program. If not, see <http://www.gnu.org/licenses/>. + * Copyright (C) 2014-2018 Etnaviv Project */ #include <linux/platform_device.h> @@ -47,11 +36,10 @@ static int __etnaviv_iommu_init(struct etnaviv_iommuv1_domain *etnaviv_domain) u32 *p; int i; - etnaviv_domain->base.bad_page_cpu = dma_alloc_coherent( - etnaviv_domain->base.dev, - SZ_4K, - &etnaviv_domain->base.bad_page_dma, - GFP_KERNEL); + etnaviv_domain->base.bad_page_cpu = + dma_alloc_wc(etnaviv_domain->base.dev, SZ_4K, + &etnaviv_domain->base.bad_page_dma, + GFP_KERNEL); if (!etnaviv_domain->base.bad_page_cpu) return -ENOMEM; @@ -59,14 +47,14 @@ static int __etnaviv_iommu_init(struct etnaviv_iommuv1_domain *etnaviv_domain) for (i = 0; i < SZ_4K / 4; i++) *p++ = 0xdead55aa; - etnaviv_domain->pgtable_cpu = - dma_alloc_coherent(etnaviv_domain->base.dev, PT_SIZE, - &etnaviv_domain->pgtable_dma, - GFP_KERNEL); + etnaviv_domain->pgtable_cpu = dma_alloc_wc(etnaviv_domain->base.dev, + PT_SIZE, + &etnaviv_domain->pgtable_dma, + GFP_KERNEL); if (!etnaviv_domain->pgtable_cpu) { - dma_free_coherent(etnaviv_domain->base.dev, SZ_4K, - etnaviv_domain->base.bad_page_cpu, - etnaviv_domain->base.bad_page_dma); + dma_free_wc(etnaviv_domain->base.dev, SZ_4K, + etnaviv_domain->base.bad_page_cpu, + etnaviv_domain->base.bad_page_dma); return -ENOMEM; } @@ -81,13 +69,12 @@ static void etnaviv_iommuv1_domain_free(struct etnaviv_iommu_domain *domain) struct etnaviv_iommuv1_domain *etnaviv_domain = to_etnaviv_domain(domain); - dma_free_coherent(etnaviv_domain->base.dev, PT_SIZE, - etnaviv_domain->pgtable_cpu, - etnaviv_domain->pgtable_dma); + dma_free_wc(etnaviv_domain->base.dev, PT_SIZE, + etnaviv_domain->pgtable_cpu, etnaviv_domain->pgtable_dma); - dma_free_coherent(etnaviv_domain->base.dev, SZ_4K, - etnaviv_domain->base.bad_page_cpu, - etnaviv_domain->base.bad_page_dma); + dma_free_wc(etnaviv_domain->base.dev, SZ_4K, + etnaviv_domain->base.bad_page_cpu, + etnaviv_domain->base.bad_page_dma); kfree(etnaviv_domain); } diff --git a/drivers/gpu/drm/etnaviv/etnaviv_iommu.h b/drivers/gpu/drm/etnaviv/etnaviv_iommu.h index 01d59bf70d78..b279404ce91a 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_iommu.h +++ b/drivers/gpu/drm/etnaviv/etnaviv_iommu.h @@ -1,17 +1,6 @@ +/* SPDX-License-Identifier: GPL-2.0 */ /* - * Copyright (C) 2014 Christian Gmeiner <christian.gmeiner@gmail.com> - * - * This program is free software; you can redistribute it and/or modify it - * under the terms of the GNU General Public License version 2 as published by - * the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License along with - * this program. If not, see <http://www.gnu.org/licenses/>. + * Copyright (C) 2014-2018 Etnaviv Project */ #ifndef __ETNAVIV_IOMMU_H__ diff --git a/drivers/gpu/drm/etnaviv/etnaviv_iommu_v2.c b/drivers/gpu/drm/etnaviv/etnaviv_iommu_v2.c index 9752dbd5d28b..71fbc1f96cb6 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_iommu_v2.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_iommu_v2.c @@ -1,17 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* - * Copyright (C) 2016 Etnaviv Project - * - * This program is free software; you can redistribute it and/or modify it - * under the terms of the GNU General Public License version 2 as published by - * the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License along with - * this program. If not, see <http://www.gnu.org/licenses/>. + * Copyright (C) 2016-2018 Etnaviv Project */ #include <linux/platform_device.h> @@ -47,8 +36,8 @@ struct etnaviv_iommuv2_domain { u32 *mtlb_cpu; dma_addr_t mtlb_dma; /* S(lave) TLB aka second level pagetable */ - u32 *stlb_cpu[1024]; - dma_addr_t stlb_dma[1024]; + u32 *stlb_cpu[MMUv2_MAX_STLB_ENTRIES]; + dma_addr_t stlb_dma[MMUv2_MAX_STLB_ENTRIES]; }; static struct etnaviv_iommuv2_domain * @@ -57,24 +46,54 @@ to_etnaviv_domain(struct etnaviv_iommu_domain *domain) return container_of(domain, struct etnaviv_iommuv2_domain, base); } +static int +etnaviv_iommuv2_ensure_stlb(struct etnaviv_iommuv2_domain *etnaviv_domain, + int stlb) +{ + if (etnaviv_domain->stlb_cpu[stlb]) + return 0; + + etnaviv_domain->stlb_cpu[stlb] = + dma_alloc_wc(etnaviv_domain->base.dev, SZ_4K, + &etnaviv_domain->stlb_dma[stlb], + GFP_KERNEL); + + if (!etnaviv_domain->stlb_cpu[stlb]) + return -ENOMEM; + + memset32(etnaviv_domain->stlb_cpu[stlb], MMUv2_PTE_EXCEPTION, + SZ_4K / sizeof(u32)); + + etnaviv_domain->mtlb_cpu[stlb] = etnaviv_domain->stlb_dma[stlb] | + MMUv2_PTE_PRESENT; + return 0; +} + static int etnaviv_iommuv2_map(struct etnaviv_iommu_domain *domain, unsigned long iova, phys_addr_t paddr, size_t size, int prot) { struct etnaviv_iommuv2_domain *etnaviv_domain = to_etnaviv_domain(domain); - int mtlb_entry, stlb_entry; - u32 entry = (u32)paddr | MMUv2_PTE_PRESENT; + int mtlb_entry, stlb_entry, ret; + u32 entry = lower_32_bits(paddr) | MMUv2_PTE_PRESENT; if (size != SZ_4K) return -EINVAL; + if (IS_ENABLED(CONFIG_PHYS_ADDR_T_64BIT)) + entry |= (upper_32_bits(paddr) & 0xff) << 4; + if (prot & ETNAVIV_PROT_WRITE) entry |= MMUv2_PTE_WRITEABLE; mtlb_entry = (iova & MMUv2_MTLB_MASK) >> MMUv2_MTLB_SHIFT; stlb_entry = (iova & MMUv2_STLB_MASK) >> MMUv2_STLB_SHIFT; + ret = etnaviv_iommuv2_ensure_stlb(etnaviv_domain, mtlb_entry); + if (ret) + return ret; + etnaviv_domain->stlb_cpu[mtlb_entry][stlb_entry] = entry; return 0; @@ -101,14 +120,13 @@ static size_t etnaviv_iommuv2_unmap(struct etnaviv_iommu_domain *domain, static int etnaviv_iommuv2_init(struct etnaviv_iommuv2_domain *etnaviv_domain) { u32 *p; - int ret, i, j; + int ret, i; /* allocate scratch page */ - etnaviv_domain->base.bad_page_cpu = dma_alloc_coherent( - etnaviv_domain->base.dev, - SZ_4K, - &etnaviv_domain->base.bad_page_dma, - GFP_KERNEL); + etnaviv_domain->base.bad_page_cpu = + dma_alloc_wc(etnaviv_domain->base.dev, SZ_4K, + &etnaviv_domain->base.bad_page_dma, + GFP_KERNEL); if (!etnaviv_domain->base.bad_page_cpu) { ret = -ENOMEM; goto fail_mem; @@ -117,67 +135,40 @@ static int etnaviv_iommuv2_init(struct etnaviv_iommuv2_domain *etnaviv_domain) for (i = 0; i < SZ_4K / 4; i++) *p++ = 0xdead55aa; - etnaviv_domain->pta_cpu = dma_alloc_coherent(etnaviv_domain->base.dev, - SZ_4K, - &etnaviv_domain->pta_dma, - GFP_KERNEL); + etnaviv_domain->pta_cpu = dma_alloc_wc(etnaviv_domain->base.dev, + SZ_4K, &etnaviv_domain->pta_dma, + GFP_KERNEL); if (!etnaviv_domain->pta_cpu) { ret = -ENOMEM; goto fail_mem; } - etnaviv_domain->mtlb_cpu = dma_alloc_coherent(etnaviv_domain->base.dev, - SZ_4K, - &etnaviv_domain->mtlb_dma, - GFP_KERNEL); + etnaviv_domain->mtlb_cpu = dma_alloc_wc(etnaviv_domain->base.dev, + SZ_4K, &etnaviv_domain->mtlb_dma, + GFP_KERNEL); if (!etnaviv_domain->mtlb_cpu) { ret = -ENOMEM; goto fail_mem; } - /* pre-populate STLB pages (may want to switch to on-demand later) */ - for (i = 0; i < MMUv2_MAX_STLB_ENTRIES; i++) { - etnaviv_domain->stlb_cpu[i] = - dma_alloc_coherent(etnaviv_domain->base.dev, - SZ_4K, - &etnaviv_domain->stlb_dma[i], - GFP_KERNEL); - if (!etnaviv_domain->stlb_cpu[i]) { - ret = -ENOMEM; - goto fail_mem; - } - p = etnaviv_domain->stlb_cpu[i]; - for (j = 0; j < SZ_4K / 4; j++) - *p++ = MMUv2_PTE_EXCEPTION; - - etnaviv_domain->mtlb_cpu[i] = etnaviv_domain->stlb_dma[i] | - MMUv2_PTE_PRESENT; - } + memset32(etnaviv_domain->mtlb_cpu, MMUv2_PTE_EXCEPTION, + MMUv2_MAX_STLB_ENTRIES); return 0; fail_mem: if (etnaviv_domain->base.bad_page_cpu) - dma_free_coherent(etnaviv_domain->base.dev, SZ_4K, - etnaviv_domain->base.bad_page_cpu, - etnaviv_domain->base.bad_page_dma); + dma_free_wc(etnaviv_domain->base.dev, SZ_4K, + etnaviv_domain->base.bad_page_cpu, + etnaviv_domain->base.bad_page_dma); if (etnaviv_domain->pta_cpu) - dma_free_coherent(etnaviv_domain->base.dev, SZ_4K, - etnaviv_domain->pta_cpu, - etnaviv_domain->pta_dma); + dma_free_wc(etnaviv_domain->base.dev, SZ_4K, + etnaviv_domain->pta_cpu, etnaviv_domain->pta_dma); if (etnaviv_domain->mtlb_cpu) - dma_free_coherent(etnaviv_domain->base.dev, SZ_4K, - etnaviv_domain->mtlb_cpu, - etnaviv_domain->mtlb_dma); - - for (i = 0; i < MMUv2_MAX_STLB_ENTRIES; i++) { - if (etnaviv_domain->stlb_cpu[i]) - dma_free_coherent(etnaviv_domain->base.dev, SZ_4K, - etnaviv_domain->stlb_cpu[i], - etnaviv_domain->stlb_dma[i]); - } + dma_free_wc(etnaviv_domain->base.dev, SZ_4K, + etnaviv_domain->mtlb_cpu, etnaviv_domain->mtlb_dma); return ret; } @@ -188,23 +179,21 @@ static void etnaviv_iommuv2_domain_free(struct etnaviv_iommu_domain *domain) to_etnaviv_domain(domain); int i; - dma_free_coherent(etnaviv_domain->base.dev, SZ_4K, - etnaviv_domain->base.bad_page_cpu, - etnaviv_domain->base.bad_page_dma); + dma_free_wc(etnaviv_domain->base.dev, SZ_4K, + etnaviv_domain->base.bad_page_cpu, + etnaviv_domain->base.bad_page_dma); - dma_free_coherent(etnaviv_domain->base.dev, SZ_4K, - etnaviv_domain->pta_cpu, - etnaviv_domain->pta_dma); + dma_free_wc(etnaviv_domain->base.dev, SZ_4K, + etnaviv_domain->pta_cpu, etnaviv_domain->pta_dma); - dma_free_coherent(etnaviv_domain->base.dev, SZ_4K, - etnaviv_domain->mtlb_cpu, - etnaviv_domain->mtlb_dma); + dma_free_wc(etnaviv_domain->base.dev, SZ_4K, + etnaviv_domain->mtlb_cpu, etnaviv_domain->mtlb_dma); for (i = 0; i < MMUv2_MAX_STLB_ENTRIES; i++) { if (etnaviv_domain->stlb_cpu[i]) - dma_free_coherent(etnaviv_domain->base.dev, SZ_4K, - etnaviv_domain->stlb_cpu[i], - etnaviv_domain->stlb_dma[i]); + dma_free_wc(etnaviv_domain->base.dev, SZ_4K, + etnaviv_domain->stlb_cpu[i], + etnaviv_domain->stlb_dma[i]); } vfree(etnaviv_domain); diff --git a/drivers/gpu/drm/etnaviv/etnaviv_mmu.c b/drivers/gpu/drm/etnaviv/etnaviv_mmu.c index 49e049713a52..8069f9f36a2e 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_mmu.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_mmu.c @@ -1,17 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* - * Copyright (C) 2015 Etnaviv Project - * - * This program is free software; you can redistribute it and/or modify it - * under the terms of the GNU General Public License version 2 as published by - * the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License along with - * this program. If not, see <http://www.gnu.org/licenses/>. + * Copyright (C) 2015-2018 Etnaviv Project */ #include "common.xml.h" @@ -162,22 +151,10 @@ static int etnaviv_iommu_find_iova(struct etnaviv_iommu *mmu, bool found; ret = drm_mm_insert_node_in_range(&mmu->mm, node, - size, 0, 0, - mmu->last_iova, U64_MAX, - mode); + size, 0, 0, 0, U64_MAX, mode); if (ret != -ENOSPC) break; - /* - * If we did not search from the start of the MMU region, - * try again in case there are free slots. - */ - if (mmu->last_iova) { - mmu->last_iova = 0; - mmu->need_flush = true; - continue; - } - /* Try to retire some entries */ drm_mm_scan_init(&scan, &mmu->mm, size, 0, 0, mode); @@ -274,7 +251,6 @@ int etnaviv_iommu_map_gem(struct etnaviv_iommu *mmu, if (ret < 0) goto unlock; - mmu->last_iova = node->start + etnaviv_obj->base.size; mapping->iova = node->start; ret = etnaviv_iommu_map(mmu, node->start, sgt, etnaviv_obj->base.size, ETNAVIV_PROT_READ | ETNAVIV_PROT_WRITE); @@ -381,7 +357,6 @@ int etnaviv_iommu_get_suballoc_va(struct etnaviv_gpu *gpu, dma_addr_t paddr, mutex_unlock(&mmu->lock); return ret; } - mmu->last_iova = vram_node->start + size; gpu->mmu->need_flush = true; mutex_unlock(&mmu->lock); diff --git a/drivers/gpu/drm/etnaviv/etnaviv_mmu.h b/drivers/gpu/drm/etnaviv/etnaviv_mmu.h index ab603f5166b1..a0db17ffb686 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_mmu.h +++ b/drivers/gpu/drm/etnaviv/etnaviv_mmu.h @@ -1,17 +1,6 @@ +/* SPDX-License-Identifier: GPL-2.0 */ /* - * Copyright (C) 2015 Etnaviv Project - * - * This program is free software; you can redistribute it and/or modify it - * under the terms of the GNU General Public License version 2 as published by - * the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License along with - * this program. If not, see <http://www.gnu.org/licenses/>. + * Copyright (C) 2015-2018 Etnaviv Project */ #ifndef __ETNAVIV_MMU_H__ @@ -59,7 +48,6 @@ struct etnaviv_iommu { struct mutex lock; struct list_head mappings; struct drm_mm mm; - u32 last_iova; bool need_flush; }; diff --git a/drivers/gpu/drm/etnaviv/etnaviv_perfmon.c b/drivers/gpu/drm/etnaviv/etnaviv_perfmon.c index 26dddfc41aac..9980d81a26e3 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_perfmon.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_perfmon.c @@ -1,18 +1,7 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2017 Etnaviv Project * Copyright (C) 2017 Zodiac Inflight Innovations - * - * This program is free software; you can redistribute it and/or modify it - * under the terms of the GNU General Public License version 2 as published by - * the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License along with - * this program. If not, see <http://www.gnu.org/licenses/>. */ #include "etnaviv_gpu.h" diff --git a/drivers/gpu/drm/etnaviv/etnaviv_perfmon.h b/drivers/gpu/drm/etnaviv/etnaviv_perfmon.h index c1653c64ab6b..4a9d508f6e10 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_perfmon.h +++ b/drivers/gpu/drm/etnaviv/etnaviv_perfmon.h @@ -1,18 +1,7 @@ +/* SPDX-License-Identifier: GPL-2.0 */ /* * Copyright (C) 2017 Etnaviv Project * Copyright (C) 2017 Zodiac Inflight Innovations - * - * This program is free software; you can redistribute it and/or modify it - * under the terms of the GNU General Public License version 2 as published by - * the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License along with - * this program. If not, see <http://www.gnu.org/licenses/>. */ #ifndef __ETNAVIV_PERFMON_H__ diff --git a/drivers/gpu/drm/etnaviv/etnaviv_sched.c b/drivers/gpu/drm/etnaviv/etnaviv_sched.c index 6cf0775dbcd7..a74eb57af15b 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_sched.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_sched.c @@ -1,17 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2017 Etnaviv Project - * - * This program is free software; you can redistribute it and/or modify it - * under the terms of the GNU General Public License version 2 as published by - * the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License along with - * this program. If not, see <http://www.gnu.org/licenses/>. */ #include <linux/kthread.h> diff --git a/drivers/gpu/drm/etnaviv/etnaviv_sched.h b/drivers/gpu/drm/etnaviv/etnaviv_sched.h index 097635fa78ae..c0a6796e22c9 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_sched.h +++ b/drivers/gpu/drm/etnaviv/etnaviv_sched.h @@ -1,17 +1,6 @@ +/* SPDX-License-Identifier: GPL-2.0 */ /* * Copyright (C) 2017 Etnaviv Project - * - * This program is free software; you can redistribute it and/or modify it - * under the terms of the GNU General Public License version 2 as published by - * the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License along with - * this program. If not, see <http://www.gnu.org/licenses/>. */ #ifndef __ETNAVIV_SCHED_H__ diff --git a/drivers/gpu/drm/exynos/exynos5433_drm_decon.c b/drivers/gpu/drm/exynos/exynos5433_drm_decon.c index 1c330f2a7a5d..82c95c34447f 100644 --- a/drivers/gpu/drm/exynos/exynos5433_drm_decon.c +++ b/drivers/gpu/drm/exynos/exynos5433_drm_decon.c @@ -31,7 +31,10 @@ #define DSD_CFG_MUX 0x1004 #define DSD_CFG_MUX_TE_UNMASK_GLOBAL BIT(13) -#define WINDOWS_NR 3 +#define WINDOWS_NR 5 +#define PRIMARY_WIN 2 +#define CURSON_WIN 4 + #define MIN_FB_WIDTH_FOR_16WORD_BURST 128 #define I80_HW_TRG (1 << 0) @@ -43,6 +46,9 @@ static const char * const decon_clks_name[] = { "aclk_smmu_decon0x", "aclk_xiu_decon0x", "pclk_smmu_decon0x", + "aclk_smmu_decon1x", + "aclk_xiu_decon1x", + "pclk_smmu_decon1x", "sclk_decon_vclk", "sclk_decon_eclk", }; @@ -74,9 +80,8 @@ static const uint32_t decon_formats[] = { }; static const enum drm_plane_type decon_win_types[WINDOWS_NR] = { - DRM_PLANE_TYPE_PRIMARY, - DRM_PLANE_TYPE_OVERLAY, - DRM_PLANE_TYPE_CURSOR, + [PRIMARY_WIN] = DRM_PLANE_TYPE_PRIMARY, + [CURSON_WIN] = DRM_PLANE_TYPE_CURSOR, }; static inline void decon_set_bits(struct decon_context *ctx, u32 reg, u32 mask, @@ -552,12 +557,10 @@ static int decon_bind(struct device *dev, struct device *master, void *data) drm_dev->max_vblank_count = 0xffffffff; for (win = ctx->first_win; win < WINDOWS_NR; win++) { - int tmp = (win == ctx->first_win) ? 0 : win; - ctx->configs[win].pixel_formats = decon_formats; ctx->configs[win].num_pixel_formats = ARRAY_SIZE(decon_formats); - ctx->configs[win].zpos = win; - ctx->configs[win].type = decon_win_types[tmp]; + ctx->configs[win].zpos = win - ctx->first_win; + ctx->configs[win].type = decon_win_types[win]; ret = exynos_plane_init(drm_dev, &ctx->planes[win], win, &ctx->configs[win]); @@ -565,7 +568,7 @@ static int decon_bind(struct device *dev, struct device *master, void *data) return ret; } - exynos_plane = &ctx->planes[ctx->first_win]; + exynos_plane = &ctx->planes[PRIMARY_WIN]; out_type = (ctx->out_type & IFTYPE_HDMI) ? EXYNOS_DISPLAY_TYPE_HDMI : EXYNOS_DISPLAY_TYPE_LCD; ctx->crtc = exynos_drm_crtc_create(drm_dev, &exynos_plane->base, diff --git a/drivers/gpu/drm/exynos/exynos_drm_crtc.c b/drivers/gpu/drm/exynos/exynos_drm_crtc.c index dc01342e759a..eea90251808f 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_crtc.c +++ b/drivers/gpu/drm/exynos/exynos_drm_crtc.c @@ -228,7 +228,7 @@ struct exynos_drm_crtc *exynos_drm_crtc_get_by_type(struct drm_device *drm_dev, if (to_exynos_crtc(crtc)->type == out_type) return to_exynos_crtc(crtc); - return ERR_PTR(-EPERM); + return ERR_PTR(-ENODEV); } int exynos_drm_set_possible_crtcs(struct drm_encoder *encoder, diff --git a/drivers/gpu/drm/exynos/exynos_drm_dsi.c b/drivers/gpu/drm/exynos/exynos_drm_dsi.c index eae44fd714f0..7c3030b7e586 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_dsi.c +++ b/drivers/gpu/drm/exynos/exynos_drm_dsi.c @@ -1264,15 +1264,15 @@ static irqreturn_t exynos_dsi_irq(int irq, void *dev_id) if (status & DSIM_INT_SW_RST_RELEASE) { u32 mask = ~(DSIM_INT_RX_DONE | DSIM_INT_SFR_FIFO_EMPTY | - DSIM_INT_SFR_HDR_FIFO_EMPTY | DSIM_INT_FRAME_DONE | - DSIM_INT_RX_ECC_ERR | DSIM_INT_SW_RST_RELEASE); + DSIM_INT_SFR_HDR_FIFO_EMPTY | DSIM_INT_RX_ECC_ERR | + DSIM_INT_SW_RST_RELEASE); exynos_dsi_write(dsi, DSIM_INTMSK_REG, mask); complete(&dsi->completed); return IRQ_HANDLED; } if (!(status & (DSIM_INT_RX_DONE | DSIM_INT_SFR_FIFO_EMPTY | - DSIM_INT_FRAME_DONE | DSIM_INT_PLL_STABLE))) + DSIM_INT_PLL_STABLE))) return IRQ_HANDLED; if (exynos_dsi_transfer_finish(dsi)) diff --git a/drivers/gpu/drm/exynos/exynos_drm_fimc.c b/drivers/gpu/drm/exynos/exynos_drm_fimc.c index 4dfbfc7f3b84..5ce84025d1cb 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_fimc.c +++ b/drivers/gpu/drm/exynos/exynos_drm_fimc.c @@ -1200,7 +1200,7 @@ e_clk_free: int exynos_drm_check_fimc_device(struct device *dev) { - unsigned int id = of_alias_get_id(dev->of_node, "fimc"); + int id = of_alias_get_id(dev->of_node, "fimc"); if (id >= 0 && (BIT(id) & fimc_mask)) return 0; diff --git a/drivers/gpu/drm/exynos/exynos_drm_plane.c b/drivers/gpu/drm/exynos/exynos_drm_plane.c index d2a90dae5c71..38a2a7f1204b 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_plane.c +++ b/drivers/gpu/drm/exynos/exynos_drm_plane.c @@ -289,13 +289,12 @@ static const struct drm_plane_helper_funcs plane_helper_funcs = { }; static void exynos_plane_attach_zpos_property(struct drm_plane *plane, - bool immutable) + int zpos, bool immutable) { - /* FIXME */ if (immutable) - drm_plane_create_zpos_immutable_property(plane, 0); + drm_plane_create_zpos_immutable_property(plane, zpos); else - drm_plane_create_zpos_property(plane, 0, 0, MAX_PLANE - 1); + drm_plane_create_zpos_property(plane, zpos, 0, MAX_PLANE - 1); } int exynos_plane_init(struct drm_device *dev, @@ -320,7 +319,7 @@ int exynos_plane_init(struct drm_device *dev, exynos_plane->index = index; exynos_plane->config = config; - exynos_plane_attach_zpos_property(&exynos_plane->base, + exynos_plane_attach_zpos_property(&exynos_plane->base, config->zpos, !(config->capabilities & EXYNOS_DRM_PLANE_CAP_ZPOS)); return 0; diff --git a/drivers/gpu/drm/exynos/exynos_drm_scaler.c b/drivers/gpu/drm/exynos/exynos_drm_scaler.c index 63b05b7c846a..91d4382343d0 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_scaler.c +++ b/drivers/gpu/drm/exynos/exynos_drm_scaler.c @@ -397,7 +397,7 @@ static inline u32 scaler_get_int_status(struct scaler_context *scaler) return scaler_read(SCALER_INT_STATUS); } -static inline bool scaler_task_done(u32 val) +static inline int scaler_task_done(u32 val) { return val & SCALER_INT_STATUS_FRAME_END ? 0 : -EINVAL; } diff --git a/drivers/gpu/drm/i2c/Kconfig b/drivers/gpu/drm/i2c/Kconfig index a6c92beb410a..65d3acb61c03 100644 --- a/drivers/gpu/drm/i2c/Kconfig +++ b/drivers/gpu/drm/i2c/Kconfig @@ -22,8 +22,14 @@ config DRM_I2C_SIL164 config DRM_I2C_NXP_TDA998X tristate "NXP Semiconductors TDA998X HDMI encoder" default m if DRM_TILCDC + select CEC_CORE if CEC_NOTIFIER select SND_SOC_HDMI_CODEC if SND_SOC help Support for NXP Semiconductors TDA998X HDMI encoders. +config DRM_I2C_NXP_TDA9950 + tristate "NXP Semiconductors TDA9950/TDA998X HDMI CEC" + select CEC_NOTIFIER + select CEC_CORE + endmenu diff --git a/drivers/gpu/drm/i2c/Makefile b/drivers/gpu/drm/i2c/Makefile index b20100c18ffb..a962f6f08568 100644 --- a/drivers/gpu/drm/i2c/Makefile +++ b/drivers/gpu/drm/i2c/Makefile @@ -7,3 +7,4 @@ obj-$(CONFIG_DRM_I2C_SIL164) += sil164.o tda998x-y := tda998x_drv.o obj-$(CONFIG_DRM_I2C_NXP_TDA998X) += tda998x.o +obj-$(CONFIG_DRM_I2C_NXP_TDA9950) += tda9950.o diff --git a/drivers/gpu/drm/i2c/tda9950.c b/drivers/gpu/drm/i2c/tda9950.c new file mode 100644 index 000000000000..3f7396caad48 --- /dev/null +++ b/drivers/gpu/drm/i2c/tda9950.c @@ -0,0 +1,509 @@ +/* + * TDA9950 Consumer Electronics Control driver + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * The NXP TDA9950 implements the HDMI Consumer Electronics Control + * interface. The host interface is similar to a mailbox: the data + * registers starting at REG_CDR0 are written to send a command to the + * internal CPU, and replies are read from these registers. + * + * As the data registers represent a mailbox, they must be accessed + * as a single I2C transaction. See the TDA9950 data sheet for details. + */ +#include <linux/delay.h> +#include <linux/i2c.h> +#include <linux/interrupt.h> +#include <linux/module.h> +#include <linux/platform_data/tda9950.h> +#include <linux/slab.h> +#include <drm/drm_edid.h> +#include <media/cec.h> +#include <media/cec-notifier.h> + +enum { + REG_CSR = 0x00, + CSR_BUSY = BIT(7), + CSR_INT = BIT(6), + CSR_ERR = BIT(5), + + REG_CER = 0x01, + + REG_CVR = 0x02, + + REG_CCR = 0x03, + CCR_RESET = BIT(7), + CCR_ON = BIT(6), + + REG_ACKH = 0x04, + REG_ACKL = 0x05, + + REG_CCONR = 0x06, + CCONR_ENABLE_ERROR = BIT(4), + CCONR_RETRY_MASK = 7, + + REG_CDR0 = 0x07, + + CDR1_REQ = 0x00, + CDR1_CNF = 0x01, + CDR1_IND = 0x81, + CDR1_ERR = 0x82, + CDR1_IER = 0x83, + + CDR2_CNF_SUCCESS = 0x00, + CDR2_CNF_OFF_STATE = 0x80, + CDR2_CNF_BAD_REQ = 0x81, + CDR2_CNF_CEC_ACCESS = 0x82, + CDR2_CNF_ARB_ERROR = 0x83, + CDR2_CNF_BAD_TIMING = 0x84, + CDR2_CNF_NACK_ADDR = 0x85, + CDR2_CNF_NACK_DATA = 0x86, +}; + +struct tda9950_priv { + struct i2c_client *client; + struct device *hdmi; + struct cec_adapter *adap; + struct tda9950_glue *glue; + u16 addresses; + struct cec_msg rx_msg; + struct cec_notifier *notify; + bool open; +}; + +static int tda9950_write_range(struct i2c_client *client, u8 addr, u8 *p, int cnt) +{ + struct i2c_msg msg; + u8 buf[cnt + 1]; + int ret; + + buf[0] = addr; + memcpy(buf + 1, p, cnt); + + msg.addr = client->addr; + msg.flags = 0; + msg.len = cnt + 1; + msg.buf = buf; + + dev_dbg(&client->dev, "wr 0x%02x: %*ph\n", addr, cnt, p); + + ret = i2c_transfer(client->adapter, &msg, 1); + if (ret < 0) + dev_err(&client->dev, "Error %d writing to cec:0x%x\n", ret, addr); + return ret < 0 ? ret : 0; +} + +static void tda9950_write(struct i2c_client *client, u8 addr, u8 val) +{ + tda9950_write_range(client, addr, &val, 1); +} + +static int tda9950_read_range(struct i2c_client *client, u8 addr, u8 *p, int cnt) +{ + struct i2c_msg msg[2]; + int ret; + + msg[0].addr = client->addr; + msg[0].flags = 0; + msg[0].len = 1; + msg[0].buf = &addr; + msg[1].addr = client->addr; + msg[1].flags = I2C_M_RD; + msg[1].len = cnt; + msg[1].buf = p; + + ret = i2c_transfer(client->adapter, msg, 2); + if (ret < 0) + dev_err(&client->dev, "Error %d reading from cec:0x%x\n", ret, addr); + + dev_dbg(&client->dev, "rd 0x%02x: %*ph\n", addr, cnt, p); + + return ret; +} + +static u8 tda9950_read(struct i2c_client *client, u8 addr) +{ + int ret; + u8 val; + + ret = tda9950_read_range(client, addr, &val, 1); + if (ret < 0) + val = 0; + + return val; +} + +static irqreturn_t tda9950_irq(int irq, void *data) +{ + struct tda9950_priv *priv = data; + unsigned int tx_status; + u8 csr, cconr, buf[19]; + u8 arb_lost_cnt, nack_cnt, err_cnt; + + if (!priv->open) + return IRQ_NONE; + + csr = tda9950_read(priv->client, REG_CSR); + if (!(csr & CSR_INT)) + return IRQ_NONE; + + cconr = tda9950_read(priv->client, REG_CCONR) & CCONR_RETRY_MASK; + + tda9950_read_range(priv->client, REG_CDR0, buf, sizeof(buf)); + + /* + * This should never happen: the data sheet says that there will + * always be a valid message if the interrupt line is asserted. + */ + if (buf[0] == 0) { + dev_warn(&priv->client->dev, "interrupt pending, but no message?\n"); + return IRQ_NONE; + } + + switch (buf[1]) { + case CDR1_CNF: /* transmit result */ + arb_lost_cnt = nack_cnt = err_cnt = 0; + switch (buf[2]) { + case CDR2_CNF_SUCCESS: + tx_status = CEC_TX_STATUS_OK; + break; + + case CDR2_CNF_ARB_ERROR: + tx_status = CEC_TX_STATUS_ARB_LOST; + arb_lost_cnt = cconr; + break; + + case CDR2_CNF_NACK_ADDR: + tx_status = CEC_TX_STATUS_NACK; + nack_cnt = cconr; + break; + + default: /* some other error, refer to TDA9950 docs */ + dev_err(&priv->client->dev, "CNF reply error 0x%02x\n", + buf[2]); + tx_status = CEC_TX_STATUS_ERROR; + err_cnt = cconr; + break; + } + /* TDA9950 executes all retries for us */ + tx_status |= CEC_TX_STATUS_MAX_RETRIES; + cec_transmit_done(priv->adap, tx_status, arb_lost_cnt, + nack_cnt, 0, err_cnt); + break; + + case CDR1_IND: + priv->rx_msg.len = buf[0] - 2; + if (priv->rx_msg.len > CEC_MAX_MSG_SIZE) + priv->rx_msg.len = CEC_MAX_MSG_SIZE; + + memcpy(priv->rx_msg.msg, buf + 2, priv->rx_msg.len); + cec_received_msg(priv->adap, &priv->rx_msg); + break; + + default: /* unknown */ + dev_err(&priv->client->dev, "unknown service id 0x%02x\n", + buf[1]); + break; + } + + return IRQ_HANDLED; +} + +static int tda9950_cec_transmit(struct cec_adapter *adap, u8 attempts, + u32 signal_free_time, struct cec_msg *msg) +{ + struct tda9950_priv *priv = adap->priv; + u8 buf[CEC_MAX_MSG_SIZE + 2]; + + buf[0] = 2 + msg->len; + buf[1] = CDR1_REQ; + memcpy(buf + 2, msg->msg, msg->len); + + if (attempts > 5) + attempts = 5; + + tda9950_write(priv->client, REG_CCONR, attempts); + + return tda9950_write_range(priv->client, REG_CDR0, buf, 2 + msg->len); +} + +static int tda9950_cec_adap_log_addr(struct cec_adapter *adap, u8 addr) +{ + struct tda9950_priv *priv = adap->priv; + u16 addresses; + u8 buf[2]; + + if (addr == CEC_LOG_ADDR_INVALID) + addresses = priv->addresses = 0; + else + addresses = priv->addresses |= BIT(addr); + + /* TDA9950 doesn't want address 15 set */ + addresses &= 0x7fff; + buf[0] = addresses >> 8; + buf[1] = addresses; + + return tda9950_write_range(priv->client, REG_ACKH, buf, 2); +} + +/* + * When operating as part of the TDA998x, we need additional handling + * to initialise and shut down the TDA9950 part of the device. These + * two hooks are provided to allow the TDA998x code to perform those + * activities. + */ +static int tda9950_glue_open(struct tda9950_priv *priv) +{ + int ret = 0; + + if (priv->glue && priv->glue->open) + ret = priv->glue->open(priv->glue->data); + + priv->open = true; + + return ret; +} + +static void tda9950_glue_release(struct tda9950_priv *priv) +{ + priv->open = false; + + if (priv->glue && priv->glue->release) + priv->glue->release(priv->glue->data); +} + +static int tda9950_open(struct tda9950_priv *priv) +{ + struct i2c_client *client = priv->client; + int ret; + + ret = tda9950_glue_open(priv); + if (ret) + return ret; + + /* Reset the TDA9950, and wait 250ms for it to recover */ + tda9950_write(client, REG_CCR, CCR_RESET); + msleep(250); + + tda9950_cec_adap_log_addr(priv->adap, CEC_LOG_ADDR_INVALID); + + /* Start the command processor */ + tda9950_write(client, REG_CCR, CCR_ON); + + return 0; +} + +static void tda9950_release(struct tda9950_priv *priv) +{ + struct i2c_client *client = priv->client; + int timeout = 50; + u8 csr; + + /* Stop the command processor */ + tda9950_write(client, REG_CCR, 0); + + /* Wait up to .5s for it to signal non-busy */ + do { + csr = tda9950_read(client, REG_CSR); + if (!(csr & CSR_BUSY) || --timeout) + break; + msleep(10); + } while (1); + + /* Warn the user that their IRQ may die if it's shared. */ + if (csr & CSR_BUSY) + dev_warn(&client->dev, "command processor failed to stop, irq%d may die (csr=0x%02x)\n", + client->irq, csr); + + tda9950_glue_release(priv); +} + +static int tda9950_cec_adap_enable(struct cec_adapter *adap, bool enable) +{ + struct tda9950_priv *priv = adap->priv; + + if (!enable) { + tda9950_release(priv); + return 0; + } else { + return tda9950_open(priv); + } +} + +static const struct cec_adap_ops tda9950_cec_ops = { + .adap_enable = tda9950_cec_adap_enable, + .adap_log_addr = tda9950_cec_adap_log_addr, + .adap_transmit = tda9950_cec_transmit, +}; + +/* + * When operating as part of the TDA998x, we need to claim additional + * resources. These two hooks permit the management of those resources. + */ +static void tda9950_devm_glue_exit(void *data) +{ + struct tda9950_glue *glue = data; + + if (glue && glue->exit) + glue->exit(glue->data); +} + +static int tda9950_devm_glue_init(struct device *dev, struct tda9950_glue *glue) +{ + int ret; + + if (glue && glue->init) { + ret = glue->init(glue->data); + if (ret) + return ret; + } + + ret = devm_add_action(dev, tda9950_devm_glue_exit, glue); + if (ret) + tda9950_devm_glue_exit(glue); + + return ret; +} + +static void tda9950_cec_del(void *data) +{ + struct tda9950_priv *priv = data; + + cec_delete_adapter(priv->adap); +} + +static int tda9950_probe(struct i2c_client *client, + const struct i2c_device_id *id) +{ + struct tda9950_glue *glue = client->dev.platform_data; + struct device *dev = &client->dev; + struct tda9950_priv *priv; + unsigned long irqflags; + int ret; + u8 cvr; + + /* + * We must have I2C functionality: our multi-byte accesses + * must be performed as a single contiguous transaction. + */ + if (!i2c_check_functionality(client->adapter, I2C_FUNC_I2C)) { + dev_err(&client->dev, + "adapter does not support I2C functionality\n"); + return -ENXIO; + } + + /* We must have an interrupt to be functional. */ + if (client->irq <= 0) { + dev_err(&client->dev, "driver requires an interrupt\n"); + return -ENXIO; + } + + priv = devm_kzalloc(dev, sizeof(*priv), GFP_KERNEL); + if (!priv) + return -ENOMEM; + + priv->client = client; + priv->glue = glue; + + i2c_set_clientdata(client, priv); + + /* + * If we're part of a TDA998x, we want the class devices to be + * associated with the HDMI Tx so we have a tight relationship + * between the HDMI interface and the CEC interface. + */ + priv->hdmi = dev; + if (glue && glue->parent) + priv->hdmi = glue->parent; + + priv->adap = cec_allocate_adapter(&tda9950_cec_ops, priv, "tda9950", + CEC_CAP_DEFAULTS, + CEC_MAX_LOG_ADDRS); + if (IS_ERR(priv->adap)) + return PTR_ERR(priv->adap); + + ret = devm_add_action(dev, tda9950_cec_del, priv); + if (ret) { + cec_delete_adapter(priv->adap); + return ret; + } + + ret = tda9950_devm_glue_init(dev, glue); + if (ret) + return ret; + + ret = tda9950_glue_open(priv); + if (ret) + return ret; + + cvr = tda9950_read(client, REG_CVR); + + dev_info(&client->dev, + "TDA9950 CEC interface, hardware version %u.%u\n", + cvr >> 4, cvr & 15); + + tda9950_glue_release(priv); + + irqflags = IRQF_TRIGGER_FALLING; + if (glue) + irqflags = glue->irq_flags; + + ret = devm_request_threaded_irq(dev, client->irq, NULL, tda9950_irq, + irqflags | IRQF_SHARED | IRQF_ONESHOT, + dev_name(&client->dev), priv); + if (ret < 0) + return ret; + + priv->notify = cec_notifier_get(priv->hdmi); + if (!priv->notify) + return -ENOMEM; + + ret = cec_register_adapter(priv->adap, priv->hdmi); + if (ret < 0) { + cec_notifier_put(priv->notify); + return ret; + } + + /* + * CEC documentation says we must not call cec_delete_adapter + * after a successful call to cec_register_adapter(). + */ + devm_remove_action(dev, tda9950_cec_del, priv); + + cec_register_cec_notifier(priv->adap, priv->notify); + + return 0; +} + +static int tda9950_remove(struct i2c_client *client) +{ + struct tda9950_priv *priv = i2c_get_clientdata(client); + + cec_unregister_adapter(priv->adap); + cec_notifier_put(priv->notify); + + return 0; +} + +static struct i2c_device_id tda9950_ids[] = { + { "tda9950", 0 }, + { }, +}; +MODULE_DEVICE_TABLE(i2c, tda9950_ids); + +static struct i2c_driver tda9950_driver = { + .probe = tda9950_probe, + .remove = tda9950_remove, + .driver = { + .name = "tda9950", + }, + .id_table = tda9950_ids, +}; + +module_i2c_driver(tda9950_driver); + +MODULE_AUTHOR("Russell King <rmk+kernel@armlinux.org.uk>"); +MODULE_DESCRIPTION("TDA9950/TDA998x Consumer Electronics Control Driver"); +MODULE_LICENSE("GPL v2"); diff --git a/drivers/gpu/drm/i2c/tda998x_drv.c b/drivers/gpu/drm/i2c/tda998x_drv.c index 421c8a72369e..6ebd8842dbcc 100644 --- a/drivers/gpu/drm/i2c/tda998x_drv.c +++ b/drivers/gpu/drm/i2c/tda998x_drv.c @@ -16,8 +16,10 @@ */ #include <linux/component.h> +#include <linux/gpio/consumer.h> #include <linux/hdmi.h> #include <linux/module.h> +#include <linux/platform_data/tda9950.h> #include <linux/irq.h> #include <sound/asoundef.h> #include <sound/hdmi-codec.h> @@ -29,6 +31,8 @@ #include <drm/drm_of.h> #include <drm/i2c/tda998x.h> +#include <media/cec-notifier.h> + #define DBG(fmt, ...) DRM_DEBUG(fmt"\n", ##__VA_ARGS__) struct tda998x_audio_port { @@ -55,6 +59,7 @@ struct tda998x_priv { struct platform_device *audio_pdev; struct mutex audio_mutex; + struct mutex edid_mutex; wait_queue_head_t wq_edid; volatile int wq_edid_wait; @@ -67,6 +72,9 @@ struct tda998x_priv { struct drm_connector connector; struct tda998x_audio_port audio_port[2]; + struct tda9950_glue cec_glue; + struct gpio_desc *calib; + struct cec_notifier *cec_notify; }; #define conn_to_tda998x_priv(x) \ @@ -345,6 +353,12 @@ struct tda998x_priv { #define REG_CEC_INTSTATUS 0xee /* read */ # define CEC_INTSTATUS_CEC (1 << 0) # define CEC_INTSTATUS_HDMI (1 << 1) +#define REG_CEC_CAL_XOSC_CTRL1 0xf2 +# define CEC_CAL_XOSC_CTRL1_ENA_CAL BIT(0) +#define REG_CEC_DES_FREQ2 0xf5 +# define CEC_DES_FREQ2_DIS_AUTOCAL BIT(7) +#define REG_CEC_CLK 0xf6 +# define CEC_CLK_FRO 0x11 #define REG_CEC_FRO_IM_CLK_CTRL 0xfb /* read/write */ # define CEC_FRO_IM_CLK_CTRL_GHOST_DIS (1 << 7) # define CEC_FRO_IM_CLK_CTRL_ENA_OTP (1 << 6) @@ -359,6 +373,7 @@ struct tda998x_priv { # define CEC_RXSHPDLEV_HPD (1 << 1) #define REG_CEC_ENAMODS 0xff /* read/write */ +# define CEC_ENAMODS_EN_CEC_CLK (1 << 7) # define CEC_ENAMODS_DIS_FRO (1 << 6) # define CEC_ENAMODS_DIS_CCLK (1 << 5) # define CEC_ENAMODS_EN_RXSENS (1 << 2) @@ -417,6 +432,114 @@ cec_read(struct tda998x_priv *priv, u8 addr) return val; } +static void cec_enamods(struct tda998x_priv *priv, u8 mods, bool enable) +{ + int val = cec_read(priv, REG_CEC_ENAMODS); + + if (val < 0) + return; + + if (enable) + val |= mods; + else + val &= ~mods; + + cec_write(priv, REG_CEC_ENAMODS, val); +} + +static void tda998x_cec_set_calibration(struct tda998x_priv *priv, bool enable) +{ + if (enable) { + u8 val; + + cec_write(priv, 0xf3, 0xc0); + cec_write(priv, 0xf4, 0xd4); + + /* Enable automatic calibration mode */ + val = cec_read(priv, REG_CEC_DES_FREQ2); + val &= ~CEC_DES_FREQ2_DIS_AUTOCAL; + cec_write(priv, REG_CEC_DES_FREQ2, val); + + /* Enable free running oscillator */ + cec_write(priv, REG_CEC_CLK, CEC_CLK_FRO); + cec_enamods(priv, CEC_ENAMODS_DIS_FRO, false); + + cec_write(priv, REG_CEC_CAL_XOSC_CTRL1, + CEC_CAL_XOSC_CTRL1_ENA_CAL); + } else { + cec_write(priv, REG_CEC_CAL_XOSC_CTRL1, 0); + } +} + +/* + * Calibration for the internal oscillator: we need to set calibration mode, + * and then pulse the IRQ line low for a 10ms ± 1% period. + */ +static void tda998x_cec_calibration(struct tda998x_priv *priv) +{ + struct gpio_desc *calib = priv->calib; + + mutex_lock(&priv->edid_mutex); + if (priv->hdmi->irq > 0) + disable_irq(priv->hdmi->irq); + gpiod_direction_output(calib, 1); + tda998x_cec_set_calibration(priv, true); + + local_irq_disable(); + gpiod_set_value(calib, 0); + mdelay(10); + gpiod_set_value(calib, 1); + local_irq_enable(); + + tda998x_cec_set_calibration(priv, false); + gpiod_direction_input(calib); + if (priv->hdmi->irq > 0) + enable_irq(priv->hdmi->irq); + mutex_unlock(&priv->edid_mutex); +} + +static int tda998x_cec_hook_init(void *data) +{ + struct tda998x_priv *priv = data; + struct gpio_desc *calib; + + calib = gpiod_get(&priv->hdmi->dev, "nxp,calib", GPIOD_ASIS); + if (IS_ERR(calib)) { + dev_warn(&priv->hdmi->dev, "failed to get calibration gpio: %ld\n", + PTR_ERR(calib)); + return PTR_ERR(calib); + } + + priv->calib = calib; + + return 0; +} + +static void tda998x_cec_hook_exit(void *data) +{ + struct tda998x_priv *priv = data; + + gpiod_put(priv->calib); + priv->calib = NULL; +} + +static int tda998x_cec_hook_open(void *data) +{ + struct tda998x_priv *priv = data; + + cec_enamods(priv, CEC_ENAMODS_EN_CEC_CLK | CEC_ENAMODS_EN_CEC, true); + tda998x_cec_calibration(priv); + + return 0; +} + +static void tda998x_cec_hook_release(void *data) +{ + struct tda998x_priv *priv = data; + + cec_enamods(priv, CEC_ENAMODS_EN_CEC_CLK | CEC_ENAMODS_EN_CEC, false); +} + static int set_page(struct tda998x_priv *priv, u16 reg) { @@ -657,10 +780,13 @@ static irqreturn_t tda998x_irq_thread(int irq, void *data) sta, cec, lvl, flag0, flag1, flag2); if (cec & CEC_RXSHPDINT_HPD) { - if (lvl & CEC_RXSHPDLEV_HPD) + if (lvl & CEC_RXSHPDLEV_HPD) { tda998x_edid_delay_start(priv); - else + } else { schedule_work(&priv->detect_work); + cec_notifier_set_phys_addr(priv->cec_notify, + CEC_PHYS_ADDR_INVALID); + } handled = true; } @@ -981,6 +1107,8 @@ static int tda998x_connector_fill_modes(struct drm_connector *connector, if (connector->edid_blob_ptr) { struct edid *edid = (void *)connector->edid_blob_ptr->data; + cec_notifier_set_phys_addr_from_edid(priv->cec_notify, edid); + priv->sink_has_audio = drm_detect_monitor_audio(edid); } else { priv->sink_has_audio = false; @@ -1024,6 +1152,8 @@ static int read_edid_block(void *data, u8 *buf, unsigned int blk, size_t length) offset = (blk & 1) ? 128 : 0; segptr = blk / 2; + mutex_lock(&priv->edid_mutex); + reg_write(priv, REG_DDC_ADDR, 0xa0); reg_write(priv, REG_DDC_OFFS, offset); reg_write(priv, REG_DDC_SEGM_ADDR, 0x60); @@ -1043,14 +1173,15 @@ static int read_edid_block(void *data, u8 *buf, unsigned int blk, size_t length) msecs_to_jiffies(100)); if (i < 0) { dev_err(&priv->hdmi->dev, "read edid wait err %d\n", i); - return i; + ret = i; + goto failed; } } else { for (i = 100; i > 0; i--) { msleep(1); ret = reg_read(priv, REG_INT_FLAGS_2); if (ret < 0) - return ret; + goto failed; if (ret & INT_FLAGS_2_EDID_BLK_RD) break; } @@ -1058,17 +1189,22 @@ static int read_edid_block(void *data, u8 *buf, unsigned int blk, size_t length) if (i == 0) { dev_err(&priv->hdmi->dev, "read edid timeout\n"); - return -ETIMEDOUT; + ret = -ETIMEDOUT; + goto failed; } ret = reg_read_range(priv, REG_EDID_DATA_0, buf, length); if (ret != length) { dev_err(&priv->hdmi->dev, "failed to read edid block %d: %d\n", blk, ret); - return ret; + goto failed; } - return 0; + ret = 0; + + failed: + mutex_unlock(&priv->edid_mutex); + return ret; } static int tda998x_connector_get_modes(struct drm_connector *connector) @@ -1423,6 +1559,9 @@ static void tda998x_destroy(struct tda998x_priv *priv) cancel_work_sync(&priv->detect_work); i2c_unregister_device(priv->cec); + + if (priv->cec_notify) + cec_notifier_put(priv->cec_notify); } /* I2C driver functions */ @@ -1472,10 +1611,16 @@ static int tda998x_get_audio_ports(struct tda998x_priv *priv, static int tda998x_create(struct i2c_client *client, struct tda998x_priv *priv) { struct device_node *np = client->dev.of_node; + struct i2c_board_info cec_info; u32 video; int rev_lo, rev_hi, ret; - mutex_init(&priv->audio_mutex); /* Protect access from audio thread */ + mutex_init(&priv->mutex); /* protect the page access */ + mutex_init(&priv->audio_mutex); /* protect access from audio thread */ + mutex_init(&priv->edid_mutex); + init_waitqueue_head(&priv->edid_delay_waitq); + timer_setup(&priv->edid_delay_timer, tda998x_edid_delay_done, 0); + INIT_WORK(&priv->detect_work, tda998x_detect_work); priv->vip_cntrl_0 = VIP_CNTRL_0_SWAP_A(2) | VIP_CNTRL_0_SWAP_B(3); priv->vip_cntrl_1 = VIP_CNTRL_1_SWAP_C(0) | VIP_CNTRL_1_SWAP_D(1); @@ -1485,14 +1630,6 @@ static int tda998x_create(struct i2c_client *client, struct tda998x_priv *priv) priv->cec_addr = 0x34 + (client->addr & 0x03); priv->current_page = 0xff; priv->hdmi = client; - priv->cec = i2c_new_dummy(client->adapter, priv->cec_addr); - if (!priv->cec) - return -ENODEV; - - mutex_init(&priv->mutex); /* protect the page access */ - init_waitqueue_head(&priv->edid_delay_waitq); - timer_setup(&priv->edid_delay_timer, tda998x_edid_delay_done, 0); - INIT_WORK(&priv->detect_work, tda998x_detect_work); /* wake up the device: */ cec_write(priv, REG_CEC_ENAMODS, @@ -1502,10 +1639,15 @@ static int tda998x_create(struct i2c_client *client, struct tda998x_priv *priv) /* read version: */ rev_lo = reg_read(priv, REG_VERSION_LSB); + if (rev_lo < 0) { + dev_err(&client->dev, "failed to read version: %d\n", rev_lo); + return rev_lo; + } + rev_hi = reg_read(priv, REG_VERSION_MSB); - if (rev_lo < 0 || rev_hi < 0) { - ret = rev_lo < 0 ? rev_lo : rev_hi; - goto fail; + if (rev_hi < 0) { + dev_err(&client->dev, "failed to read version: %d\n", rev_hi); + return rev_hi; } priv->rev = rev_lo | rev_hi << 8; @@ -1529,7 +1671,7 @@ static int tda998x_create(struct i2c_client *client, struct tda998x_priv *priv) default: dev_err(&client->dev, "found unsupported device: %04x\n", priv->rev); - goto fail; + return -ENXIO; } /* after reset, enable DDC: */ @@ -1545,6 +1687,15 @@ static int tda998x_create(struct i2c_client *client, struct tda998x_priv *priv) cec_write(priv, REG_CEC_FRO_IM_CLK_CTRL, CEC_FRO_IM_CLK_CTRL_GHOST_DIS | CEC_FRO_IM_CLK_CTRL_IMCLK_SEL); + /* ensure interrupts are disabled */ + cec_write(priv, REG_CEC_RXSHPDINTENA, 0); + + /* clear pending interrupts */ + cec_read(priv, REG_CEC_RXSHPDINT); + reg_read(priv, REG_INT_FLAGS_0); + reg_read(priv, REG_INT_FLAGS_1); + reg_read(priv, REG_INT_FLAGS_2); + /* initialize the optional IRQ */ if (client->irq) { unsigned long irq_flags; @@ -1552,13 +1703,11 @@ static int tda998x_create(struct i2c_client *client, struct tda998x_priv *priv) /* init read EDID waitqueue and HDP work */ init_waitqueue_head(&priv->wq_edid); - /* clear pending interrupts */ - reg_read(priv, REG_INT_FLAGS_0); - reg_read(priv, REG_INT_FLAGS_1); - reg_read(priv, REG_INT_FLAGS_2); - irq_flags = irqd_get_trigger_type(irq_get_irq_data(client->irq)); + + priv->cec_glue.irq_flags = irq_flags; + irq_flags |= IRQF_SHARED | IRQF_ONESHOT; ret = request_threaded_irq(client->irq, NULL, tda998x_irq_thread, irq_flags, @@ -1567,13 +1716,46 @@ static int tda998x_create(struct i2c_client *client, struct tda998x_priv *priv) dev_err(&client->dev, "failed to request IRQ#%u: %d\n", client->irq, ret); - goto fail; + goto err_irq; } /* enable HPD irq */ cec_write(priv, REG_CEC_RXSHPDINTENA, CEC_RXSHPDLEV_HPD); } + priv->cec_notify = cec_notifier_get(&client->dev); + if (!priv->cec_notify) { + ret = -ENOMEM; + goto fail; + } + + priv->cec_glue.parent = &client->dev; + priv->cec_glue.data = priv; + priv->cec_glue.init = tda998x_cec_hook_init; + priv->cec_glue.exit = tda998x_cec_hook_exit; + priv->cec_glue.open = tda998x_cec_hook_open; + priv->cec_glue.release = tda998x_cec_hook_release; + + /* + * Some TDA998x are actually two I2C devices merged onto one piece + * of silicon: TDA9989 and TDA19989 combine the HDMI transmitter + * with a slightly modified TDA9950 CEC device. The CEC device + * is at the TDA9950 address, with the address pins strapped across + * to the TDA998x address pins. Hence, it always has the same + * offset. + */ + memset(&cec_info, 0, sizeof(cec_info)); + strlcpy(cec_info.type, "tda9950", sizeof(cec_info.type)); + cec_info.addr = priv->cec_addr; + cec_info.platform_data = &priv->cec_glue; + cec_info.irq = client->irq; + + priv->cec = i2c_new_device(client->adapter, &cec_info); + if (!priv->cec) { + ret = -ENODEV; + goto fail; + } + /* enable EDID read irq: */ reg_set(priv, REG_INT_FLAGS_2, INT_FLAGS_2_EDID_BLK_RD); @@ -1596,12 +1778,18 @@ static int tda998x_create(struct i2c_client *client, struct tda998x_priv *priv) tda998x_audio_codec_init(priv, &client->dev); return 0; + fail: /* if encoder_init fails, the encoder slave is never registered, * so cleanup here: */ i2c_unregister_device(priv->cec); - return -ENXIO; + if (priv->cec_notify) + cec_notifier_put(priv->cec_notify); + if (client->irq) + free_irq(client->irq, priv); +err_irq: + return ret; } static void tda998x_encoder_prepare(struct drm_encoder *encoder) diff --git a/drivers/gpu/drm/msm/disp/mdp4/mdp4_plane.c b/drivers/gpu/drm/msm/disp/mdp4/mdp4_plane.c index 7a1ad3af08e3..20e956e14c21 100644 --- a/drivers/gpu/drm/msm/disp/mdp4/mdp4_plane.c +++ b/drivers/gpu/drm/msm/disp/mdp4/mdp4_plane.c @@ -98,21 +98,6 @@ static const struct drm_plane_funcs mdp4_plane_funcs = { .atomic_destroy_state = drm_atomic_helper_plane_destroy_state, }; -static int mdp4_plane_prepare_fb(struct drm_plane *plane, - struct drm_plane_state *new_state) -{ - struct mdp4_plane *mdp4_plane = to_mdp4_plane(plane); - struct mdp4_kms *mdp4_kms = get_kms(plane); - struct msm_kms *kms = &mdp4_kms->base.base; - struct drm_framebuffer *fb = new_state->fb; - - if (!fb) - return 0; - - DBG("%s: prepare: FB[%u]", mdp4_plane->name, fb->base.id); - return msm_framebuffer_prepare(fb, kms->aspace); -} - static void mdp4_plane_cleanup_fb(struct drm_plane *plane, struct drm_plane_state *old_state) { @@ -152,7 +137,7 @@ static void mdp4_plane_atomic_update(struct drm_plane *plane, } static const struct drm_plane_helper_funcs mdp4_plane_helper_funcs = { - .prepare_fb = mdp4_plane_prepare_fb, + .prepare_fb = msm_atomic_prepare_fb, .cleanup_fb = mdp4_plane_cleanup_fb, .atomic_check = mdp4_plane_atomic_check, .atomic_update = mdp4_plane_atomic_update, diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5_crtc.c b/drivers/gpu/drm/msm/disp/mdp5/mdp5_crtc.c index 76b96081916f..10271359789e 100644 --- a/drivers/gpu/drm/msm/disp/mdp5/mdp5_crtc.c +++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_crtc.c @@ -430,6 +430,7 @@ static void mdp5_crtc_atomic_disable(struct drm_crtc *crtc, struct mdp5_crtc_state *mdp5_cstate = to_mdp5_crtc_state(crtc->state); struct mdp5_kms *mdp5_kms = get_kms(crtc); struct device *dev = &mdp5_kms->pdev->dev; + unsigned long flags; DBG("%s", crtc->name); @@ -445,6 +446,14 @@ static void mdp5_crtc_atomic_disable(struct drm_crtc *crtc, mdp_irq_unregister(&mdp5_kms->base, &mdp5_crtc->err); pm_runtime_put_sync(dev); + if (crtc->state->event && !crtc->state->active) { + WARN_ON(mdp5_crtc->event); + spin_lock_irqsave(&mdp5_kms->dev->event_lock, flags); + drm_crtc_send_vblank_event(crtc, crtc->state->event); + crtc->state->event = NULL; + spin_unlock_irqrestore(&mdp5_kms->dev->event_lock, flags); + } + mdp5_crtc->enabled = false; } diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5_kms.c b/drivers/gpu/drm/msm/disp/mdp5/mdp5_kms.c index 6d8e3a9a6fc0..6e12e275deba 100644 --- a/drivers/gpu/drm/msm/disp/mdp5/mdp5_kms.c +++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_kms.c @@ -70,60 +70,110 @@ static int mdp5_hw_init(struct msm_kms *kms) return 0; } -struct mdp5_state *mdp5_get_state(struct drm_atomic_state *s) +/* Global/shared object state funcs */ + +/* + * This is a helper that returns the private state currently in operation. + * Note that this would return the "old_state" if called in the atomic check + * path, and the "new_state" after the atomic swap has been done. + */ +struct mdp5_global_state * +mdp5_get_existing_global_state(struct mdp5_kms *mdp5_kms) +{ + return to_mdp5_global_state(mdp5_kms->glob_state.state); +} + +/* + * This acquires the modeset lock set aside for global state, creates + * a new duplicated private object state. + */ +struct mdp5_global_state *mdp5_get_global_state(struct drm_atomic_state *s) { struct msm_drm_private *priv = s->dev->dev_private; struct mdp5_kms *mdp5_kms = to_mdp5_kms(to_mdp_kms(priv->kms)); - struct msm_kms_state *state = to_kms_state(s); - struct mdp5_state *new_state; + struct drm_private_state *priv_state; int ret; - if (state->state) - return state->state; - - ret = drm_modeset_lock(&mdp5_kms->state_lock, s->acquire_ctx); + ret = drm_modeset_lock(&mdp5_kms->glob_state_lock, s->acquire_ctx); if (ret) return ERR_PTR(ret); - new_state = kmalloc(sizeof(*mdp5_kms->state), GFP_KERNEL); - if (!new_state) - return ERR_PTR(-ENOMEM); + priv_state = drm_atomic_get_private_obj_state(s, &mdp5_kms->glob_state); + if (IS_ERR(priv_state)) + return ERR_CAST(priv_state); - /* Copy state: */ - new_state->hwpipe = mdp5_kms->state->hwpipe; - new_state->hwmixer = mdp5_kms->state->hwmixer; - if (mdp5_kms->smp) - new_state->smp = mdp5_kms->state->smp; + return to_mdp5_global_state(priv_state); +} + +static struct drm_private_state * +mdp5_global_duplicate_state(struct drm_private_obj *obj) +{ + struct mdp5_global_state *state; + + state = kmemdup(obj->state, sizeof(*state), GFP_KERNEL); + if (!state) + return NULL; - state->state = new_state; + __drm_atomic_helper_private_obj_duplicate_state(obj, &state->base); - return new_state; + return &state->base; } -static void mdp5_swap_state(struct msm_kms *kms, struct drm_atomic_state *state) +static void mdp5_global_destroy_state(struct drm_private_obj *obj, + struct drm_private_state *state) { - struct mdp5_kms *mdp5_kms = to_mdp5_kms(to_mdp_kms(kms)); - swap(to_kms_state(state)->state, mdp5_kms->state); + struct mdp5_global_state *mdp5_state = to_mdp5_global_state(state); + + kfree(mdp5_state); +} + +static const struct drm_private_state_funcs mdp5_global_state_funcs = { + .atomic_duplicate_state = mdp5_global_duplicate_state, + .atomic_destroy_state = mdp5_global_destroy_state, +}; + +static int mdp5_global_obj_init(struct mdp5_kms *mdp5_kms) +{ + struct mdp5_global_state *state; + + drm_modeset_lock_init(&mdp5_kms->glob_state_lock); + + state = kzalloc(sizeof(*state), GFP_KERNEL); + if (!state) + return -ENOMEM; + + state->mdp5_kms = mdp5_kms; + + drm_atomic_private_obj_init(&mdp5_kms->glob_state, + &state->base, + &mdp5_global_state_funcs); + return 0; } static void mdp5_prepare_commit(struct msm_kms *kms, struct drm_atomic_state *state) { struct mdp5_kms *mdp5_kms = to_mdp5_kms(to_mdp_kms(kms)); struct device *dev = &mdp5_kms->pdev->dev; + struct mdp5_global_state *global_state; + + global_state = mdp5_get_existing_global_state(mdp5_kms); pm_runtime_get_sync(dev); if (mdp5_kms->smp) - mdp5_smp_prepare_commit(mdp5_kms->smp, &mdp5_kms->state->smp); + mdp5_smp_prepare_commit(mdp5_kms->smp, &global_state->smp); } static void mdp5_complete_commit(struct msm_kms *kms, struct drm_atomic_state *state) { struct mdp5_kms *mdp5_kms = to_mdp5_kms(to_mdp_kms(kms)); struct device *dev = &mdp5_kms->pdev->dev; + struct mdp5_global_state *global_state; + + global_state = mdp5_get_existing_global_state(mdp5_kms); if (mdp5_kms->smp) - mdp5_smp_complete_commit(mdp5_kms->smp, &mdp5_kms->state->smp); + mdp5_smp_complete_commit(mdp5_kms->smp, &global_state->smp); pm_runtime_put_sync(dev); } @@ -229,7 +279,6 @@ static const struct mdp_kms_funcs kms_funcs = { .irq = mdp5_irq, .enable_vblank = mdp5_enable_vblank, .disable_vblank = mdp5_disable_vblank, - .swap_state = mdp5_swap_state, .prepare_commit = mdp5_prepare_commit, .complete_commit = mdp5_complete_commit, .wait_for_crtc_commit_done = mdp5_wait_for_crtc_commit_done, @@ -727,7 +776,8 @@ static void mdp5_destroy(struct platform_device *pdev) if (mdp5_kms->rpm_enabled) pm_runtime_disable(&pdev->dev); - kfree(mdp5_kms->state); + drm_atomic_private_obj_fini(&mdp5_kms->glob_state); + drm_modeset_lock_fini(&mdp5_kms->glob_state_lock); } static int construct_pipes(struct mdp5_kms *mdp5_kms, int cnt, @@ -880,12 +930,9 @@ static int mdp5_init(struct platform_device *pdev, struct drm_device *dev) mdp5_kms->dev = dev; mdp5_kms->pdev = pdev; - drm_modeset_lock_init(&mdp5_kms->state_lock); - mdp5_kms->state = kzalloc(sizeof(*mdp5_kms->state), GFP_KERNEL); - if (!mdp5_kms->state) { - ret = -ENOMEM; + ret = mdp5_global_obj_init(mdp5_kms); + if (ret) goto fail; - } mdp5_kms->mmio = msm_ioremap(pdev, "mdp_phys", "MDP5"); if (IS_ERR(mdp5_kms->mmio)) { diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5_kms.h b/drivers/gpu/drm/msm/disp/mdp5/mdp5_kms.h index 425a03d213e5..854dfd30e829 100644 --- a/drivers/gpu/drm/msm/disp/mdp5/mdp5_kms.h +++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_kms.h @@ -28,8 +28,6 @@ #include "mdp5_ctl.h" #include "mdp5_smp.h" -struct mdp5_state; - struct mdp5_kms { struct mdp_kms base; @@ -49,11 +47,12 @@ struct mdp5_kms { struct mdp5_cfg_handler *cfg; uint32_t caps; /* MDP capabilities (MDP_CAP_XXX bits) */ - /** - * Global atomic state. Do not access directly, use mdp5_get_state() + /* + * Global private object state, Do not access directly, use + * mdp5_global_get_state() */ - struct mdp5_state *state; - struct drm_modeset_lock state_lock; + struct drm_modeset_lock glob_state_lock; + struct drm_private_obj glob_state; struct mdp5_smp *smp; struct mdp5_ctl_manager *ctlm; @@ -81,19 +80,23 @@ struct mdp5_kms { }; #define to_mdp5_kms(x) container_of(x, struct mdp5_kms, base) -/* Global atomic state for tracking resources that are shared across +/* Global private object state for tracking resources that are shared across * multiple kms objects (planes/crtcs/etc). - * - * For atomic updates which require modifying global state, */ -struct mdp5_state { +#define to_mdp5_global_state(x) container_of(x, struct mdp5_global_state, base) +struct mdp5_global_state { + struct drm_private_state base; + + struct drm_atomic_state *state; + struct mdp5_kms *mdp5_kms; + struct mdp5_hw_pipe_state hwpipe; struct mdp5_hw_mixer_state hwmixer; struct mdp5_smp_state smp; }; -struct mdp5_state *__must_check -mdp5_get_state(struct drm_atomic_state *s); +struct mdp5_global_state * mdp5_get_existing_global_state(struct mdp5_kms *mdp5_kms); +struct mdp5_global_state *__must_check mdp5_get_global_state(struct drm_atomic_state *s); /* Atomic plane state. Subclasses the base drm_plane_state in order to * track assigned hwpipe and hw specific state. diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5_mixer.c b/drivers/gpu/drm/msm/disp/mdp5/mdp5_mixer.c index 8a00991f03c7..113e6b569562 100644 --- a/drivers/gpu/drm/msm/disp/mdp5/mdp5_mixer.c +++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_mixer.c @@ -52,14 +52,14 @@ int mdp5_mixer_assign(struct drm_atomic_state *s, struct drm_crtc *crtc, { struct msm_drm_private *priv = s->dev->dev_private; struct mdp5_kms *mdp5_kms = to_mdp5_kms(to_mdp_kms(priv->kms)); - struct mdp5_state *state = mdp5_get_state(s); + struct mdp5_global_state *global_state = mdp5_get_global_state(s); struct mdp5_hw_mixer_state *new_state; int i; - if (IS_ERR(state)) - return PTR_ERR(state); + if (IS_ERR(global_state)) + return PTR_ERR(global_state); - new_state = &state->hwmixer; + new_state = &global_state->hwmixer; for (i = 0; i < mdp5_kms->num_hwmixers; i++) { struct mdp5_hw_mixer *cur = mdp5_kms->hwmixers[i]; @@ -129,8 +129,8 @@ int mdp5_mixer_assign(struct drm_atomic_state *s, struct drm_crtc *crtc, void mdp5_mixer_release(struct drm_atomic_state *s, struct mdp5_hw_mixer *mixer) { - struct mdp5_state *state = mdp5_get_state(s); - struct mdp5_hw_mixer_state *new_state = &state->hwmixer; + struct mdp5_global_state *global_state = mdp5_get_global_state(s); + struct mdp5_hw_mixer_state *new_state = &global_state->hwmixer; if (!mixer) return; diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5_pipe.c b/drivers/gpu/drm/msm/disp/mdp5/mdp5_pipe.c index ff52c49095f9..1ef26bc63163 100644 --- a/drivers/gpu/drm/msm/disp/mdp5/mdp5_pipe.c +++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_pipe.c @@ -24,17 +24,19 @@ int mdp5_pipe_assign(struct drm_atomic_state *s, struct drm_plane *plane, { struct msm_drm_private *priv = s->dev->dev_private; struct mdp5_kms *mdp5_kms = to_mdp5_kms(to_mdp_kms(priv->kms)); - struct mdp5_state *state; + struct mdp5_global_state *new_global_state, *old_global_state; struct mdp5_hw_pipe_state *old_state, *new_state; int i, j; - state = mdp5_get_state(s); - if (IS_ERR(state)) - return PTR_ERR(state); + new_global_state = mdp5_get_global_state(s); + if (IS_ERR(new_global_state)) + return PTR_ERR(new_global_state); - /* grab old_state after mdp5_get_state(), since now we hold lock: */ - old_state = &mdp5_kms->state->hwpipe; - new_state = &state->hwpipe; + /* grab old_state after mdp5_get_global_state(), since now we hold lock: */ + old_global_state = mdp5_get_existing_global_state(mdp5_kms); + + old_state = &old_global_state->hwpipe; + new_state = &new_global_state->hwpipe; for (i = 0; i < mdp5_kms->num_hwpipes; i++) { struct mdp5_hw_pipe *cur = mdp5_kms->hwpipes[i]; @@ -107,7 +109,7 @@ int mdp5_pipe_assign(struct drm_atomic_state *s, struct drm_plane *plane, WARN_ON(r_hwpipe); DBG("%s: alloc SMP blocks", (*hwpipe)->name); - ret = mdp5_smp_assign(mdp5_kms->smp, &state->smp, + ret = mdp5_smp_assign(mdp5_kms->smp, &new_global_state->smp, (*hwpipe)->pipe, blkcfg); if (ret) return -ENOMEM; @@ -132,7 +134,7 @@ void mdp5_pipe_release(struct drm_atomic_state *s, struct mdp5_hw_pipe *hwpipe) { struct msm_drm_private *priv = s->dev->dev_private; struct mdp5_kms *mdp5_kms = to_mdp5_kms(to_mdp_kms(priv->kms)); - struct mdp5_state *state = mdp5_get_state(s); + struct mdp5_global_state *state = mdp5_get_global_state(s); struct mdp5_hw_pipe_state *new_state = &state->hwpipe; if (!hwpipe) diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5_plane.c b/drivers/gpu/drm/msm/disp/mdp5/mdp5_plane.c index a9f31da7d45a..e09bc53a0e65 100644 --- a/drivers/gpu/drm/msm/disp/mdp5/mdp5_plane.c +++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_plane.c @@ -245,20 +245,6 @@ static const struct drm_plane_funcs mdp5_plane_funcs = { .atomic_print_state = mdp5_plane_atomic_print_state, }; -static int mdp5_plane_prepare_fb(struct drm_plane *plane, - struct drm_plane_state *new_state) -{ - struct mdp5_kms *mdp5_kms = get_kms(plane); - struct msm_kms *kms = &mdp5_kms->base.base; - struct drm_framebuffer *fb = new_state->fb; - - if (!new_state->fb) - return 0; - - DBG("%s: prepare: FB[%u]", plane->name, fb->base.id); - return msm_framebuffer_prepare(fb, kms->aspace); -} - static void mdp5_plane_cleanup_fb(struct drm_plane *plane, struct drm_plane_state *old_state) { @@ -543,7 +529,7 @@ static void mdp5_plane_atomic_async_update(struct drm_plane *plane, } static const struct drm_plane_helper_funcs mdp5_plane_helper_funcs = { - .prepare_fb = mdp5_plane_prepare_fb, + .prepare_fb = msm_atomic_prepare_fb, .cleanup_fb = mdp5_plane_cleanup_fb, .atomic_check = mdp5_plane_atomic_check, .atomic_update = mdp5_plane_atomic_update, diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5_smp.c b/drivers/gpu/drm/msm/disp/mdp5/mdp5_smp.c index ae4983d9d0a5..96c2b828dba4 100644 --- a/drivers/gpu/drm/msm/disp/mdp5/mdp5_smp.c +++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_smp.c @@ -340,17 +340,20 @@ void mdp5_smp_dump(struct mdp5_smp *smp, struct drm_printer *p) struct mdp5_kms *mdp5_kms = get_kms(smp); struct mdp5_hw_pipe_state *hwpstate; struct mdp5_smp_state *state; + struct mdp5_global_state *global_state; int total = 0, i, j; drm_printf(p, "name\tinuse\tplane\n"); drm_printf(p, "----\t-----\t-----\n"); if (drm_can_sleep()) - drm_modeset_lock(&mdp5_kms->state_lock, NULL); + drm_modeset_lock(&mdp5_kms->glob_state_lock, NULL); + + global_state = mdp5_get_existing_global_state(mdp5_kms); /* grab these *after* we hold the state_lock */ - hwpstate = &mdp5_kms->state->hwpipe; - state = &mdp5_kms->state->smp; + hwpstate = &global_state->hwpipe; + state = &global_state->smp; for (i = 0; i < mdp5_kms->num_hwpipes; i++) { struct mdp5_hw_pipe *hwpipe = mdp5_kms->hwpipes[i]; @@ -374,7 +377,7 @@ void mdp5_smp_dump(struct mdp5_smp *smp, struct drm_printer *p) bitmap_weight(state->state, smp->blk_cnt)); if (drm_can_sleep()) - drm_modeset_unlock(&mdp5_kms->state_lock); + drm_modeset_unlock(&mdp5_kms->glob_state_lock); } void mdp5_smp_destroy(struct mdp5_smp *smp) @@ -384,7 +387,8 @@ void mdp5_smp_destroy(struct mdp5_smp *smp) struct mdp5_smp *mdp5_smp_init(struct mdp5_kms *mdp5_kms, const struct mdp5_smp_block *cfg) { - struct mdp5_smp_state *state = &mdp5_kms->state->smp; + struct mdp5_smp_state *state; + struct mdp5_global_state *global_state; struct mdp5_smp *smp = NULL; int ret; @@ -398,6 +402,9 @@ struct mdp5_smp *mdp5_smp_init(struct mdp5_kms *mdp5_kms, const struct mdp5_smp_ smp->blk_cnt = cfg->mmb_count; smp->blk_size = cfg->mmb_size; + global_state = mdp5_get_existing_global_state(mdp5_kms); + state = &global_state->smp; + /* statically tied MMBs cannot be re-allocated: */ bitmap_copy(state->state, cfg->reserved_state, smp->blk_cnt); memcpy(smp->reserved, cfg->reserved, sizeof(smp->reserved)); diff --git a/drivers/gpu/drm/msm/dsi/dsi_host.c b/drivers/gpu/drm/msm/dsi/dsi_host.c index 8baba30d6c65..2f1a2780658a 100644 --- a/drivers/gpu/drm/msm/dsi/dsi_host.c +++ b/drivers/gpu/drm/msm/dsi/dsi_host.c @@ -1036,7 +1036,6 @@ static int dsi_tx_buf_alloc(struct msm_dsi_host *msm_host, int size) ret = msm_gem_get_iova(msm_host->tx_gem_obj, priv->kms->aspace, &iova); - mutex_unlock(&dev->struct_mutex); if (ret) { pr_err("%s: failed to get iova, %d\n", __func__, ret); return ret; @@ -1067,9 +1066,20 @@ static int dsi_tx_buf_alloc(struct msm_dsi_host *msm_host, int size) static void dsi_tx_buf_free(struct msm_dsi_host *msm_host) { struct drm_device *dev = msm_host->dev; + struct msm_drm_private *priv; + /* + * This is possible if we're tearing down before we've had a chance to + * fully initialize. A very real possibility if our probe is deferred, + * in which case we'll hit msm_dsi_host_destroy() without having run + * through the dsi_tx_buf_alloc(). + */ + if (!dev) + return; + + priv = dev->dev_private; if (msm_host->tx_gem_obj) { - msm_gem_put_iova(msm_host->tx_gem_obj, 0); + msm_gem_put_iova(msm_host->tx_gem_obj, priv->kms->aspace); drm_gem_object_put_unlocked(msm_host->tx_gem_obj); msm_host->tx_gem_obj = NULL; } diff --git a/drivers/gpu/drm/msm/msm_atomic.c b/drivers/gpu/drm/msm/msm_atomic.c index bf5f8c39f34d..f0635c3da7f4 100644 --- a/drivers/gpu/drm/msm/msm_atomic.c +++ b/drivers/gpu/drm/msm/msm_atomic.c @@ -16,69 +16,8 @@ */ #include "msm_drv.h" -#include "msm_kms.h" #include "msm_gem.h" -#include "msm_fence.h" - -struct msm_commit { - struct drm_device *dev; - struct drm_atomic_state *state; - struct work_struct work; - uint32_t crtc_mask; -}; - -static void commit_worker(struct work_struct *work); - -/* block until specified crtcs are no longer pending update, and - * atomically mark them as pending update - */ -static int start_atomic(struct msm_drm_private *priv, uint32_t crtc_mask) -{ - int ret; - - spin_lock(&priv->pending_crtcs_event.lock); - ret = wait_event_interruptible_locked(priv->pending_crtcs_event, - !(priv->pending_crtcs & crtc_mask)); - if (ret == 0) { - DBG("start: %08x", crtc_mask); - priv->pending_crtcs |= crtc_mask; - } - spin_unlock(&priv->pending_crtcs_event.lock); - - return ret; -} - -/* clear specified crtcs (no longer pending update) - */ -static void end_atomic(struct msm_drm_private *priv, uint32_t crtc_mask) -{ - spin_lock(&priv->pending_crtcs_event.lock); - DBG("end: %08x", crtc_mask); - priv->pending_crtcs &= ~crtc_mask; - wake_up_all_locked(&priv->pending_crtcs_event); - spin_unlock(&priv->pending_crtcs_event.lock); -} - -static struct msm_commit *commit_init(struct drm_atomic_state *state) -{ - struct msm_commit *c = kzalloc(sizeof(*c), GFP_KERNEL); - - if (!c) - return NULL; - - c->dev = state->dev; - c->state = state; - - INIT_WORK(&c->work, commit_worker); - - return c; -} - -static void commit_destroy(struct msm_commit *c) -{ - end_atomic(c->dev->dev_private, c->crtc_mask); - kfree(c); -} +#include "msm_kms.h" static void msm_atomic_wait_for_commit_done(struct drm_device *dev, struct drm_atomic_state *old_state) @@ -97,195 +36,48 @@ static void msm_atomic_wait_for_commit_done(struct drm_device *dev, } } -/* The (potentially) asynchronous part of the commit. At this point - * nothing can fail short of armageddon. - */ -static void complete_commit(struct msm_commit *c, bool async) +int msm_atomic_prepare_fb(struct drm_plane *plane, + struct drm_plane_state *new_state) { - struct drm_atomic_state *state = c->state; - struct drm_device *dev = state->dev; - struct msm_drm_private *priv = dev->dev_private; + struct msm_drm_private *priv = plane->dev->dev_private; struct msm_kms *kms = priv->kms; + struct drm_gem_object *obj; + struct msm_gem_object *msm_obj; + struct dma_fence *fence; - drm_atomic_helper_wait_for_fences(dev, state, false); - - kms->funcs->prepare_commit(kms, state); - - drm_atomic_helper_commit_modeset_disables(dev, state); - - drm_atomic_helper_commit_planes(dev, state, 0); - - drm_atomic_helper_commit_modeset_enables(dev, state); - - /* NOTE: _wait_for_vblanks() only waits for vblank on - * enabled CRTCs. So we end up faulting when disabling - * due to (potentially) unref'ing the outgoing fb's - * before the vblank when the disable has latched. - * - * But if it did wait on disabled (or newly disabled) - * CRTCs, that would be racy (ie. we could have missed - * the irq. We need some way to poll for pipe shut - * down. Or just live with occasionally hitting the - * timeout in the CRTC disable path (which really should - * not be critical path) - */ - - msm_atomic_wait_for_commit_done(dev, state); - - drm_atomic_helper_cleanup_planes(dev, state); + if (!new_state->fb) + return 0; - kms->funcs->complete_commit(kms, state); + obj = msm_framebuffer_bo(new_state->fb, 0); + msm_obj = to_msm_bo(obj); + fence = reservation_object_get_excl_rcu(msm_obj->resv); - drm_atomic_state_put(state); + drm_atomic_set_fence_for_plane(new_state, fence); - commit_destroy(c); -} - -static void commit_worker(struct work_struct *work) -{ - complete_commit(container_of(work, struct msm_commit, work), true); + return msm_framebuffer_prepare(new_state->fb, kms->aspace); } -/** - * drm_atomic_helper_commit - commit validated state object - * @dev: DRM device - * @state: the driver state object - * @nonblock: nonblocking commit - * - * This function commits a with drm_atomic_helper_check() pre-validated state - * object. This can still fail when e.g. the framebuffer reservation fails. - * - * RETURNS - * Zero for success or -errno. - */ -int msm_atomic_commit(struct drm_device *dev, - struct drm_atomic_state *state, bool nonblock) +void msm_atomic_commit_tail(struct drm_atomic_state *state) { + struct drm_device *dev = state->dev; struct msm_drm_private *priv = dev->dev_private; - struct msm_commit *c; - struct drm_crtc *crtc; - struct drm_crtc_state *crtc_state; - struct drm_plane *plane; - struct drm_plane_state *old_plane_state, *new_plane_state; - int i, ret; - - ret = drm_atomic_helper_prepare_planes(dev, state); - if (ret) - return ret; - - /* - * Note that plane->atomic_async_check() should fail if we need - * to re-assign hwpipe or anything that touches global atomic - * state, so we'll never go down the async update path in those - * cases. - */ - if (state->async_update) { - drm_atomic_helper_async_commit(dev, state); - drm_atomic_helper_cleanup_planes(dev, state); - return 0; - } - - c = commit_init(state); - if (!c) { - ret = -ENOMEM; - goto error; - } - - /* - * Figure out what crtcs we have: - */ - for_each_new_crtc_in_state(state, crtc, crtc_state, i) - c->crtc_mask |= drm_crtc_mask(crtc); - - /* - * Figure out what fence to wait for: - */ - for_each_oldnew_plane_in_state(state, plane, old_plane_state, new_plane_state, i) { - if ((new_plane_state->fb != old_plane_state->fb) && new_plane_state->fb) { - struct drm_gem_object *obj = msm_framebuffer_bo(new_plane_state->fb, 0); - struct msm_gem_object *msm_obj = to_msm_bo(obj); - struct dma_fence *fence = reservation_object_get_excl_rcu(msm_obj->resv); + struct msm_kms *kms = priv->kms; - drm_atomic_set_fence_for_plane(new_plane_state, fence); - } - } + kms->funcs->prepare_commit(kms, state); - /* - * Wait for pending updates on any of the same crtc's and then - * mark our set of crtc's as busy: - */ - ret = start_atomic(dev->dev_private, c->crtc_mask); - if (ret) - goto err_free; + drm_atomic_helper_commit_modeset_disables(dev, state); - BUG_ON(drm_atomic_helper_swap_state(state, false) < 0); + drm_atomic_helper_commit_planes(dev, state, 0); - /* - * This is the point of no return - everything below never fails except - * when the hw goes bonghits. Which means we can commit the new state on - * the software side now. - * - * swap driver private state while still holding state_lock - */ - if (to_kms_state(state)->state) - priv->kms->funcs->swap_state(priv->kms, state); + drm_atomic_helper_commit_modeset_enables(dev, state); - /* - * Everything below can be run asynchronously without the need to grab - * any modeset locks at all under one conditions: It must be guaranteed - * that the asynchronous work has either been cancelled (if the driver - * supports it, which at least requires that the framebuffers get - * cleaned up with drm_atomic_helper_cleanup_planes()) or completed - * before the new state gets committed on the software side with - * drm_atomic_helper_swap_state(). - * - * This scheme allows new atomic state updates to be prepared and - * checked in parallel to the asynchronous completion of the previous - * update. Which is important since compositors need to figure out the - * composition of the next frame right after having submitted the - * current layout. - */ + msm_atomic_wait_for_commit_done(dev, state); - drm_atomic_state_get(state); - if (nonblock) { - queue_work(priv->atomic_wq, &c->work); - return 0; - } + kms->funcs->complete_commit(kms, state); - complete_commit(c, false); + drm_atomic_helper_wait_for_vblanks(dev, state); - return 0; + drm_atomic_helper_commit_hw_done(state); -err_free: - kfree(c); -error: drm_atomic_helper_cleanup_planes(dev, state); - return ret; -} - -struct drm_atomic_state *msm_atomic_state_alloc(struct drm_device *dev) -{ - struct msm_kms_state *state = kzalloc(sizeof(*state), GFP_KERNEL); - - if (!state || drm_atomic_state_init(dev, &state->base) < 0) { - kfree(state); - return NULL; - } - - return &state->base; -} - -void msm_atomic_state_clear(struct drm_atomic_state *s) -{ - struct msm_kms_state *state = to_kms_state(s); - drm_atomic_state_default_clear(&state->base); - kfree(state->state); - state->state = NULL; -} - -void msm_atomic_state_free(struct drm_atomic_state *state) -{ - kfree(to_kms_state(state)->state); - drm_atomic_state_default_release(state); - kfree(state); } diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c index 30cd514d8f7c..021a0b6f9a59 100644 --- a/drivers/gpu/drm/msm/msm_drv.c +++ b/drivers/gpu/drm/msm/msm_drv.c @@ -41,10 +41,11 @@ static const struct drm_mode_config_funcs mode_config_funcs = { .fb_create = msm_framebuffer_create, .output_poll_changed = drm_fb_helper_output_poll_changed, .atomic_check = drm_atomic_helper_check, - .atomic_commit = msm_atomic_commit, - .atomic_state_alloc = msm_atomic_state_alloc, - .atomic_state_clear = msm_atomic_state_clear, - .atomic_state_free = msm_atomic_state_free, + .atomic_commit = drm_atomic_helper_commit, +}; + +static const struct drm_mode_config_helper_funcs mode_config_helper_funcs = { + .atomic_commit_tail = msm_atomic_commit_tail, }; #ifdef CONFIG_DRM_MSM_REGISTER_LOGGING @@ -384,7 +385,6 @@ static int msm_drm_init(struct device *dev, struct drm_driver *drv) priv->wq = alloc_ordered_workqueue("msm", 0); priv->atomic_wq = alloc_ordered_workqueue("msm:atomic", 0); - init_waitqueue_head(&priv->pending_crtcs_event); INIT_LIST_HEAD(&priv->inactive_list); INIT_LIST_HEAD(&priv->vblank_ctrl.event_list); @@ -442,6 +442,7 @@ static int msm_drm_init(struct device *dev, struct drm_driver *drv) } ddev->mode_config.funcs = &mode_config_funcs; + ddev->mode_config.helper_private = &mode_config_helper_funcs; ret = drm_vblank_init(ddev, priv->num_crtcs); if (ret < 0) { diff --git a/drivers/gpu/drm/msm/msm_drv.h b/drivers/gpu/drm/msm/msm_drv.h index 48ed5b9a8580..b2da1fbf81e0 100644 --- a/drivers/gpu/drm/msm/msm_drv.h +++ b/drivers/gpu/drm/msm/msm_drv.h @@ -117,10 +117,6 @@ struct msm_drm_private { struct workqueue_struct *wq; struct workqueue_struct *atomic_wq; - /* crtcs pending async atomic updates: */ - uint32_t pending_crtcs; - wait_queue_head_t pending_crtcs_event; - unsigned int num_planes; struct drm_plane *planes[16]; @@ -160,8 +156,9 @@ struct msm_format { uint32_t pixel_format; }; -int msm_atomic_commit(struct drm_device *dev, - struct drm_atomic_state *state, bool nonblock); +int msm_atomic_prepare_fb(struct drm_plane *plane, + struct drm_plane_state *new_state); +void msm_atomic_commit_tail(struct drm_atomic_state *state); struct drm_atomic_state *msm_atomic_state_alloc(struct drm_device *dev); void msm_atomic_state_clear(struct drm_atomic_state *state); void msm_atomic_state_free(struct drm_atomic_state *state); diff --git a/drivers/gpu/drm/msm/msm_kms.h b/drivers/gpu/drm/msm/msm_kms.h index aaa329dc020e..dfd92947de2c 100644 --- a/drivers/gpu/drm/msm/msm_kms.h +++ b/drivers/gpu/drm/msm/msm_kms.h @@ -40,8 +40,6 @@ struct msm_kms_funcs { irqreturn_t (*irq)(struct msm_kms *kms); int (*enable_vblank)(struct msm_kms *kms, struct drm_crtc *crtc); void (*disable_vblank)(struct msm_kms *kms, struct drm_crtc *crtc); - /* swap global atomic state: */ - void (*swap_state)(struct msm_kms *kms, struct drm_atomic_state *state); /* modeset, bracketing atomic_commit(): */ void (*prepare_commit)(struct msm_kms *kms, struct drm_atomic_state *state); void (*complete_commit)(struct msm_kms *kms, struct drm_atomic_state *state); @@ -80,18 +78,6 @@ struct msm_kms { struct msm_gem_address_space *aspace; }; -/** - * Subclass of drm_atomic_state, to allow kms backend to have driver - * private global state. The kms backend can do whatever it wants - * with the ->state ptr. On ->atomic_state_clear() the ->state ptr - * is kfree'd and set back to NULL. - */ -struct msm_kms_state { - struct drm_atomic_state base; - void *state; -}; -#define to_kms_state(x) container_of(x, struct msm_kms_state, base) - static inline void msm_kms_init(struct msm_kms *kms, const struct msm_kms_funcs *funcs) { diff --git a/drivers/gpu/drm/nouveau/Kbuild b/drivers/gpu/drm/nouveau/Kbuild index 9c0c650655e9..b17843dd050d 100644 --- a/drivers/gpu/drm/nouveau/Kbuild +++ b/drivers/gpu/drm/nouveau/Kbuild @@ -38,14 +38,16 @@ nouveau-y += nouveau_vmm.o # DRM - modesetting nouveau-$(CONFIG_DRM_NOUVEAU_BACKLIGHT) += nouveau_backlight.o +nouveau-y += nouveau_bios.o nouveau-y += nouveau_connector.o nouveau-y += nouveau_display.o -nouveau-y += nv50_display.o nouveau-y += nouveau_dp.o nouveau-y += nouveau_fbcon.o nouveau-y += nv04_fbcon.o nouveau-y += nv50_fbcon.o nouveau-y += nvc0_fbcon.o +include $(src)/dispnv04/Kbuild +include $(src)/dispnv50/Kbuild # DRM - command submission nouveau-y += nouveau_abi16.o @@ -59,8 +61,4 @@ nouveau-y += nv50_fence.o nouveau-y += nv84_fence.o nouveau-y += nvc0_fence.o -# DRM - prehistoric modesetting (NV04-G7x) -nouveau-y += nouveau_bios.o -include $(src)/dispnv04/Kbuild - obj-$(CONFIG_DRM_NOUVEAU) += nouveau.o diff --git a/drivers/gpu/drm/nouveau/dispnv50/Kbuild b/drivers/gpu/drm/nouveau/dispnv50/Kbuild new file mode 100644 index 000000000000..849b0f45afb8 --- /dev/null +++ b/drivers/gpu/drm/nouveau/dispnv50/Kbuild @@ -0,0 +1,51 @@ +nouveau-y += dispnv50/disp.o +nouveau-y += dispnv50/lut.o + +nouveau-y += dispnv50/core.o +nouveau-y += dispnv50/core507d.o +nouveau-y += dispnv50/core827d.o +nouveau-y += dispnv50/core907d.o +nouveau-y += dispnv50/core917d.o +nouveau-y += dispnv50/corec37d.o + +nouveau-y += dispnv50/dac507d.o +nouveau-y += dispnv50/dac907d.o + +nouveau-y += dispnv50/pior507d.o + +nouveau-y += dispnv50/sor507d.o +nouveau-y += dispnv50/sor907d.o +nouveau-y += dispnv50/sorc37d.o + +nouveau-y += dispnv50/head.o +nouveau-y += dispnv50/head507d.o +nouveau-y += dispnv50/head827d.o +nouveau-y += dispnv50/head907d.o +nouveau-y += dispnv50/head917d.o +nouveau-y += dispnv50/headc37d.o + +nouveau-y += dispnv50/wimm.o +nouveau-y += dispnv50/wimmc37b.o + +nouveau-y += dispnv50/wndw.o +nouveau-y += dispnv50/wndwc37e.o + +nouveau-y += dispnv50/base.o +nouveau-y += dispnv50/base507c.o +nouveau-y += dispnv50/base827c.o +nouveau-y += dispnv50/base907c.o +nouveau-y += dispnv50/base917c.o + +nouveau-y += dispnv50/curs.o +nouveau-y += dispnv50/curs507a.o +nouveau-y += dispnv50/curs907a.o +nouveau-y += dispnv50/cursc37a.o + +nouveau-y += dispnv50/oimm.o +nouveau-y += dispnv50/oimm507b.o + +nouveau-y += dispnv50/ovly.o +nouveau-y += dispnv50/ovly507e.o +nouveau-y += dispnv50/ovly827e.o +nouveau-y += dispnv50/ovly907e.o +nouveau-y += dispnv50/ovly917e.o diff --git a/drivers/gpu/drm/nouveau/dispnv50/atom.h b/drivers/gpu/drm/nouveau/dispnv50/atom.h new file mode 100644 index 000000000000..908feb1fc60f --- /dev/null +++ b/drivers/gpu/drm/nouveau/dispnv50/atom.h @@ -0,0 +1,222 @@ +#ifndef __NV50_KMS_ATOM_H__ +#define __NV50_KMS_ATOM_H__ +#define nv50_atom(p) container_of((p), struct nv50_atom, state) +#include <drm/drm_atomic.h> + +struct nv50_atom { + struct drm_atomic_state state; + + struct list_head outp; + bool lock_core; + bool flush_disable; +}; + +#define nv50_head_atom(p) container_of((p), struct nv50_head_atom, state) + +struct nv50_head_atom { + struct drm_crtc_state state; + + struct { + u32 mask; + u32 olut; + } wndw; + + struct { + u16 iW; + u16 iH; + u16 oW; + u16 oH; + } view; + + struct nv50_head_mode { + bool interlace; + u32 clock; + struct { + u16 active; + u16 synce; + u16 blanke; + u16 blanks; + } h; + struct { + u32 active; + u16 synce; + u16 blanke; + u16 blanks; + u16 blank2s; + u16 blank2e; + u16 blankus; + } v; + } mode; + + struct { + bool visible; + u32 handle; + u64 offset:40; + u8 buffer:1; + u8 mode:4; + u8 size:2; + u8 range:2; + u8 output_mode:2; + } olut; + + struct { + bool visible; + u32 handle; + u64 offset:40; + u8 format; + u8 kind:7; + u8 layout:1; + u8 blockh:4; + u16 blocks:12; + u32 pitch:20; + u16 x; + u16 y; + u16 w; + u16 h; + } core; + + struct { + bool visible; + u32 handle; + u64 offset:40; + u8 layout:2; + u8 format:8; + } curs; + + struct { + u8 depth; + u8 cpp; + u16 x; + u16 y; + u16 w; + u16 h; + } base; + + struct { + u8 cpp; + } ovly; + + struct { + bool enable:1; + u8 bits:2; + u8 mode:4; + } dither; + + struct { + struct { + u16 cos:12; + u16 sin:12; + } sat; + } procamp; + + struct { + u8 nhsync:1; + u8 nvsync:1; + u8 depth:4; + } or; + + union nv50_head_atom_mask { + struct { + bool olut:1; + bool core:1; + bool curs:1; + bool view:1; + bool mode:1; + bool base:1; + bool ovly:1; + bool dither:1; + bool procamp:1; + bool or:1; + }; + u16 mask; + } set, clr; +}; + +static inline struct nv50_head_atom * +nv50_head_atom_get(struct drm_atomic_state *state, struct drm_crtc *crtc) +{ + struct drm_crtc_state *statec = drm_atomic_get_crtc_state(state, crtc); + if (IS_ERR(statec)) + return (void *)statec; + return nv50_head_atom(statec); +} + +#define nv50_wndw_atom(p) container_of((p), struct nv50_wndw_atom, state) + +struct nv50_wndw_atom { + struct drm_plane_state state; + + struct drm_property_blob *ilut; + bool visible; + + struct { + u32 handle; + u16 offset:12; + bool awaken:1; + } ntfy; + + struct { + u32 handle; + u16 offset:12; + u32 acquire; + u32 release; + } sema; + + struct { + u32 handle; + struct { + u64 offset:40; + u8 buffer:1; + u8 enable:2; + u8 mode:4; + u8 size:2; + u8 range:2; + u8 output_mode:2; + } i; + } xlut; + + struct { + u8 mode:2; + u8 interval:4; + + u8 colorspace:2; + u8 format; + u8 kind:7; + u8 layout:1; + u8 blockh:4; + u16 blocks[3]; + u32 pitch[3]; + u16 w; + u16 h; + + u32 handle[6]; + u64 offset[6]; + } image; + + struct { + u16 sx; + u16 sy; + u16 sw; + u16 sh; + u16 dw; + u16 dh; + } scale; + + struct { + u16 x; + u16 y; + } point; + + union nv50_wndw_atom_mask { + struct { + bool ntfy:1; + bool sema:1; + bool xlut:1; + bool image:1; + bool scale:1; + bool point:1; + }; + u8 mask; + } set, clr; +}; +#endif diff --git a/drivers/gpu/drm/nouveau/dispnv50/base.c b/drivers/gpu/drm/nouveau/dispnv50/base.c new file mode 100644 index 000000000000..7c752acf2b48 --- /dev/null +++ b/drivers/gpu/drm/nouveau/dispnv50/base.c @@ -0,0 +1,53 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "base.h" + +#include <nvif/class.h> + +int +nv50_base_new(struct nouveau_drm *drm, int head, struct nv50_wndw **pwndw) +{ + struct { + s32 oclass; + int version; + int (*new)(struct nouveau_drm *, int, s32, struct nv50_wndw **); + } bases[] = { + { GK110_DISP_BASE_CHANNEL_DMA, 0, base917c_new }, + { GK104_DISP_BASE_CHANNEL_DMA, 0, base917c_new }, + { GF110_DISP_BASE_CHANNEL_DMA, 0, base907c_new }, + { GT214_DISP_BASE_CHANNEL_DMA, 0, base827c_new }, + { GT200_DISP_BASE_CHANNEL_DMA, 0, base827c_new }, + { G82_DISP_BASE_CHANNEL_DMA, 0, base827c_new }, + { NV50_DISP_BASE_CHANNEL_DMA, 0, base507c_new }, + {} + }; + struct nv50_disp *disp = nv50_disp(drm->dev); + int cid; + + cid = nvif_mclass(&disp->disp->object, bases); + if (cid < 0) { + NV_ERROR(drm, "No supported base class\n"); + return cid; + } + + return bases[cid].new(drm, head, bases[cid].oclass, pwndw); +} diff --git a/drivers/gpu/drm/nouveau/dispnv50/base.h b/drivers/gpu/drm/nouveau/dispnv50/base.h new file mode 100644 index 000000000000..e7f14f230f35 --- /dev/null +++ b/drivers/gpu/drm/nouveau/dispnv50/base.h @@ -0,0 +1,31 @@ +#ifndef __NV50_KMS_BASE_H__ +#define __NV50_KMS_BASE_H__ +#include "wndw.h" + +int base507c_new(struct nouveau_drm *, int, s32, struct nv50_wndw **); +int base507c_new_(const struct nv50_wndw_func *, const u32 *format, + struct nouveau_drm *, int head, s32 oclass, + u32 interlock_data, struct nv50_wndw **); +extern const u32 base507c_format[]; +int base507c_acquire(struct nv50_wndw *, struct nv50_wndw_atom *, + struct nv50_head_atom *); +void base507c_release(struct nv50_wndw *, struct nv50_wndw_atom *, + struct nv50_head_atom *); +void base507c_sema_set(struct nv50_wndw *, struct nv50_wndw_atom *); +void base507c_sema_clr(struct nv50_wndw *); +void base507c_ntfy_set(struct nv50_wndw *, struct nv50_wndw_atom *); +void base507c_ntfy_clr(struct nv50_wndw *); +void base507c_xlut_set(struct nv50_wndw *, struct nv50_wndw_atom *); +void base507c_xlut_clr(struct nv50_wndw *); +void base507c_image_clr(struct nv50_wndw *); +void base507c_update(struct nv50_wndw *, u32 *); + +int base827c_new(struct nouveau_drm *, int, s32, struct nv50_wndw **); + +int base907c_new(struct nouveau_drm *, int, s32, struct nv50_wndw **); +extern const struct nv50_wndw_func base907c; + +int base917c_new(struct nouveau_drm *, int, s32, struct nv50_wndw **); + +int nv50_base_new(struct nouveau_drm *, int head, struct nv50_wndw **); +#endif diff --git a/drivers/gpu/drm/nouveau/dispnv50/base507c.c b/drivers/gpu/drm/nouveau/dispnv50/base507c.c new file mode 100644 index 000000000000..d5e295ca2caa --- /dev/null +++ b/drivers/gpu/drm/nouveau/dispnv50/base507c.c @@ -0,0 +1,286 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "base.h" + +#include <nvif/cl507c.h> +#include <nvif/event.h> + +#include <drm/drm_atomic_helper.h> +#include <drm/drm_plane_helper.h> +#include "nouveau_bo.h" + +void +base507c_update(struct nv50_wndw *wndw, u32 *interlock) +{ + u32 *push; + if ((push = evo_wait(&wndw->wndw, 2))) { + evo_mthd(push, 0x0080, 1); + evo_data(push, interlock[NV50_DISP_INTERLOCK_CORE]); + evo_kick(push, &wndw->wndw); + } +} + +void +base507c_image_clr(struct nv50_wndw *wndw) +{ + u32 *push; + if ((push = evo_wait(&wndw->wndw, 4))) { + evo_mthd(push, 0x0084, 1); + evo_data(push, 0x00000000); + evo_mthd(push, 0x00c0, 1); + evo_data(push, 0x00000000); + evo_kick(push, &wndw->wndw); + } +} + +static void +base507c_image_set(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw) +{ + u32 *push; + if ((push = evo_wait(&wndw->wndw, 10))) { + evo_mthd(push, 0x0084, 1); + evo_data(push, asyw->image.mode << 8 | + asyw->image.interval << 4); + evo_mthd(push, 0x00c0, 1); + evo_data(push, asyw->image.handle[0]); + evo_mthd(push, 0x0800, 5); + evo_data(push, asyw->image.offset[0] >> 8); + evo_data(push, 0x00000000); + evo_data(push, asyw->image.h << 16 | asyw->image.w); + evo_data(push, asyw->image.layout << 20 | + (asyw->image.pitch[0] >> 8) << 8 | + asyw->image.blocks[0] << 8 | + asyw->image.blockh); + evo_data(push, asyw->image.kind << 16 | + asyw->image.format << 8); + evo_kick(push, &wndw->wndw); + } +} + +void +base507c_xlut_clr(struct nv50_wndw *wndw) +{ + u32 *push; + if ((push = evo_wait(&wndw->wndw, 2))) { + evo_mthd(push, 0x00e0, 1); + evo_data(push, 0x00000000); + evo_kick(push, &wndw->wndw); + } +} + +void +base507c_xlut_set(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw) +{ + u32 *push; + if ((push = evo_wait(&wndw->wndw, 2))) { + evo_mthd(push, 0x00e0, 1); + evo_data(push, 0x40000000); + evo_kick(push, &wndw->wndw); + } +} + +int +base507c_ntfy_wait_begun(struct nouveau_bo *bo, u32 offset, + struct nvif_device *device) +{ + s64 time = nvif_msec(device, 2000ULL, + u32 data = nouveau_bo_rd32(bo, offset / 4); + if ((data & 0xc0000000) == 0x40000000) + break; + usleep_range(1, 2); + ); + return time < 0 ? time : 0; +} + +void +base507c_ntfy_clr(struct nv50_wndw *wndw) +{ + u32 *push; + if ((push = evo_wait(&wndw->wndw, 2))) { + evo_mthd(push, 0x00a4, 1); + evo_data(push, 0x00000000); + evo_kick(push, &wndw->wndw); + } +} + +void +base507c_ntfy_set(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw) +{ + u32 *push; + if ((push = evo_wait(&wndw->wndw, 3))) { + evo_mthd(push, 0x00a0, 2); + evo_data(push, asyw->ntfy.awaken << 30 | asyw->ntfy.offset); + evo_data(push, asyw->ntfy.handle); + evo_kick(push, &wndw->wndw); + } +} + +void +base507c_ntfy_reset(struct nouveau_bo *bo, u32 offset) +{ + nouveau_bo_wr32(bo, offset / 4, 0x00000000); +} + +void +base507c_sema_clr(struct nv50_wndw *wndw) +{ + u32 *push; + if ((push = evo_wait(&wndw->wndw, 2))) { + evo_mthd(push, 0x0094, 1); + evo_data(push, 0x00000000); + evo_kick(push, &wndw->wndw); + } +} + +void +base507c_sema_set(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw) +{ + u32 *push; + if ((push = evo_wait(&wndw->wndw, 5))) { + evo_mthd(push, 0x0088, 4); + evo_data(push, asyw->sema.offset); + evo_data(push, asyw->sema.acquire); + evo_data(push, asyw->sema.release); + evo_data(push, asyw->sema.handle); + evo_kick(push, &wndw->wndw); + } +} + +void +base507c_release(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw, + struct nv50_head_atom *asyh) +{ + asyh->base.cpp = 0; +} + +int +base507c_acquire(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw, + struct nv50_head_atom *asyh) +{ + const struct drm_framebuffer *fb = asyw->state.fb; + int ret; + + if (!fb->format->depth) + return -EINVAL; + + ret = drm_atomic_helper_check_plane_state(&asyw->state, &asyh->state, + DRM_PLANE_HELPER_NO_SCALING, + DRM_PLANE_HELPER_NO_SCALING, + false, true); + if (ret) + return ret; + + if (!wndw->func->ilut) { + if ((asyh->base.cpp != 1) ^ (fb->format->cpp[0] != 1)) + asyh->state.color_mgmt_changed = true; + } + + asyh->base.depth = fb->format->depth; + asyh->base.cpp = fb->format->cpp[0]; + asyh->base.x = asyw->state.src.x1 >> 16; + asyh->base.y = asyw->state.src.y1 >> 16; + asyh->base.w = asyw->state.fb->width; + asyh->base.h = asyw->state.fb->height; + return 0; +} + +const u32 +base507c_format[] = { + DRM_FORMAT_C8, + DRM_FORMAT_RGB565, + DRM_FORMAT_XRGB1555, + DRM_FORMAT_ARGB1555, + DRM_FORMAT_XRGB8888, + DRM_FORMAT_ARGB8888, + DRM_FORMAT_XBGR2101010, + DRM_FORMAT_ABGR2101010, + DRM_FORMAT_XBGR8888, + DRM_FORMAT_ABGR8888, + 0 +}; + +static const struct nv50_wndw_func +base507c = { + .acquire = base507c_acquire, + .release = base507c_release, + .sema_set = base507c_sema_set, + .sema_clr = base507c_sema_clr, + .ntfy_reset = base507c_ntfy_reset, + .ntfy_set = base507c_ntfy_set, + .ntfy_clr = base507c_ntfy_clr, + .ntfy_wait_begun = base507c_ntfy_wait_begun, + .olut_core = 1, + .xlut_set = base507c_xlut_set, + .xlut_clr = base507c_xlut_clr, + .image_set = base507c_image_set, + .image_clr = base507c_image_clr, + .update = base507c_update, +}; + +int +base507c_new_(const struct nv50_wndw_func *func, const u32 *format, + struct nouveau_drm *drm, int head, s32 oclass, u32 interlock_data, + struct nv50_wndw **pwndw) +{ + struct nv50_disp_base_channel_dma_v0 args = { + .head = head, + }; + struct nv50_disp *disp = nv50_disp(drm->dev); + struct nv50_wndw *wndw; + int ret; + + ret = nv50_wndw_new_(func, drm->dev, DRM_PLANE_TYPE_PRIMARY, + "base", head, format, BIT(head), + NV50_DISP_INTERLOCK_BASE, interlock_data, &wndw); + if (*pwndw = wndw, ret) + return ret; + + ret = nv50_dmac_create(&drm->client.device, &disp->disp->object, + &oclass, head, &args, sizeof(args), + disp->sync->bo.offset, &wndw->wndw); + if (ret) { + NV_ERROR(drm, "base%04x allocation failed: %d\n", oclass, ret); + return ret; + } + + ret = nvif_notify_init(&wndw->wndw.base.user, wndw->notify.func, + false, NV50_DISP_BASE_CHANNEL_DMA_V0_NTFY_UEVENT, + &(struct nvif_notify_uevent_req) {}, + sizeof(struct nvif_notify_uevent_req), + sizeof(struct nvif_notify_uevent_rep), + &wndw->notify); + if (ret) + return ret; + + wndw->ntfy = NV50_DISP_BASE_NTFY(wndw->id); + wndw->sema = NV50_DISP_BASE_SEM0(wndw->id); + wndw->data = 0x00000000; + return 0; +} + +int +base507c_new(struct nouveau_drm *drm, int head, s32 oclass, + struct nv50_wndw **pwndw) +{ + return base507c_new_(&base507c, base507c_format, drm, head, oclass, + 0x00000002 << (head * 8), pwndw); +} diff --git a/drivers/gpu/drm/nouveau/dispnv50/base827c.c b/drivers/gpu/drm/nouveau/dispnv50/base827c.c new file mode 100644 index 000000000000..73646819a0d6 --- /dev/null +++ b/drivers/gpu/drm/nouveau/dispnv50/base827c.c @@ -0,0 +1,71 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "base.h" + +static void +base827c_image_set(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw) +{ + u32 *push; + if ((push = evo_wait(&wndw->wndw, 10))) { + evo_mthd(push, 0x0084, 1); + evo_data(push, asyw->image.mode << 8 | + asyw->image.interval << 4); + evo_mthd(push, 0x00c0, 1); + evo_data(push, asyw->image.handle[0]); + evo_mthd(push, 0x0800, 5); + evo_data(push, asyw->image.offset[0] >> 8); + evo_data(push, 0x00000000); + evo_data(push, asyw->image.h << 16 | asyw->image.w); + evo_data(push, asyw->image.layout << 20 | + (asyw->image.pitch[0] >> 8) << 8 | + asyw->image.blocks[0] << 8 | + asyw->image.blockh); + evo_data(push, asyw->image.format << 8); + evo_kick(push, &wndw->wndw); + } +} + +static const struct nv50_wndw_func +base827c = { + .acquire = base507c_acquire, + .release = base507c_release, + .sema_set = base507c_sema_set, + .sema_clr = base507c_sema_clr, + .ntfy_reset = base507c_ntfy_reset, + .ntfy_set = base507c_ntfy_set, + .ntfy_clr = base507c_ntfy_clr, + .ntfy_wait_begun = base507c_ntfy_wait_begun, + .olut_core = 1, + .xlut_set = base507c_xlut_set, + .xlut_clr = base507c_xlut_clr, + .image_set = base827c_image_set, + .image_clr = base507c_image_clr, + .update = base507c_update, +}; + +int +base827c_new(struct nouveau_drm *drm, int head, s32 oclass, + struct nv50_wndw **pwndw) +{ + return base507c_new_(&base827c, base507c_format, drm, head, oclass, + 0x00000002 << (head * 8), pwndw); +} diff --git a/drivers/gpu/drm/nouveau/dispnv50/base907c.c b/drivers/gpu/drm/nouveau/dispnv50/base907c.c new file mode 100644 index 000000000000..a562fc94ce59 --- /dev/null +++ b/drivers/gpu/drm/nouveau/dispnv50/base907c.c @@ -0,0 +1,110 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "base.h" + +static void +base907c_image_set(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw) +{ + u32 *push; + if ((push = evo_wait(&wndw->wndw, 10))) { + evo_mthd(push, 0x0084, 1); + evo_data(push, asyw->image.mode << 8 | + asyw->image.interval << 4); + evo_mthd(push, 0x00c0, 1); + evo_data(push, asyw->image.handle[0]); + evo_mthd(push, 0x0400, 5); + evo_data(push, asyw->image.offset[0] >> 8); + evo_data(push, 0x00000000); + evo_data(push, asyw->image.h << 16 | asyw->image.w); + evo_data(push, asyw->image.layout << 24 | + (asyw->image.pitch[0] >> 8) << 8 | + asyw->image.blocks[0] << 8 | + asyw->image.blockh); + evo_data(push, asyw->image.format << 8); + evo_kick(push, &wndw->wndw); + } +} + +static void +base907c_xlut_clr(struct nv50_wndw *wndw) +{ + u32 *push; + if ((push = evo_wait(&wndw->wndw, 6))) { + evo_mthd(push, 0x00e0, 1); + evo_data(push, 0x00000000); + evo_mthd(push, 0x00e8, 1); + evo_data(push, 0x00000000); + evo_mthd(push, 0x00fc, 1); + evo_data(push, 0x00000000); + evo_kick(push, &wndw->wndw); + } +} + +static void +base907c_xlut_set(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw) +{ + u32 *push; + if ((push = evo_wait(&wndw->wndw, 6))) { + evo_mthd(push, 0x00e0, 3); + evo_data(push, asyw->xlut.i.enable << 30 | + asyw->xlut.i.mode << 24); + evo_data(push, asyw->xlut.i.offset >> 8); + evo_data(push, 0x40000000); + evo_mthd(push, 0x00fc, 1); + evo_data(push, asyw->xlut.handle); + evo_kick(push, &wndw->wndw); + } +} + +static void +base907c_ilut(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw) +{ + asyw->xlut.i.mode = 7; + asyw->xlut.i.enable = 2; +} + +const struct nv50_wndw_func +base907c = { + .acquire = base507c_acquire, + .release = base507c_release, + .sema_set = base507c_sema_set, + .sema_clr = base507c_sema_clr, + .ntfy_reset = base507c_ntfy_reset, + .ntfy_set = base507c_ntfy_set, + .ntfy_clr = base507c_ntfy_clr, + .ntfy_wait_begun = base507c_ntfy_wait_begun, + .ilut = base907c_ilut, + .olut_core = true, + .xlut_set = base907c_xlut_set, + .xlut_clr = base907c_xlut_clr, + .image_set = base907c_image_set, + .image_clr = base507c_image_clr, + .update = base507c_update, +}; + +int +base907c_new(struct nouveau_drm *drm, int head, s32 oclass, + struct nv50_wndw **pwndw) +{ + return base507c_new_(&base907c, base507c_format, drm, head, oclass, + 0x00000002 << (head * 4), pwndw); +} diff --git a/drivers/gpu/drm/nouveau/dispnv50/base917c.c b/drivers/gpu/drm/nouveau/dispnv50/base917c.c new file mode 100644 index 000000000000..54d705bb81a5 --- /dev/null +++ b/drivers/gpu/drm/nouveau/dispnv50/base917c.c @@ -0,0 +1,48 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "base.h" +#include "atom.h" + +const u32 +base917c_format[] = { + DRM_FORMAT_C8, + DRM_FORMAT_XRGB8888, + DRM_FORMAT_ARGB8888, + DRM_FORMAT_RGB565, + DRM_FORMAT_XRGB1555, + DRM_FORMAT_ARGB1555, + DRM_FORMAT_XBGR2101010, + DRM_FORMAT_ABGR2101010, + DRM_FORMAT_XBGR8888, + DRM_FORMAT_ABGR8888, + DRM_FORMAT_XRGB2101010, + DRM_FORMAT_ARGB2101010, + 0 +}; + +int +base917c_new(struct nouveau_drm *drm, int head, s32 oclass, + struct nv50_wndw **pwndw) +{ + return base507c_new_(&base907c, base917c_format, drm, head, oclass, + 0x00000002 << (head * 4), pwndw); +} diff --git a/drivers/gpu/drm/nouveau/dispnv50/core.c b/drivers/gpu/drm/nouveau/dispnv50/core.c new file mode 100644 index 000000000000..f3c49adb1bdb --- /dev/null +++ b/drivers/gpu/drm/nouveau/dispnv50/core.c @@ -0,0 +1,70 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "core.h" + +#include <nvif/class.h> + +void +nv50_core_del(struct nv50_core **pcore) +{ + struct nv50_core *core = *pcore; + if (core) { + nv50_dmac_destroy(&core->chan); + kfree(*pcore); + *pcore = NULL; + } +} + +int +nv50_core_new(struct nouveau_drm *drm, struct nv50_core **pcore) +{ + struct { + s32 oclass; + int version; + int (*new)(struct nouveau_drm *, s32, struct nv50_core **); + } cores[] = { + { GV100_DISP_CORE_CHANNEL_DMA, 0, corec37d_new }, + { GP102_DISP_CORE_CHANNEL_DMA, 0, core917d_new }, + { GP100_DISP_CORE_CHANNEL_DMA, 0, core917d_new }, + { GM200_DISP_CORE_CHANNEL_DMA, 0, core917d_new }, + { GM107_DISP_CORE_CHANNEL_DMA, 0, core917d_new }, + { GK110_DISP_CORE_CHANNEL_DMA, 0, core917d_new }, + { GK104_DISP_CORE_CHANNEL_DMA, 0, core917d_new }, + { GF110_DISP_CORE_CHANNEL_DMA, 0, core907d_new }, + { GT214_DISP_CORE_CHANNEL_DMA, 0, core827d_new }, + { GT206_DISP_CORE_CHANNEL_DMA, 0, core827d_new }, + { GT200_DISP_CORE_CHANNEL_DMA, 0, core827d_new }, + { G82_DISP_CORE_CHANNEL_DMA, 0, core827d_new }, + { NV50_DISP_CORE_CHANNEL_DMA, 0, core507d_new }, + {} + }; + struct nv50_disp *disp = nv50_disp(drm->dev); + int cid; + + cid = nvif_mclass(&disp->disp->object, cores); + if (cid < 0) { + NV_ERROR(drm, "No supported core channel class\n"); + return cid; + } + + return cores[cid].new(drm, cores[cid].oclass, pcore); +} diff --git a/drivers/gpu/drm/nouveau/dispnv50/core.h b/drivers/gpu/drm/nouveau/dispnv50/core.h new file mode 100644 index 000000000000..8470df9dd13d --- /dev/null +++ b/drivers/gpu/drm/nouveau/dispnv50/core.h @@ -0,0 +1,50 @@ +#ifndef __NV50_KMS_CORE_H__ +#define __NV50_KMS_CORE_H__ +#include "disp.h" +#include "atom.h" + +struct nv50_core { + const struct nv50_core_func *func; + struct nv50_dmac chan; +}; + +int nv50_core_new(struct nouveau_drm *, struct nv50_core **); +void nv50_core_del(struct nv50_core **); + +struct nv50_core_func { + void (*init)(struct nv50_core *); + void (*ntfy_init)(struct nouveau_bo *, u32 offset); + int (*ntfy_wait_done)(struct nouveau_bo *, u32 offset, + struct nvif_device *); + void (*update)(struct nv50_core *, u32 *interlock, bool ntfy); + + const struct nv50_head_func *head; + const struct nv50_outp_func { + void (*ctrl)(struct nv50_core *, int or, u32 ctrl, + struct nv50_head_atom *); + } *dac, *pior, *sor; +}; + +int core507d_new(struct nouveau_drm *, s32, struct nv50_core **); +int core507d_new_(const struct nv50_core_func *, struct nouveau_drm *, s32, + struct nv50_core **); +void core507d_init(struct nv50_core *); +void core507d_ntfy_init(struct nouveau_bo *, u32); +int core507d_ntfy_wait_done(struct nouveau_bo *, u32, struct nvif_device *); +void core507d_update(struct nv50_core *, u32 *, bool); + +extern const struct nv50_outp_func dac507d; +extern const struct nv50_outp_func sor507d; +extern const struct nv50_outp_func pior507d; + +int core827d_new(struct nouveau_drm *, s32, struct nv50_core **); + +int core907d_new(struct nouveau_drm *, s32, struct nv50_core **); +extern const struct nv50_outp_func dac907d; +extern const struct nv50_outp_func sor907d; + +int core917d_new(struct nouveau_drm *, s32, struct nv50_core **); + +int corec37d_new(struct nouveau_drm *, s32, struct nv50_core **); +extern const struct nv50_outp_func sorc37d; +#endif diff --git a/drivers/gpu/drm/nouveau/dispnv50/core507d.c b/drivers/gpu/drm/nouveau/dispnv50/core507d.c new file mode 100644 index 000000000000..e7fcfa6e6467 --- /dev/null +++ b/drivers/gpu/drm/nouveau/dispnv50/core507d.c @@ -0,0 +1,115 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "core.h" +#include "head.h" + +#include <nvif/cl507d.h> + +#include "nouveau_bo.h" + +void +core507d_update(struct nv50_core *core, u32 *interlock, bool ntfy) +{ + u32 *push; + if ((push = evo_wait(&core->chan, 5))) { + if (ntfy) { + evo_mthd(push, 0x0084, 1); + evo_data(push, 0x80000000 | NV50_DISP_CORE_NTFY); + } + evo_mthd(push, 0x0080, 2); + evo_data(push, interlock[NV50_DISP_INTERLOCK_BASE] | + interlock[NV50_DISP_INTERLOCK_OVLY]); + evo_data(push, 0x00000000); + evo_kick(push, &core->chan); + } +} + +int +core507d_ntfy_wait_done(struct nouveau_bo *bo, u32 offset, + struct nvif_device *device) +{ + s64 time = nvif_msec(device, 2000ULL, + if (nouveau_bo_rd32(bo, offset / 4)) + break; + usleep_range(1, 2); + ); + return time < 0 ? time : 0; +} + +void +core507d_ntfy_init(struct nouveau_bo *bo, u32 offset) +{ + nouveau_bo_wr32(bo, offset / 4, 0x00000000); +} + +void +core507d_init(struct nv50_core *core) +{ + u32 *push; + if ((push = evo_wait(&core->chan, 2))) { + evo_mthd(push, 0x0088, 1); + evo_data(push, core->chan.sync.handle); + evo_kick(push, &core->chan); + } +} + +static const struct nv50_core_func +core507d = { + .init = core507d_init, + .ntfy_init = core507d_ntfy_init, + .ntfy_wait_done = core507d_ntfy_wait_done, + .update = core507d_update, + .head = &head507d, + .dac = &dac507d, + .sor = &sor507d, + .pior = &pior507d, +}; + +int +core507d_new_(const struct nv50_core_func *func, struct nouveau_drm *drm, + s32 oclass, struct nv50_core **pcore) +{ + struct nv50_disp_core_channel_dma_v0 args = {}; + struct nv50_disp *disp = nv50_disp(drm->dev); + struct nv50_core *core; + int ret; + + if (!(core = *pcore = kzalloc(sizeof(*core), GFP_KERNEL))) + return -ENOMEM; + core->func = func; + + ret = nv50_dmac_create(&drm->client.device, &disp->disp->object, + &oclass, 0, &args, sizeof(args), + disp->sync->bo.offset, &core->chan); + if (ret) { + NV_ERROR(drm, "core%04x allocation failed: %d\n", oclass, ret); + return ret; + } + + return 0; +} + +int +core507d_new(struct nouveau_drm *drm, s32 oclass, struct nv50_core **pcore) +{ + return core507d_new_(&core507d, drm, oclass, pcore); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/basegt215.c b/drivers/gpu/drm/nouveau/dispnv50/core827d.c index 08e2b1fa3806..6123a068f836 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/basegt215.c +++ b/drivers/gpu/drm/nouveau/dispnv50/core827d.c @@ -1,5 +1,5 @@ /* - * Copyright 2015 Red Hat Inc. + * Copyright 2018 Red Hat Inc. * * Permission is hereby granted, free of charge, to any person obtaining a * copy of this software and associated documentation files (the "Software"), @@ -18,21 +18,24 @@ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR * OTHER DEALINGS IN THE SOFTWARE. - * - * Authors: Ben Skeggs <bskeggs@redhat.com> */ -#include "dmacnv50.h" -#include "rootnv50.h" - -#include <nvif/class.h> +#include "core.h" +#include "head.h" -const struct nv50_disp_dmac_oclass -gt215_disp_base_oclass = { - .base.oclass = GT214_DISP_BASE_CHANNEL_DMA, - .base.minver = 0, - .base.maxver = 0, - .ctor = nv50_disp_base_new, - .func = &nv50_disp_dmac_func, - .mthd = &g84_disp_base_chan_mthd, - .chid = 1, +static const struct nv50_core_func +core827d = { + .init = core507d_init, + .ntfy_init = core507d_ntfy_init, + .ntfy_wait_done = core507d_ntfy_wait_done, + .update = core507d_update, + .head = &head827d, + .dac = &dac507d, + .sor = &sor507d, + .pior = &pior507d, }; + +int +core827d_new(struct nouveau_drm *drm, s32 oclass, struct nv50_core **pcore) +{ + return core507d_new_(&core827d, drm, oclass, pcore); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/basegk110.c b/drivers/gpu/drm/nouveau/dispnv50/core907d.c index d8bdd246c8ed..ef822f813435 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/basegk110.c +++ b/drivers/gpu/drm/nouveau/dispnv50/core907d.c @@ -1,5 +1,5 @@ /* - * Copyright 2015 Red Hat Inc. + * Copyright 2018 Red Hat Inc. * * Permission is hereby granted, free of charge, to any person obtaining a * copy of this software and associated documentation files (the "Software"), @@ -18,21 +18,23 @@ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR * OTHER DEALINGS IN THE SOFTWARE. - * - * Authors: Ben Skeggs <bskeggs@redhat.com> */ -#include "dmacnv50.h" -#include "rootnv50.h" - -#include <nvif/class.h> +#include "core.h" +#include "head.h" -const struct nv50_disp_dmac_oclass -gk110_disp_base_oclass = { - .base.oclass = GK110_DISP_BASE_CHANNEL_DMA, - .base.minver = 0, - .base.maxver = 0, - .ctor = nv50_disp_base_new, - .func = &gf119_disp_dmac_func, - .mthd = &gf119_disp_base_chan_mthd, - .chid = 1, +static const struct nv50_core_func +core907d = { + .init = core507d_init, + .ntfy_init = core507d_ntfy_init, + .ntfy_wait_done = core507d_ntfy_wait_done, + .update = core507d_update, + .head = &head907d, + .dac = &dac907d, + .sor = &sor907d, }; + +int +core907d_new(struct nouveau_drm *drm, s32 oclass, struct nv50_core **pcore) +{ + return core507d_new_(&core907d, drm, oclass, pcore); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/basegt200.c b/drivers/gpu/drm/nouveau/dispnv50/core917d.c index 93451e46570c..392338df5bfd 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/basegt200.c +++ b/drivers/gpu/drm/nouveau/dispnv50/core917d.c @@ -1,5 +1,5 @@ /* - * Copyright 2015 Red Hat Inc. + * Copyright 2018 Red Hat Inc. * * Permission is hereby granted, free of charge, to any person obtaining a * copy of this software and associated documentation files (the "Software"), @@ -18,21 +18,23 @@ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR * OTHER DEALINGS IN THE SOFTWARE. - * - * Authors: Ben Skeggs <bskeggs@redhat.com> */ -#include "dmacnv50.h" -#include "rootnv50.h" - -#include <nvif/class.h> +#include "core.h" +#include "head.h" -const struct nv50_disp_dmac_oclass -gt200_disp_base_oclass = { - .base.oclass = GT200_DISP_BASE_CHANNEL_DMA, - .base.minver = 0, - .base.maxver = 0, - .ctor = nv50_disp_base_new, - .func = &nv50_disp_dmac_func, - .mthd = &g84_disp_base_chan_mthd, - .chid = 1, +static const struct nv50_core_func +core917d = { + .init = core507d_init, + .ntfy_init = core507d_ntfy_init, + .ntfy_wait_done = core507d_ntfy_wait_done, + .update = core507d_update, + .head = &head917d, + .dac = &dac907d, + .sor = &sor907d, }; + +int +core917d_new(struct nouveau_drm *drm, s32 oclass, struct nv50_core **pcore) +{ + return core507d_new_(&core917d, drm, oclass, pcore); +} diff --git a/drivers/gpu/drm/nouveau/dispnv50/corec37d.c b/drivers/gpu/drm/nouveau/dispnv50/corec37d.c new file mode 100644 index 000000000000..b5c17c948918 --- /dev/null +++ b/drivers/gpu/drm/nouveau/dispnv50/corec37d.c @@ -0,0 +1,110 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "core.h" +#include "head.h" + +#include <nouveau_bo.h> + +static void +corec37d_update(struct nv50_core *core, u32 *interlock, bool ntfy) +{ + u32 *push; + if ((push = evo_wait(&core->chan, 9))) { + if (ntfy) { + evo_mthd(push, 0x020c, 1); + evo_data(push, 0x00001000 | NV50_DISP_CORE_NTFY); + } + + evo_mthd(push, 0x0218, 2); + evo_data(push, interlock[NV50_DISP_INTERLOCK_CURS]); + evo_data(push, interlock[NV50_DISP_INTERLOCK_WNDW]); + evo_mthd(push, 0x0200, 1); + evo_data(push, 0x00000001); + + if (ntfy) { + evo_mthd(push, 0x020c, 1); + evo_data(push, 0x00000000); + } + evo_kick(push, &core->chan); + } +} + +int +corec37d_ntfy_wait_done(struct nouveau_bo *bo, u32 offset, + struct nvif_device *device) +{ + u32 data; + s64 time = nvif_msec(device, 2000ULL, + data = nouveau_bo_rd32(bo, offset / 4 + 0); + if ((data & 0xc0000000) == 0x80000000) + break; + usleep_range(1, 2); + ); + return time < 0 ? time : 0; +} + +void +corec37d_ntfy_init(struct nouveau_bo *bo, u32 offset) +{ + nouveau_bo_wr32(bo, offset / 4 + 0, 0x00000000); + nouveau_bo_wr32(bo, offset / 4 + 1, 0x00000000); + nouveau_bo_wr32(bo, offset / 4 + 2, 0x00000000); + nouveau_bo_wr32(bo, offset / 4 + 3, 0x00000000); +} + +void +corec37d_init(struct nv50_core *core) +{ + const u32 windows = 8; /*XXX*/ + u32 *push, i; + if ((push = evo_wait(&core->chan, 2 + 6 * windows + 2))) { + evo_mthd(push, 0x0208, 1); + evo_data(push, core->chan.sync.handle); + for (i = 0; i < windows; i++) { + evo_mthd(push, 0x1000 + (i * 0x080), 3); + evo_data(push, i >> 1); + evo_data(push, 0x00000017); + evo_data(push, 0x00000000); + evo_mthd(push, 0x1010 + (i * 0x080), 1); + evo_data(push, 0x00127fff); + } + evo_mthd(push, 0x0200, 1); + evo_data(push, 0x00000001); + evo_kick(push, &core->chan); + } +} + +static const struct nv50_core_func +corec37d = { + .init = corec37d_init, + .ntfy_init = corec37d_ntfy_init, + .ntfy_wait_done = corec37d_ntfy_wait_done, + .update = corec37d_update, + .head = &headc37d, + .sor = &sorc37d, +}; + +int +corec37d_new(struct nouveau_drm *drm, s32 oclass, struct nv50_core **pcore) +{ + return core507d_new_(&corec37d, drm, oclass, pcore); +} diff --git a/drivers/gpu/drm/nouveau/dispnv50/curs.c b/drivers/gpu/drm/nouveau/dispnv50/curs.c new file mode 100644 index 000000000000..f592087338c4 --- /dev/null +++ b/drivers/gpu/drm/nouveau/dispnv50/curs.c @@ -0,0 +1,52 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "curs.h" + +#include <nvif/class.h> + +int +nv50_curs_new(struct nouveau_drm *drm, int head, struct nv50_wndw **pwndw) +{ + struct { + s32 oclass; + int version; + int (*new)(struct nouveau_drm *, int, s32, struct nv50_wndw **); + } curses[] = { + { GV100_DISP_CURSOR, 0, cursc37a_new }, + { GK104_DISP_CURSOR, 0, curs907a_new }, + { GF110_DISP_CURSOR, 0, curs907a_new }, + { GT214_DISP_CURSOR, 0, curs507a_new }, + { G82_DISP_CURSOR, 0, curs507a_new }, + { NV50_DISP_CURSOR, 0, curs507a_new }, + {} + }; + struct nv50_disp *disp = nv50_disp(drm->dev); + int cid; + + cid = nvif_mclass(&disp->disp->object, curses); + if (cid < 0) { + NV_ERROR(drm, "No supported cursor immediate class\n"); + return cid; + } + + return curses[cid].new(drm, head, curses[cid].oclass, pwndw); +} diff --git a/drivers/gpu/drm/nouveau/dispnv50/curs.h b/drivers/gpu/drm/nouveau/dispnv50/curs.h new file mode 100644 index 000000000000..23aff5fd6747 --- /dev/null +++ b/drivers/gpu/drm/nouveau/dispnv50/curs.h @@ -0,0 +1,14 @@ +#ifndef __NV50_KMS_CURS_H__ +#define __NV50_KMS_CURS_H__ +#include "wndw.h" + +int curs507a_new(struct nouveau_drm *, int, s32, struct nv50_wndw **); +int curs507a_new_(const struct nv50_wimm_func *, struct nouveau_drm *, + int head, s32 oclass, u32 interlock_data, + struct nv50_wndw **); + +int curs907a_new(struct nouveau_drm *, int, s32, struct nv50_wndw **); +int cursc37a_new(struct nouveau_drm *, int, s32, struct nv50_wndw **); + +int nv50_curs_new(struct nouveau_drm *, int head, struct nv50_wndw **); +#endif diff --git a/drivers/gpu/drm/nouveau/dispnv50/curs507a.c b/drivers/gpu/drm/nouveau/dispnv50/curs507a.c new file mode 100644 index 000000000000..291c08117ab6 --- /dev/null +++ b/drivers/gpu/drm/nouveau/dispnv50/curs507a.c @@ -0,0 +1,145 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "curs.h" +#include "core.h" +#include "head.h" + +#include <nvif/cl507a.h> + +#include <drm/drm_atomic_helper.h> +#include <drm/drm_plane_helper.h> + +static void +curs507a_update(struct nv50_wndw *wndw, u32 *interlock) +{ + nvif_wr32(&wndw->wimm.base.user, 0x0080, 0x00000000); +} + +static void +curs507a_point(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw) +{ + nvif_wr32(&wndw->wimm.base.user, 0x0084, asyw->point.y << 16 | + asyw->point.x); +} + +const struct nv50_wimm_func +curs507a = { + .point = curs507a_point, + .update = curs507a_update, +}; + +static void +curs507a_prepare(struct nv50_wndw *wndw, struct nv50_head_atom *asyh, + struct nv50_wndw_atom *asyw) +{ + u32 handle = nv50_disp(wndw->plane.dev)->core->chan.vram.handle; + u32 offset = asyw->image.offset[0]; + if (asyh->curs.handle != handle || asyh->curs.offset != offset) { + asyh->curs.handle = handle; + asyh->curs.offset = offset; + asyh->set.curs = asyh->curs.visible; + } +} + +static void +curs507a_release(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw, + struct nv50_head_atom *asyh) +{ + asyh->curs.visible = false; +} + +static int +curs507a_acquire(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw, + struct nv50_head_atom *asyh) +{ + struct nv50_head *head = nv50_head(asyw->state.crtc); + int ret; + + ret = drm_atomic_helper_check_plane_state(&asyw->state, &asyh->state, + DRM_PLANE_HELPER_NO_SCALING, + DRM_PLANE_HELPER_NO_SCALING, + true, true); + asyh->curs.visible = asyw->state.visible; + if (ret || !asyh->curs.visible) + return ret; + + if (asyw->image.w != asyw->image.h) + return -EINVAL; + + ret = head->func->curs_layout(head, asyw, asyh); + if (ret) + return ret; + + return head->func->curs_format(head, asyw, asyh); +} + +static const u32 +curs507a_format[] = { + DRM_FORMAT_ARGB8888, + 0 +}; + +static const struct nv50_wndw_func +curs507a_wndw = { + .acquire = curs507a_acquire, + .release = curs507a_release, + .prepare = curs507a_prepare, +}; + +int +curs507a_new_(const struct nv50_wimm_func *func, struct nouveau_drm *drm, + int head, s32 oclass, u32 interlock_data, + struct nv50_wndw **pwndw) +{ + struct nv50_disp_cursor_v0 args = { + .head = head, + }; + struct nv50_disp *disp = nv50_disp(drm->dev); + struct nv50_wndw *wndw; + int ret; + + ret = nv50_wndw_new_(&curs507a_wndw, drm->dev, DRM_PLANE_TYPE_CURSOR, + "curs", head, curs507a_format, BIT(head), + NV50_DISP_INTERLOCK_CURS, interlock_data, &wndw); + if (*pwndw = wndw, ret) + return ret; + + ret = nvif_object_init(&disp->disp->object, 0, oclass, &args, + sizeof(args), &wndw->wimm.base.user); + if (ret) { + NV_ERROR(drm, "curs%04x allocation failed: %d\n", oclass, ret); + return ret; + } + + nvif_object_map(&wndw->wimm.base.user, NULL, 0); + wndw->immd = func; + wndw->ctxdma.parent = &disp->core->chan.base.user; + return 0; +} + +int +curs507a_new(struct nouveau_drm *drm, int head, s32 oclass, + struct nv50_wndw **pwndw) +{ + return curs507a_new_(&curs507a, drm, head, oclass, + 0x00000001 << (head * 8), pwndw); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gpfifogk110.c b/drivers/gpu/drm/nouveau/dispnv50/curs907a.c index a9aa69c82e8e..d742362de03e 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gpfifogk110.c +++ b/drivers/gpu/drm/nouveau/dispnv50/curs907a.c @@ -1,5 +1,5 @@ /* - * Copyright 2016 Red Hat Inc. + * Copyright 2018 Red Hat Inc. * * Permission is hereby granted, free of charge, to any person obtaining a * copy of this software and associated documentation files (the "Software"), @@ -18,17 +18,13 @@ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR * OTHER DEALINGS IN THE SOFTWARE. - * - * Authors: Ben Skeggs */ -#include "changk104.h" - -#include <nvif/class.h> +#include "curs.h" -const struct nvkm_fifo_chan_oclass -gk110_fifo_gpfifo_oclass = { - .base.oclass = KEPLER_CHANNEL_GPFIFO_B, - .base.minver = 0, - .base.maxver = 0, - .ctor = gk104_fifo_gpfifo_new, -}; +int +curs907a_new(struct nouveau_drm *drm, int head, s32 oclass, + struct nv50_wndw **pwndw) +{ + return curs507a_new_(&curs507a, drm, head, oclass, + 0x00000001 << (head * 4), pwndw); +} diff --git a/drivers/gpu/drm/nouveau/dispnv50/cursc37a.c b/drivers/gpu/drm/nouveau/dispnv50/cursc37a.c new file mode 100644 index 000000000000..23fb29d41efe --- /dev/null +++ b/drivers/gpu/drm/nouveau/dispnv50/cursc37a.c @@ -0,0 +1,50 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "curs.h" +#include "atom.h" + +static void +cursc37a_update(struct nv50_wndw *wndw, u32 *interlock) +{ + nvif_wr32(&wndw->wimm.base.user, 0x0200, 0x00000001); +} + +static void +cursc37a_point(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw) +{ + nvif_wr32(&wndw->wimm.base.user, 0x0208, asyw->point.y << 16 | + asyw->point.x); +} + +static const struct nv50_wimm_func +cursc37a = { + .point = cursc37a_point, + .update = cursc37a_update, +}; + +int +cursc37a_new(struct nouveau_drm *drm, int head, s32 oclass, + struct nv50_wndw **pwndw) +{ + return curs507a_new_(&cursc37a, drm, head, oclass, + 0x00000001 << head, pwndw); +} diff --git a/drivers/gpu/drm/nouveau/dispnv50/dac507d.c b/drivers/gpu/drm/nouveau/dispnv50/dac507d.c new file mode 100644 index 000000000000..2a10ef7d30a8 --- /dev/null +++ b/drivers/gpu/drm/nouveau/dispnv50/dac507d.c @@ -0,0 +1,44 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "core.h" + +static void +dac507d_ctrl(struct nv50_core *core, int or, u32 ctrl, + struct nv50_head_atom *asyh) +{ + u32 *push, sync = 0; + if ((push = evo_wait(&core->chan, 3))) { + if (asyh) { + sync |= asyh->or.nvsync << 1; + sync |= asyh->or.nhsync; + } + evo_mthd(push, 0x0400 + (or * 0x080), 2); + evo_data(push, ctrl); + evo_data(push, sync); + evo_kick(push, &core->chan); + } +} + +const struct nv50_outp_func +dac507d = { + .ctrl = dac507d_ctrl, +}; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/cursgt215.c b/drivers/gpu/drm/nouveau/dispnv50/dac907d.c index 00a7f3564450..11e87fa53fac 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/cursgt215.c +++ b/drivers/gpu/drm/nouveau/dispnv50/dac907d.c @@ -1,5 +1,5 @@ /* - * Copyright 2015 Red Hat Inc. + * Copyright 2018 Red Hat Inc. * * Permission is hereby granted, free of charge, to any person obtaining a * copy of this software and associated documentation files (the "Software"), @@ -18,20 +18,22 @@ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR * OTHER DEALINGS IN THE SOFTWARE. - * - * Authors: Ben Skeggs <bskeggs@redhat.com> */ -#include "channv50.h" -#include "rootnv50.h" +#include "core.h" -#include <nvif/class.h> +static void +dac907d_ctrl(struct nv50_core *core, int or, u32 ctrl, + struct nv50_head_atom *asyh) +{ + u32 *push; + if ((push = evo_wait(&core->chan, 2))) { + evo_mthd(push, 0x0180 + (or * 0x020), 1); + evo_data(push, ctrl); + evo_kick(push, &core->chan); + } +} -const struct nv50_disp_pioc_oclass -gt215_disp_curs_oclass = { - .base.oclass = GT214_DISP_CURSOR, - .base.minver = 0, - .base.maxver = 0, - .ctor = nv50_disp_curs_new, - .func = &nv50_disp_pioc_func, - .chid = { 7, 7 }, +const struct nv50_outp_func +dac907d = { + .ctrl = dac907d_ctrl, }; diff --git a/drivers/gpu/drm/nouveau/dispnv50/disp.c b/drivers/gpu/drm/nouveau/dispnv50/disp.c new file mode 100644 index 000000000000..b83465ae7c1b --- /dev/null +++ b/drivers/gpu/drm/nouveau/dispnv50/disp.c @@ -0,0 +1,2238 @@ +/* + * Copyright 2011 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: Ben Skeggs + */ +#include "disp.h" +#include "atom.h" +#include "core.h" +#include "head.h" +#include "wndw.h" + +#include <linux/dma-mapping.h> +#include <linux/hdmi.h> + +#include <drm/drmP.h> +#include <drm/drm_atomic_helper.h> +#include <drm/drm_crtc_helper.h> +#include <drm/drm_dp_helper.h> +#include <drm/drm_fb_helper.h> +#include <drm/drm_plane_helper.h> +#include <drm/drm_edid.h> + +#include <nvif/class.h> +#include <nvif/cl0002.h> +#include <nvif/cl5070.h> +#include <nvif/cl507d.h> +#include <nvif/event.h> + +#include "nouveau_drv.h" +#include "nouveau_dma.h" +#include "nouveau_gem.h" +#include "nouveau_connector.h" +#include "nouveau_encoder.h" +#include "nouveau_fence.h" +#include "nouveau_fbcon.h" + +#include <subdev/bios/dp.h> + +/****************************************************************************** + * Atomic state + *****************************************************************************/ + +struct nv50_outp_atom { + struct list_head head; + + struct drm_encoder *encoder; + bool flush_disable; + + union nv50_outp_atom_mask { + struct { + bool ctrl:1; + }; + u8 mask; + } set, clr; +}; + +/****************************************************************************** + * EVO channel + *****************************************************************************/ + +static int +nv50_chan_create(struct nvif_device *device, struct nvif_object *disp, + const s32 *oclass, u8 head, void *data, u32 size, + struct nv50_chan *chan) +{ + struct nvif_sclass *sclass; + int ret, i, n; + + chan->device = device; + + ret = n = nvif_object_sclass_get(disp, &sclass); + if (ret < 0) + return ret; + + while (oclass[0]) { + for (i = 0; i < n; i++) { + if (sclass[i].oclass == oclass[0]) { + ret = nvif_object_init(disp, 0, oclass[0], + data, size, &chan->user); + if (ret == 0) + nvif_object_map(&chan->user, NULL, 0); + nvif_object_sclass_put(&sclass); + return ret; + } + } + oclass++; + } + + nvif_object_sclass_put(&sclass); + return -ENOSYS; +} + +static void +nv50_chan_destroy(struct nv50_chan *chan) +{ + nvif_object_fini(&chan->user); +} + +/****************************************************************************** + * DMA EVO channel + *****************************************************************************/ + +void +nv50_dmac_destroy(struct nv50_dmac *dmac) +{ + nvif_object_fini(&dmac->vram); + nvif_object_fini(&dmac->sync); + + nv50_chan_destroy(&dmac->base); + + nvif_mem_fini(&dmac->push); +} + +int +nv50_dmac_create(struct nvif_device *device, struct nvif_object *disp, + const s32 *oclass, u8 head, void *data, u32 size, u64 syncbuf, + struct nv50_dmac *dmac) +{ + struct nouveau_cli *cli = (void *)device->object.client; + struct nv50_disp_core_channel_dma_v0 *args = data; + int ret; + + mutex_init(&dmac->lock); + + ret = nvif_mem_init_map(&cli->mmu, NVIF_MEM_COHERENT, 0x1000, + &dmac->push); + if (ret) + return ret; + + dmac->ptr = dmac->push.object.map.ptr; + + args->pushbuf = nvif_handle(&dmac->push.object); + + ret = nv50_chan_create(device, disp, oclass, head, data, size, + &dmac->base); + if (ret) + return ret; + + if (!syncbuf) + return 0; + + ret = nvif_object_init(&dmac->base.user, 0xf0000000, NV_DMA_IN_MEMORY, + &(struct nv_dma_v0) { + .target = NV_DMA_V0_TARGET_VRAM, + .access = NV_DMA_V0_ACCESS_RDWR, + .start = syncbuf + 0x0000, + .limit = syncbuf + 0x0fff, + }, sizeof(struct nv_dma_v0), + &dmac->sync); + if (ret) + return ret; + + ret = nvif_object_init(&dmac->base.user, 0xf0000001, NV_DMA_IN_MEMORY, + &(struct nv_dma_v0) { + .target = NV_DMA_V0_TARGET_VRAM, + .access = NV_DMA_V0_ACCESS_RDWR, + .start = 0, + .limit = device->info.ram_user - 1, + }, sizeof(struct nv_dma_v0), + &dmac->vram); + if (ret) + return ret; + + return ret; +} + +/****************************************************************************** + * EVO channel helpers + *****************************************************************************/ +u32 * +evo_wait(struct nv50_dmac *evoc, int nr) +{ + struct nv50_dmac *dmac = evoc; + struct nvif_device *device = dmac->base.device; + u32 put = nvif_rd32(&dmac->base.user, 0x0000) / 4; + + mutex_lock(&dmac->lock); + if (put + nr >= (PAGE_SIZE / 4) - 8) { + dmac->ptr[put] = 0x20000000; + + nvif_wr32(&dmac->base.user, 0x0000, 0x00000000); + if (nvif_msec(device, 2000, + if (!nvif_rd32(&dmac->base.user, 0x0004)) + break; + ) < 0) { + mutex_unlock(&dmac->lock); + pr_err("nouveau: evo channel stalled\n"); + return NULL; + } + + put = 0; + } + + return dmac->ptr + put; +} + +void +evo_kick(u32 *push, struct nv50_dmac *evoc) +{ + struct nv50_dmac *dmac = evoc; + nvif_wr32(&dmac->base.user, 0x0000, (push - dmac->ptr) << 2); + mutex_unlock(&dmac->lock); +} + +/****************************************************************************** + * Output path helpers + *****************************************************************************/ +static void +nv50_outp_release(struct nouveau_encoder *nv_encoder) +{ + struct nv50_disp *disp = nv50_disp(nv_encoder->base.base.dev); + struct { + struct nv50_disp_mthd_v1 base; + } args = { + .base.version = 1, + .base.method = NV50_DISP_MTHD_V1_RELEASE, + .base.hasht = nv_encoder->dcb->hasht, + .base.hashm = nv_encoder->dcb->hashm, + }; + + nvif_mthd(&disp->disp->object, 0, &args, sizeof(args)); + nv_encoder->or = -1; + nv_encoder->link = 0; +} + +static int +nv50_outp_acquire(struct nouveau_encoder *nv_encoder) +{ + struct nouveau_drm *drm = nouveau_drm(nv_encoder->base.base.dev); + struct nv50_disp *disp = nv50_disp(drm->dev); + struct { + struct nv50_disp_mthd_v1 base; + struct nv50_disp_acquire_v0 info; + } args = { + .base.version = 1, + .base.method = NV50_DISP_MTHD_V1_ACQUIRE, + .base.hasht = nv_encoder->dcb->hasht, + .base.hashm = nv_encoder->dcb->hashm, + }; + int ret; + + ret = nvif_mthd(&disp->disp->object, 0, &args, sizeof(args)); + if (ret) { + NV_ERROR(drm, "error acquiring output path: %d\n", ret); + return ret; + } + + nv_encoder->or = args.info.or; + nv_encoder->link = args.info.link; + return 0; +} + +static int +nv50_outp_atomic_check_view(struct drm_encoder *encoder, + struct drm_crtc_state *crtc_state, + struct drm_connector_state *conn_state, + struct drm_display_mode *native_mode) +{ + struct drm_display_mode *adjusted_mode = &crtc_state->adjusted_mode; + struct drm_display_mode *mode = &crtc_state->mode; + struct drm_connector *connector = conn_state->connector; + struct nouveau_conn_atom *asyc = nouveau_conn_atom(conn_state); + struct nouveau_drm *drm = nouveau_drm(encoder->dev); + + NV_ATOMIC(drm, "%s atomic_check\n", encoder->name); + asyc->scaler.full = false; + if (!native_mode) + return 0; + + if (asyc->scaler.mode == DRM_MODE_SCALE_NONE) { + switch (connector->connector_type) { + case DRM_MODE_CONNECTOR_LVDS: + case DRM_MODE_CONNECTOR_eDP: + /* Force use of scaler for non-EDID modes. */ + if (adjusted_mode->type & DRM_MODE_TYPE_DRIVER) + break; + mode = native_mode; + asyc->scaler.full = true; + break; + default: + break; + } + } else { + mode = native_mode; + } + + if (!drm_mode_equal(adjusted_mode, mode)) { + drm_mode_copy(adjusted_mode, mode); + crtc_state->mode_changed = true; + } + + return 0; +} + +static int +nv50_outp_atomic_check(struct drm_encoder *encoder, + struct drm_crtc_state *crtc_state, + struct drm_connector_state *conn_state) +{ + struct nouveau_connector *nv_connector = + nouveau_connector(conn_state->connector); + return nv50_outp_atomic_check_view(encoder, crtc_state, conn_state, + nv_connector->native_mode); +} + +/****************************************************************************** + * DAC + *****************************************************************************/ +static void +nv50_dac_disable(struct drm_encoder *encoder) +{ + struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder); + struct nv50_core *core = nv50_disp(encoder->dev)->core; + if (nv_encoder->crtc) + core->func->dac->ctrl(core, nv_encoder->or, 0x00000000, NULL); + nv_encoder->crtc = NULL; + nv50_outp_release(nv_encoder); +} + +static void +nv50_dac_enable(struct drm_encoder *encoder) +{ + struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder); + struct nouveau_crtc *nv_crtc = nouveau_crtc(encoder->crtc); + struct nv50_head_atom *asyh = nv50_head_atom(nv_crtc->base.state); + struct nv50_core *core = nv50_disp(encoder->dev)->core; + + nv50_outp_acquire(nv_encoder); + + core->func->dac->ctrl(core, nv_encoder->or, 1 << nv_crtc->index, asyh); + asyh->or.depth = 0; + + nv_encoder->crtc = encoder->crtc; +} + +static enum drm_connector_status +nv50_dac_detect(struct drm_encoder *encoder, struct drm_connector *connector) +{ + struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder); + struct nv50_disp *disp = nv50_disp(encoder->dev); + struct { + struct nv50_disp_mthd_v1 base; + struct nv50_disp_dac_load_v0 load; + } args = { + .base.version = 1, + .base.method = NV50_DISP_MTHD_V1_DAC_LOAD, + .base.hasht = nv_encoder->dcb->hasht, + .base.hashm = nv_encoder->dcb->hashm, + }; + int ret; + + args.load.data = nouveau_drm(encoder->dev)->vbios.dactestval; + if (args.load.data == 0) + args.load.data = 340; + + ret = nvif_mthd(&disp->disp->object, 0, &args, sizeof(args)); + if (ret || !args.load.load) + return connector_status_disconnected; + + return connector_status_connected; +} + +static const struct drm_encoder_helper_funcs +nv50_dac_help = { + .atomic_check = nv50_outp_atomic_check, + .enable = nv50_dac_enable, + .disable = nv50_dac_disable, + .detect = nv50_dac_detect +}; + +static void +nv50_dac_destroy(struct drm_encoder *encoder) +{ + drm_encoder_cleanup(encoder); + kfree(encoder); +} + +static const struct drm_encoder_funcs +nv50_dac_func = { + .destroy = nv50_dac_destroy, +}; + +static int +nv50_dac_create(struct drm_connector *connector, struct dcb_output *dcbe) +{ + struct nouveau_drm *drm = nouveau_drm(connector->dev); + struct nvkm_i2c *i2c = nvxx_i2c(&drm->client.device); + struct nvkm_i2c_bus *bus; + struct nouveau_encoder *nv_encoder; + struct drm_encoder *encoder; + int type = DRM_MODE_ENCODER_DAC; + + nv_encoder = kzalloc(sizeof(*nv_encoder), GFP_KERNEL); + if (!nv_encoder) + return -ENOMEM; + nv_encoder->dcb = dcbe; + + bus = nvkm_i2c_bus_find(i2c, dcbe->i2c_index); + if (bus) + nv_encoder->i2c = &bus->i2c; + + encoder = to_drm_encoder(nv_encoder); + encoder->possible_crtcs = dcbe->heads; + encoder->possible_clones = 0; + drm_encoder_init(connector->dev, encoder, &nv50_dac_func, type, + "dac-%04x-%04x", dcbe->hasht, dcbe->hashm); + drm_encoder_helper_add(encoder, &nv50_dac_help); + + drm_mode_connector_attach_encoder(connector, encoder); + return 0; +} + +/****************************************************************************** + * Audio + *****************************************************************************/ +static void +nv50_audio_disable(struct drm_encoder *encoder, struct nouveau_crtc *nv_crtc) +{ + struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder); + struct nv50_disp *disp = nv50_disp(encoder->dev); + struct { + struct nv50_disp_mthd_v1 base; + struct nv50_disp_sor_hda_eld_v0 eld; + } args = { + .base.version = 1, + .base.method = NV50_DISP_MTHD_V1_SOR_HDA_ELD, + .base.hasht = nv_encoder->dcb->hasht, + .base.hashm = (0xf0ff & nv_encoder->dcb->hashm) | + (0x0100 << nv_crtc->index), + }; + + nvif_mthd(&disp->disp->object, 0, &args, sizeof(args)); +} + +static void +nv50_audio_enable(struct drm_encoder *encoder, struct drm_display_mode *mode) +{ + struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder); + struct nouveau_crtc *nv_crtc = nouveau_crtc(encoder->crtc); + struct nouveau_connector *nv_connector; + struct nv50_disp *disp = nv50_disp(encoder->dev); + struct __packed { + struct { + struct nv50_disp_mthd_v1 mthd; + struct nv50_disp_sor_hda_eld_v0 eld; + } base; + u8 data[sizeof(nv_connector->base.eld)]; + } args = { + .base.mthd.version = 1, + .base.mthd.method = NV50_DISP_MTHD_V1_SOR_HDA_ELD, + .base.mthd.hasht = nv_encoder->dcb->hasht, + .base.mthd.hashm = (0xf0ff & nv_encoder->dcb->hashm) | + (0x0100 << nv_crtc->index), + }; + + nv_connector = nouveau_encoder_connector_get(nv_encoder); + if (!drm_detect_monitor_audio(nv_connector->edid)) + return; + + memcpy(args.data, nv_connector->base.eld, sizeof(args.data)); + + nvif_mthd(&disp->disp->object, 0, &args, + sizeof(args.base) + drm_eld_size(args.data)); +} + +/****************************************************************************** + * HDMI + *****************************************************************************/ +static void +nv50_hdmi_disable(struct drm_encoder *encoder, struct nouveau_crtc *nv_crtc) +{ + struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder); + struct nv50_disp *disp = nv50_disp(encoder->dev); + struct { + struct nv50_disp_mthd_v1 base; + struct nv50_disp_sor_hdmi_pwr_v0 pwr; + } args = { + .base.version = 1, + .base.method = NV50_DISP_MTHD_V1_SOR_HDMI_PWR, + .base.hasht = nv_encoder->dcb->hasht, + .base.hashm = (0xf0ff & nv_encoder->dcb->hashm) | + (0x0100 << nv_crtc->index), + }; + + nvif_mthd(&disp->disp->object, 0, &args, sizeof(args)); +} + +static void +nv50_hdmi_enable(struct drm_encoder *encoder, struct drm_display_mode *mode) +{ + struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder); + struct nouveau_crtc *nv_crtc = nouveau_crtc(encoder->crtc); + struct nv50_disp *disp = nv50_disp(encoder->dev); + struct { + struct nv50_disp_mthd_v1 base; + struct nv50_disp_sor_hdmi_pwr_v0 pwr; + u8 infoframes[2 * 17]; /* two frames, up to 17 bytes each */ + } args = { + .base.version = 1, + .base.method = NV50_DISP_MTHD_V1_SOR_HDMI_PWR, + .base.hasht = nv_encoder->dcb->hasht, + .base.hashm = (0xf0ff & nv_encoder->dcb->hashm) | + (0x0100 << nv_crtc->index), + .pwr.state = 1, + .pwr.rekey = 56, /* binary driver, and tegra, constant */ + }; + struct nouveau_connector *nv_connector; + u32 max_ac_packet; + union hdmi_infoframe avi_frame; + union hdmi_infoframe vendor_frame; + int ret; + int size; + + nv_connector = nouveau_encoder_connector_get(nv_encoder); + if (!drm_detect_hdmi_monitor(nv_connector->edid)) + return; + + ret = drm_hdmi_avi_infoframe_from_display_mode(&avi_frame.avi, mode, + false); + if (!ret) { + /* We have an AVI InfoFrame, populate it to the display */ + args.pwr.avi_infoframe_length + = hdmi_infoframe_pack(&avi_frame, args.infoframes, 17); + } + + ret = drm_hdmi_vendor_infoframe_from_display_mode(&vendor_frame.vendor.hdmi, + &nv_connector->base, mode); + if (!ret) { + /* We have a Vendor InfoFrame, populate it to the display */ + args.pwr.vendor_infoframe_length + = hdmi_infoframe_pack(&vendor_frame, + args.infoframes + + args.pwr.avi_infoframe_length, + 17); + } + + max_ac_packet = mode->htotal - mode->hdisplay; + max_ac_packet -= args.pwr.rekey; + max_ac_packet -= 18; /* constant from tegra */ + args.pwr.max_ac_packet = max_ac_packet / 32; + + size = sizeof(args.base) + + sizeof(args.pwr) + + args.pwr.avi_infoframe_length + + args.pwr.vendor_infoframe_length; + nvif_mthd(&disp->disp->object, 0, &args, size); + nv50_audio_enable(encoder, mode); +} + +/****************************************************************************** + * MST + *****************************************************************************/ +#define nv50_mstm(p) container_of((p), struct nv50_mstm, mgr) +#define nv50_mstc(p) container_of((p), struct nv50_mstc, connector) +#define nv50_msto(p) container_of((p), struct nv50_msto, encoder) + +struct nv50_mstm { + struct nouveau_encoder *outp; + + struct drm_dp_mst_topology_mgr mgr; + struct nv50_msto *msto[4]; + + bool modified; + bool disabled; + int links; +}; + +struct nv50_mstc { + struct nv50_mstm *mstm; + struct drm_dp_mst_port *port; + struct drm_connector connector; + + struct drm_display_mode *native; + struct edid *edid; + + int pbn; +}; + +struct nv50_msto { + struct drm_encoder encoder; + + struct nv50_head *head; + struct nv50_mstc *mstc; + bool disabled; +}; + +static struct drm_dp_payload * +nv50_msto_payload(struct nv50_msto *msto) +{ + struct nouveau_drm *drm = nouveau_drm(msto->encoder.dev); + struct nv50_mstc *mstc = msto->mstc; + struct nv50_mstm *mstm = mstc->mstm; + int vcpi = mstc->port->vcpi.vcpi, i; + + NV_ATOMIC(drm, "%s: vcpi %d\n", msto->encoder.name, vcpi); + for (i = 0; i < mstm->mgr.max_payloads; i++) { + struct drm_dp_payload *payload = &mstm->mgr.payloads[i]; + NV_ATOMIC(drm, "%s: %d: vcpi %d start 0x%02x slots 0x%02x\n", + mstm->outp->base.base.name, i, payload->vcpi, + payload->start_slot, payload->num_slots); + } + + for (i = 0; i < mstm->mgr.max_payloads; i++) { + struct drm_dp_payload *payload = &mstm->mgr.payloads[i]; + if (payload->vcpi == vcpi) + return payload; + } + + return NULL; +} + +static void +nv50_msto_cleanup(struct nv50_msto *msto) +{ + struct nouveau_drm *drm = nouveau_drm(msto->encoder.dev); + struct nv50_mstc *mstc = msto->mstc; + struct nv50_mstm *mstm = mstc->mstm; + + NV_ATOMIC(drm, "%s: msto cleanup\n", msto->encoder.name); + if (mstc->port && mstc->port->vcpi.vcpi > 0 && !nv50_msto_payload(msto)) + drm_dp_mst_deallocate_vcpi(&mstm->mgr, mstc->port); + if (msto->disabled) { + msto->mstc = NULL; + msto->head = NULL; + msto->disabled = false; + } +} + +static void +nv50_msto_prepare(struct nv50_msto *msto) +{ + struct nouveau_drm *drm = nouveau_drm(msto->encoder.dev); + struct nv50_mstc *mstc = msto->mstc; + struct nv50_mstm *mstm = mstc->mstm; + struct { + struct nv50_disp_mthd_v1 base; + struct nv50_disp_sor_dp_mst_vcpi_v0 vcpi; + } args = { + .base.version = 1, + .base.method = NV50_DISP_MTHD_V1_SOR_DP_MST_VCPI, + .base.hasht = mstm->outp->dcb->hasht, + .base.hashm = (0xf0ff & mstm->outp->dcb->hashm) | + (0x0100 << msto->head->base.index), + }; + + NV_ATOMIC(drm, "%s: msto prepare\n", msto->encoder.name); + if (mstc->port && mstc->port->vcpi.vcpi > 0) { + struct drm_dp_payload *payload = nv50_msto_payload(msto); + if (payload) { + args.vcpi.start_slot = payload->start_slot; + args.vcpi.num_slots = payload->num_slots; + args.vcpi.pbn = mstc->port->vcpi.pbn; + args.vcpi.aligned_pbn = mstc->port->vcpi.aligned_pbn; + } + } + + NV_ATOMIC(drm, "%s: %s: %02x %02x %04x %04x\n", + msto->encoder.name, msto->head->base.base.name, + args.vcpi.start_slot, args.vcpi.num_slots, + args.vcpi.pbn, args.vcpi.aligned_pbn); + nvif_mthd(&drm->display->disp.object, 0, &args, sizeof(args)); +} + +static int +nv50_msto_atomic_check(struct drm_encoder *encoder, + struct drm_crtc_state *crtc_state, + struct drm_connector_state *conn_state) +{ + struct nv50_mstc *mstc = nv50_mstc(conn_state->connector); + struct nv50_mstm *mstm = mstc->mstm; + int bpp = conn_state->connector->display_info.bpc * 3; + int slots; + + mstc->pbn = drm_dp_calc_pbn_mode(crtc_state->adjusted_mode.clock, bpp); + + slots = drm_dp_find_vcpi_slots(&mstm->mgr, mstc->pbn); + if (slots < 0) + return slots; + + return nv50_outp_atomic_check_view(encoder, crtc_state, conn_state, + mstc->native); +} + +static void +nv50_msto_enable(struct drm_encoder *encoder) +{ + struct nv50_head *head = nv50_head(encoder->crtc); + struct nv50_msto *msto = nv50_msto(encoder); + struct nv50_mstc *mstc = NULL; + struct nv50_mstm *mstm = NULL; + struct drm_connector *connector; + struct drm_connector_list_iter conn_iter; + u8 proto, depth; + int slots; + bool r; + + drm_connector_list_iter_begin(encoder->dev, &conn_iter); + drm_for_each_connector_iter(connector, &conn_iter) { + if (connector->state->best_encoder == &msto->encoder) { + mstc = nv50_mstc(connector); + mstm = mstc->mstm; + break; + } + } + drm_connector_list_iter_end(&conn_iter); + + if (WARN_ON(!mstc)) + return; + + slots = drm_dp_find_vcpi_slots(&mstm->mgr, mstc->pbn); + r = drm_dp_mst_allocate_vcpi(&mstm->mgr, mstc->port, mstc->pbn, slots); + WARN_ON(!r); + + if (!mstm->links++) + nv50_outp_acquire(mstm->outp); + + if (mstm->outp->link & 1) + proto = 0x8; + else + proto = 0x9; + + switch (mstc->connector.display_info.bpc) { + case 6: depth = 0x2; break; + case 8: depth = 0x5; break; + case 10: + default: depth = 0x6; break; + } + + mstm->outp->update(mstm->outp, head->base.index, + nv50_head_atom(head->base.base.state), proto, depth); + + msto->head = head; + msto->mstc = mstc; + mstm->modified = true; +} + +static void +nv50_msto_disable(struct drm_encoder *encoder) +{ + struct nv50_msto *msto = nv50_msto(encoder); + struct nv50_mstc *mstc = msto->mstc; + struct nv50_mstm *mstm = mstc->mstm; + + if (mstc->port) + drm_dp_mst_reset_vcpi_slots(&mstm->mgr, mstc->port); + + mstm->outp->update(mstm->outp, msto->head->base.index, NULL, 0, 0); + mstm->modified = true; + if (!--mstm->links) + mstm->disabled = true; + msto->disabled = true; +} + +static const struct drm_encoder_helper_funcs +nv50_msto_help = { + .disable = nv50_msto_disable, + .enable = nv50_msto_enable, + .atomic_check = nv50_msto_atomic_check, +}; + +static void +nv50_msto_destroy(struct drm_encoder *encoder) +{ + struct nv50_msto *msto = nv50_msto(encoder); + drm_encoder_cleanup(&msto->encoder); + kfree(msto); +} + +static const struct drm_encoder_funcs +nv50_msto = { + .destroy = nv50_msto_destroy, +}; + +static int +nv50_msto_new(struct drm_device *dev, u32 heads, const char *name, int id, + struct nv50_msto **pmsto) +{ + struct nv50_msto *msto; + int ret; + + if (!(msto = *pmsto = kzalloc(sizeof(*msto), GFP_KERNEL))) + return -ENOMEM; + + ret = drm_encoder_init(dev, &msto->encoder, &nv50_msto, + DRM_MODE_ENCODER_DPMST, "%s-mst-%d", name, id); + if (ret) { + kfree(*pmsto); + *pmsto = NULL; + return ret; + } + + drm_encoder_helper_add(&msto->encoder, &nv50_msto_help); + msto->encoder.possible_crtcs = heads; + return 0; +} + +static struct drm_encoder * +nv50_mstc_atomic_best_encoder(struct drm_connector *connector, + struct drm_connector_state *connector_state) +{ + struct nv50_head *head = nv50_head(connector_state->crtc); + struct nv50_mstc *mstc = nv50_mstc(connector); + if (mstc->port) { + struct nv50_mstm *mstm = mstc->mstm; + return &mstm->msto[head->base.index]->encoder; + } + return NULL; +} + +static struct drm_encoder * +nv50_mstc_best_encoder(struct drm_connector *connector) +{ + struct nv50_mstc *mstc = nv50_mstc(connector); + if (mstc->port) { + struct nv50_mstm *mstm = mstc->mstm; + return &mstm->msto[0]->encoder; + } + return NULL; +} + +static enum drm_mode_status +nv50_mstc_mode_valid(struct drm_connector *connector, + struct drm_display_mode *mode) +{ + return MODE_OK; +} + +static int +nv50_mstc_get_modes(struct drm_connector *connector) +{ + struct nv50_mstc *mstc = nv50_mstc(connector); + int ret = 0; + + mstc->edid = drm_dp_mst_get_edid(&mstc->connector, mstc->port->mgr, mstc->port); + drm_mode_connector_update_edid_property(&mstc->connector, mstc->edid); + if (mstc->edid) + ret = drm_add_edid_modes(&mstc->connector, mstc->edid); + + if (!mstc->connector.display_info.bpc) + mstc->connector.display_info.bpc = 8; + + if (mstc->native) + drm_mode_destroy(mstc->connector.dev, mstc->native); + mstc->native = nouveau_conn_native_mode(&mstc->connector); + return ret; +} + +static const struct drm_connector_helper_funcs +nv50_mstc_help = { + .get_modes = nv50_mstc_get_modes, + .mode_valid = nv50_mstc_mode_valid, + .best_encoder = nv50_mstc_best_encoder, + .atomic_best_encoder = nv50_mstc_atomic_best_encoder, +}; + +static enum drm_connector_status +nv50_mstc_detect(struct drm_connector *connector, bool force) +{ + struct nv50_mstc *mstc = nv50_mstc(connector); + if (!mstc->port) + return connector_status_disconnected; + return drm_dp_mst_detect_port(connector, mstc->port->mgr, mstc->port); +} + +static void +nv50_mstc_destroy(struct drm_connector *connector) +{ + struct nv50_mstc *mstc = nv50_mstc(connector); + drm_connector_cleanup(&mstc->connector); + kfree(mstc); +} + +static const struct drm_connector_funcs +nv50_mstc = { + .reset = nouveau_conn_reset, + .detect = nv50_mstc_detect, + .fill_modes = drm_helper_probe_single_connector_modes, + .destroy = nv50_mstc_destroy, + .atomic_duplicate_state = nouveau_conn_atomic_duplicate_state, + .atomic_destroy_state = nouveau_conn_atomic_destroy_state, + .atomic_set_property = nouveau_conn_atomic_set_property, + .atomic_get_property = nouveau_conn_atomic_get_property, +}; + +static int +nv50_mstc_new(struct nv50_mstm *mstm, struct drm_dp_mst_port *port, + const char *path, struct nv50_mstc **pmstc) +{ + struct drm_device *dev = mstm->outp->base.base.dev; + struct nv50_mstc *mstc; + int ret, i; + + if (!(mstc = *pmstc = kzalloc(sizeof(*mstc), GFP_KERNEL))) + return -ENOMEM; + mstc->mstm = mstm; + mstc->port = port; + + ret = drm_connector_init(dev, &mstc->connector, &nv50_mstc, + DRM_MODE_CONNECTOR_DisplayPort); + if (ret) { + kfree(*pmstc); + *pmstc = NULL; + return ret; + } + + drm_connector_helper_add(&mstc->connector, &nv50_mstc_help); + + mstc->connector.funcs->reset(&mstc->connector); + nouveau_conn_attach_properties(&mstc->connector); + + for (i = 0; i < ARRAY_SIZE(mstm->msto) && mstm->msto[i]; i++) + drm_mode_connector_attach_encoder(&mstc->connector, &mstm->msto[i]->encoder); + + drm_object_attach_property(&mstc->connector.base, dev->mode_config.path_property, 0); + drm_object_attach_property(&mstc->connector.base, dev->mode_config.tile_property, 0); + drm_mode_connector_set_path_property(&mstc->connector, path); + return 0; +} + +static void +nv50_mstm_cleanup(struct nv50_mstm *mstm) +{ + struct nouveau_drm *drm = nouveau_drm(mstm->outp->base.base.dev); + struct drm_encoder *encoder; + int ret; + + NV_ATOMIC(drm, "%s: mstm cleanup\n", mstm->outp->base.base.name); + ret = drm_dp_check_act_status(&mstm->mgr); + + ret = drm_dp_update_payload_part2(&mstm->mgr); + + drm_for_each_encoder(encoder, mstm->outp->base.base.dev) { + if (encoder->encoder_type == DRM_MODE_ENCODER_DPMST) { + struct nv50_msto *msto = nv50_msto(encoder); + struct nv50_mstc *mstc = msto->mstc; + if (mstc && mstc->mstm == mstm) + nv50_msto_cleanup(msto); + } + } + + mstm->modified = false; +} + +static void +nv50_mstm_prepare(struct nv50_mstm *mstm) +{ + struct nouveau_drm *drm = nouveau_drm(mstm->outp->base.base.dev); + struct drm_encoder *encoder; + int ret; + + NV_ATOMIC(drm, "%s: mstm prepare\n", mstm->outp->base.base.name); + ret = drm_dp_update_payload_part1(&mstm->mgr); + + drm_for_each_encoder(encoder, mstm->outp->base.base.dev) { + if (encoder->encoder_type == DRM_MODE_ENCODER_DPMST) { + struct nv50_msto *msto = nv50_msto(encoder); + struct nv50_mstc *mstc = msto->mstc; + if (mstc && mstc->mstm == mstm) + nv50_msto_prepare(msto); + } + } + + if (mstm->disabled) { + if (!mstm->links) + nv50_outp_release(mstm->outp); + mstm->disabled = false; + } +} + +static void +nv50_mstm_hotplug(struct drm_dp_mst_topology_mgr *mgr) +{ + struct nv50_mstm *mstm = nv50_mstm(mgr); + drm_kms_helper_hotplug_event(mstm->outp->base.base.dev); +} + +static void +nv50_mstm_destroy_connector(struct drm_dp_mst_topology_mgr *mgr, + struct drm_connector *connector) +{ + struct nouveau_drm *drm = nouveau_drm(connector->dev); + struct nv50_mstc *mstc = nv50_mstc(connector); + + drm_connector_unregister(&mstc->connector); + + drm_fb_helper_remove_one_connector(&drm->fbcon->helper, &mstc->connector); + + drm_modeset_lock(&drm->dev->mode_config.connection_mutex, NULL); + mstc->port = NULL; + drm_modeset_unlock(&drm->dev->mode_config.connection_mutex); + + drm_connector_unreference(&mstc->connector); +} + +static void +nv50_mstm_register_connector(struct drm_connector *connector) +{ + struct nouveau_drm *drm = nouveau_drm(connector->dev); + + drm_fb_helper_add_one_connector(&drm->fbcon->helper, connector); + + drm_connector_register(connector); +} + +static struct drm_connector * +nv50_mstm_add_connector(struct drm_dp_mst_topology_mgr *mgr, + struct drm_dp_mst_port *port, const char *path) +{ + struct nv50_mstm *mstm = nv50_mstm(mgr); + struct nv50_mstc *mstc; + int ret; + + ret = nv50_mstc_new(mstm, port, path, &mstc); + if (ret) { + if (mstc) + mstc->connector.funcs->destroy(&mstc->connector); + return NULL; + } + + return &mstc->connector; +} + +static const struct drm_dp_mst_topology_cbs +nv50_mstm = { + .add_connector = nv50_mstm_add_connector, + .register_connector = nv50_mstm_register_connector, + .destroy_connector = nv50_mstm_destroy_connector, + .hotplug = nv50_mstm_hotplug, +}; + +void +nv50_mstm_service(struct nv50_mstm *mstm) +{ + struct drm_dp_aux *aux = mstm ? mstm->mgr.aux : NULL; + bool handled = true; + int ret; + u8 esi[8] = {}; + + if (!aux) + return; + + while (handled) { + ret = drm_dp_dpcd_read(aux, DP_SINK_COUNT_ESI, esi, 8); + if (ret != 8) { + drm_dp_mst_topology_mgr_set_mst(&mstm->mgr, false); + return; + } + + drm_dp_mst_hpd_irq(&mstm->mgr, esi, &handled); + if (!handled) + break; + + drm_dp_dpcd_write(aux, DP_SINK_COUNT_ESI + 1, &esi[1], 3); + } +} + +void +nv50_mstm_remove(struct nv50_mstm *mstm) +{ + if (mstm) + drm_dp_mst_topology_mgr_set_mst(&mstm->mgr, false); +} + +static int +nv50_mstm_enable(struct nv50_mstm *mstm, u8 dpcd, int state) +{ + struct nouveau_encoder *outp = mstm->outp; + struct { + struct nv50_disp_mthd_v1 base; + struct nv50_disp_sor_dp_mst_link_v0 mst; + } args = { + .base.version = 1, + .base.method = NV50_DISP_MTHD_V1_SOR_DP_MST_LINK, + .base.hasht = outp->dcb->hasht, + .base.hashm = outp->dcb->hashm, + .mst.state = state, + }; + struct nouveau_drm *drm = nouveau_drm(outp->base.base.dev); + struct nvif_object *disp = &drm->display->disp.object; + int ret; + + if (dpcd >= 0x12) { + ret = drm_dp_dpcd_readb(mstm->mgr.aux, DP_MSTM_CTRL, &dpcd); + if (ret < 0) + return ret; + + dpcd &= ~DP_MST_EN; + if (state) + dpcd |= DP_MST_EN; + + ret = drm_dp_dpcd_writeb(mstm->mgr.aux, DP_MSTM_CTRL, dpcd); + if (ret < 0) + return ret; + } + + return nvif_mthd(disp, 0, &args, sizeof(args)); +} + +int +nv50_mstm_detect(struct nv50_mstm *mstm, u8 dpcd[8], int allow) +{ + int ret, state = 0; + + if (!mstm) + return 0; + + if (dpcd[0] >= 0x12) { + ret = drm_dp_dpcd_readb(mstm->mgr.aux, DP_MSTM_CAP, &dpcd[1]); + if (ret < 0) + return ret; + + if (!(dpcd[1] & DP_MST_CAP)) + dpcd[0] = 0x11; + else + state = allow; + } + + ret = nv50_mstm_enable(mstm, dpcd[0], state); + if (ret) + return ret; + + ret = drm_dp_mst_topology_mgr_set_mst(&mstm->mgr, state); + if (ret) + return nv50_mstm_enable(mstm, dpcd[0], 0); + + return mstm->mgr.mst_state; +} + +static void +nv50_mstm_fini(struct nv50_mstm *mstm) +{ + if (mstm && mstm->mgr.mst_state) + drm_dp_mst_topology_mgr_suspend(&mstm->mgr); +} + +static void +nv50_mstm_init(struct nv50_mstm *mstm) +{ + if (mstm && mstm->mgr.mst_state) + drm_dp_mst_topology_mgr_resume(&mstm->mgr); +} + +static void +nv50_mstm_del(struct nv50_mstm **pmstm) +{ + struct nv50_mstm *mstm = *pmstm; + if (mstm) { + kfree(*pmstm); + *pmstm = NULL; + } +} + +static int +nv50_mstm_new(struct nouveau_encoder *outp, struct drm_dp_aux *aux, int aux_max, + int conn_base_id, struct nv50_mstm **pmstm) +{ + const int max_payloads = hweight8(outp->dcb->heads); + struct drm_device *dev = outp->base.base.dev; + struct nv50_mstm *mstm; + int ret, i; + u8 dpcd; + + /* This is a workaround for some monitors not functioning + * correctly in MST mode on initial module load. I think + * some bad interaction with the VBIOS may be responsible. + * + * A good ol' off and on again seems to work here ;) + */ + ret = drm_dp_dpcd_readb(aux, DP_DPCD_REV, &dpcd); + if (ret >= 0 && dpcd >= 0x12) + drm_dp_dpcd_writeb(aux, DP_MSTM_CTRL, 0); + + if (!(mstm = *pmstm = kzalloc(sizeof(*mstm), GFP_KERNEL))) + return -ENOMEM; + mstm->outp = outp; + mstm->mgr.cbs = &nv50_mstm; + + ret = drm_dp_mst_topology_mgr_init(&mstm->mgr, dev, aux, aux_max, + max_payloads, conn_base_id); + if (ret) + return ret; + + for (i = 0; i < max_payloads; i++) { + ret = nv50_msto_new(dev, outp->dcb->heads, outp->base.base.name, + i, &mstm->msto[i]); + if (ret) + return ret; + } + + return 0; +} + +/****************************************************************************** + * SOR + *****************************************************************************/ +static void +nv50_sor_update(struct nouveau_encoder *nv_encoder, u8 head, + struct nv50_head_atom *asyh, u8 proto, u8 depth) +{ + struct nv50_disp *disp = nv50_disp(nv_encoder->base.base.dev); + struct nv50_core *core = disp->core; + + if (!asyh) { + nv_encoder->ctrl &= ~BIT(head); + if (!(nv_encoder->ctrl & 0x0000000f)) + nv_encoder->ctrl = 0; + } else { + nv_encoder->ctrl |= proto << 8; + nv_encoder->ctrl |= BIT(head); + asyh->or.depth = depth; + } + + core->func->sor->ctrl(core, nv_encoder->or, nv_encoder->ctrl, asyh); +} + +static void +nv50_sor_disable(struct drm_encoder *encoder) +{ + struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder); + struct nouveau_crtc *nv_crtc = nouveau_crtc(nv_encoder->crtc); + + nv_encoder->crtc = NULL; + + if (nv_crtc) { + struct nvkm_i2c_aux *aux = nv_encoder->aux; + u8 pwr; + + if (aux) { + int ret = nvkm_rdaux(aux, DP_SET_POWER, &pwr, 1); + if (ret == 0) { + pwr &= ~DP_SET_POWER_MASK; + pwr |= DP_SET_POWER_D3; + nvkm_wraux(aux, DP_SET_POWER, &pwr, 1); + } + } + + nv_encoder->update(nv_encoder, nv_crtc->index, NULL, 0, 0); + nv50_audio_disable(encoder, nv_crtc); + nv50_hdmi_disable(&nv_encoder->base.base, nv_crtc); + nv50_outp_release(nv_encoder); + } +} + +static void +nv50_sor_enable(struct drm_encoder *encoder) +{ + struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder); + struct nouveau_crtc *nv_crtc = nouveau_crtc(encoder->crtc); + struct nv50_head_atom *asyh = nv50_head_atom(nv_crtc->base.state); + struct drm_display_mode *mode = &asyh->state.adjusted_mode; + struct { + struct nv50_disp_mthd_v1 base; + struct nv50_disp_sor_lvds_script_v0 lvds; + } lvds = { + .base.version = 1, + .base.method = NV50_DISP_MTHD_V1_SOR_LVDS_SCRIPT, + .base.hasht = nv_encoder->dcb->hasht, + .base.hashm = nv_encoder->dcb->hashm, + }; + struct nv50_disp *disp = nv50_disp(encoder->dev); + struct drm_device *dev = encoder->dev; + struct nouveau_drm *drm = nouveau_drm(dev); + struct nouveau_connector *nv_connector; + struct nvbios *bios = &drm->vbios; + u8 proto = 0xf; + u8 depth = 0x0; + + nv_connector = nouveau_encoder_connector_get(nv_encoder); + nv_encoder->crtc = encoder->crtc; + nv50_outp_acquire(nv_encoder); + + switch (nv_encoder->dcb->type) { + case DCB_OUTPUT_TMDS: + if (nv_encoder->link & 1) { + proto = 0x1; + /* Only enable dual-link if: + * - Need to (i.e. rate > 165MHz) + * - DCB says we can + * - Not an HDMI monitor, since there's no dual-link + * on HDMI. + */ + if (mode->clock >= 165000 && + nv_encoder->dcb->duallink_possible && + !drm_detect_hdmi_monitor(nv_connector->edid)) + proto |= 0x4; + } else { + proto = 0x2; + } + + nv50_hdmi_enable(&nv_encoder->base.base, mode); + break; + case DCB_OUTPUT_LVDS: + proto = 0x0; + + if (bios->fp_no_ddc) { + if (bios->fp.dual_link) + lvds.lvds.script |= 0x0100; + if (bios->fp.if_is_24bit) + lvds.lvds.script |= 0x0200; + } else { + if (nv_connector->type == DCB_CONNECTOR_LVDS_SPWG) { + if (((u8 *)nv_connector->edid)[121] == 2) + lvds.lvds.script |= 0x0100; + } else + if (mode->clock >= bios->fp.duallink_transition_clk) { + lvds.lvds.script |= 0x0100; + } + + if (lvds.lvds.script & 0x0100) { + if (bios->fp.strapless_is_24bit & 2) + lvds.lvds.script |= 0x0200; + } else { + if (bios->fp.strapless_is_24bit & 1) + lvds.lvds.script |= 0x0200; + } + + if (nv_connector->base.display_info.bpc == 8) + lvds.lvds.script |= 0x0200; + } + + nvif_mthd(&disp->disp->object, 0, &lvds, sizeof(lvds)); + break; + case DCB_OUTPUT_DP: + if (nv_connector->base.display_info.bpc == 6) + depth = 0x2; + else + if (nv_connector->base.display_info.bpc == 8) + depth = 0x5; + else + depth = 0x6; + + if (nv_encoder->link & 1) + proto = 0x8; + else + proto = 0x9; + + nv50_audio_enable(encoder, mode); + break; + default: + BUG(); + break; + } + + nv_encoder->update(nv_encoder, nv_crtc->index, asyh, proto, depth); +} + +static const struct drm_encoder_helper_funcs +nv50_sor_help = { + .atomic_check = nv50_outp_atomic_check, + .enable = nv50_sor_enable, + .disable = nv50_sor_disable, +}; + +static void +nv50_sor_destroy(struct drm_encoder *encoder) +{ + struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder); + nv50_mstm_del(&nv_encoder->dp.mstm); + drm_encoder_cleanup(encoder); + kfree(encoder); +} + +static const struct drm_encoder_funcs +nv50_sor_func = { + .destroy = nv50_sor_destroy, +}; + +static int +nv50_sor_create(struct drm_connector *connector, struct dcb_output *dcbe) +{ + struct nouveau_connector *nv_connector = nouveau_connector(connector); + struct nouveau_drm *drm = nouveau_drm(connector->dev); + struct nvkm_bios *bios = nvxx_bios(&drm->client.device); + struct nvkm_i2c *i2c = nvxx_i2c(&drm->client.device); + struct nouveau_encoder *nv_encoder; + struct drm_encoder *encoder; + u8 ver, hdr, cnt, len; + u32 data; + int type, ret; + + switch (dcbe->type) { + case DCB_OUTPUT_LVDS: type = DRM_MODE_ENCODER_LVDS; break; + case DCB_OUTPUT_TMDS: + case DCB_OUTPUT_DP: + default: + type = DRM_MODE_ENCODER_TMDS; + break; + } + + nv_encoder = kzalloc(sizeof(*nv_encoder), GFP_KERNEL); + if (!nv_encoder) + return -ENOMEM; + nv_encoder->dcb = dcbe; + nv_encoder->update = nv50_sor_update; + + encoder = to_drm_encoder(nv_encoder); + encoder->possible_crtcs = dcbe->heads; + encoder->possible_clones = 0; + drm_encoder_init(connector->dev, encoder, &nv50_sor_func, type, + "sor-%04x-%04x", dcbe->hasht, dcbe->hashm); + drm_encoder_helper_add(encoder, &nv50_sor_help); + + drm_mode_connector_attach_encoder(connector, encoder); + + if (dcbe->type == DCB_OUTPUT_DP) { + struct nv50_disp *disp = nv50_disp(encoder->dev); + struct nvkm_i2c_aux *aux = + nvkm_i2c_aux_find(i2c, dcbe->i2c_index); + if (aux) { + if (disp->disp->object.oclass < GF110_DISP) { + /* HW has no support for address-only + * transactions, so we're required to + * use custom I2C-over-AUX code. + */ + nv_encoder->i2c = &aux->i2c; + } else { + nv_encoder->i2c = &nv_connector->aux.ddc; + } + nv_encoder->aux = aux; + } + + if ((data = nvbios_dp_table(bios, &ver, &hdr, &cnt, &len)) && + ver >= 0x40 && (nvbios_rd08(bios, data + 0x08) & 0x04)) { + ret = nv50_mstm_new(nv_encoder, &nv_connector->aux, 16, + nv_connector->base.base.id, + &nv_encoder->dp.mstm); + if (ret) + return ret; + } + } else { + struct nvkm_i2c_bus *bus = + nvkm_i2c_bus_find(i2c, dcbe->i2c_index); + if (bus) + nv_encoder->i2c = &bus->i2c; + } + + return 0; +} + +/****************************************************************************** + * PIOR + *****************************************************************************/ +static int +nv50_pior_atomic_check(struct drm_encoder *encoder, + struct drm_crtc_state *crtc_state, + struct drm_connector_state *conn_state) +{ + int ret = nv50_outp_atomic_check(encoder, crtc_state, conn_state); + if (ret) + return ret; + crtc_state->adjusted_mode.clock *= 2; + return 0; +} + +static void +nv50_pior_disable(struct drm_encoder *encoder) +{ + struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder); + struct nv50_core *core = nv50_disp(encoder->dev)->core; + if (nv_encoder->crtc) + core->func->pior->ctrl(core, nv_encoder->or, 0x00000000, NULL); + nv_encoder->crtc = NULL; + nv50_outp_release(nv_encoder); +} + +static void +nv50_pior_enable(struct drm_encoder *encoder) +{ + struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder); + struct nouveau_crtc *nv_crtc = nouveau_crtc(encoder->crtc); + struct nouveau_connector *nv_connector; + struct nv50_head_atom *asyh = nv50_head_atom(nv_crtc->base.state); + struct nv50_core *core = nv50_disp(encoder->dev)->core; + u8 owner = 1 << nv_crtc->index; + u8 proto; + + nv50_outp_acquire(nv_encoder); + + nv_connector = nouveau_encoder_connector_get(nv_encoder); + switch (nv_connector->base.display_info.bpc) { + case 10: asyh->or.depth = 0x6; break; + case 8: asyh->or.depth = 0x5; break; + case 6: asyh->or.depth = 0x2; break; + default: asyh->or.depth = 0x0; break; + } + + switch (nv_encoder->dcb->type) { + case DCB_OUTPUT_TMDS: + case DCB_OUTPUT_DP: + proto = 0x0; + break; + default: + BUG(); + break; + } + + core->func->pior->ctrl(core, nv_encoder->or, (proto << 8) | owner, asyh); + nv_encoder->crtc = encoder->crtc; +} + +static const struct drm_encoder_helper_funcs +nv50_pior_help = { + .atomic_check = nv50_pior_atomic_check, + .enable = nv50_pior_enable, + .disable = nv50_pior_disable, +}; + +static void +nv50_pior_destroy(struct drm_encoder *encoder) +{ + drm_encoder_cleanup(encoder); + kfree(encoder); +} + +static const struct drm_encoder_funcs +nv50_pior_func = { + .destroy = nv50_pior_destroy, +}; + +static int +nv50_pior_create(struct drm_connector *connector, struct dcb_output *dcbe) +{ + struct nouveau_drm *drm = nouveau_drm(connector->dev); + struct nvkm_i2c *i2c = nvxx_i2c(&drm->client.device); + struct nvkm_i2c_bus *bus = NULL; + struct nvkm_i2c_aux *aux = NULL; + struct i2c_adapter *ddc; + struct nouveau_encoder *nv_encoder; + struct drm_encoder *encoder; + int type; + + switch (dcbe->type) { + case DCB_OUTPUT_TMDS: + bus = nvkm_i2c_bus_find(i2c, NVKM_I2C_BUS_EXT(dcbe->extdev)); + ddc = bus ? &bus->i2c : NULL; + type = DRM_MODE_ENCODER_TMDS; + break; + case DCB_OUTPUT_DP: + aux = nvkm_i2c_aux_find(i2c, NVKM_I2C_AUX_EXT(dcbe->extdev)); + ddc = aux ? &aux->i2c : NULL; + type = DRM_MODE_ENCODER_TMDS; + break; + default: + return -ENODEV; + } + + nv_encoder = kzalloc(sizeof(*nv_encoder), GFP_KERNEL); + if (!nv_encoder) + return -ENOMEM; + nv_encoder->dcb = dcbe; + nv_encoder->i2c = ddc; + nv_encoder->aux = aux; + + encoder = to_drm_encoder(nv_encoder); + encoder->possible_crtcs = dcbe->heads; + encoder->possible_clones = 0; + drm_encoder_init(connector->dev, encoder, &nv50_pior_func, type, + "pior-%04x-%04x", dcbe->hasht, dcbe->hashm); + drm_encoder_helper_add(encoder, &nv50_pior_help); + + drm_mode_connector_attach_encoder(connector, encoder); + return 0; +} + +/****************************************************************************** + * Atomic + *****************************************************************************/ + +static void +nv50_disp_atomic_commit_core(struct nouveau_drm *drm, u32 *interlock) +{ + struct nv50_disp *disp = nv50_disp(drm->dev); + struct nv50_core *core = disp->core; + struct nv50_mstm *mstm; + struct drm_encoder *encoder; + + NV_ATOMIC(drm, "commit core %08x\n", interlock[NV50_DISP_INTERLOCK_BASE]); + + drm_for_each_encoder(encoder, drm->dev) { + if (encoder->encoder_type != DRM_MODE_ENCODER_DPMST) { + mstm = nouveau_encoder(encoder)->dp.mstm; + if (mstm && mstm->modified) + nv50_mstm_prepare(mstm); + } + } + + core->func->ntfy_init(disp->sync, NV50_DISP_CORE_NTFY); + core->func->update(core, interlock, true); + if (core->func->ntfy_wait_done(disp->sync, NV50_DISP_CORE_NTFY, + disp->core->chan.base.device)) + NV_ERROR(drm, "core notifier timeout\n"); + + drm_for_each_encoder(encoder, drm->dev) { + if (encoder->encoder_type != DRM_MODE_ENCODER_DPMST) { + mstm = nouveau_encoder(encoder)->dp.mstm; + if (mstm && mstm->modified) + nv50_mstm_cleanup(mstm); + } + } +} + +static void +nv50_disp_atomic_commit_tail(struct drm_atomic_state *state) +{ + struct drm_device *dev = state->dev; + struct drm_crtc_state *new_crtc_state, *old_crtc_state; + struct drm_crtc *crtc; + struct drm_plane_state *new_plane_state; + struct drm_plane *plane; + struct nouveau_drm *drm = nouveau_drm(dev); + struct nv50_disp *disp = nv50_disp(dev); + struct nv50_atom *atom = nv50_atom(state); + struct nv50_outp_atom *outp, *outt; + u32 interlock[NV50_DISP_INTERLOCK__SIZE] = {}; + int i; + + NV_ATOMIC(drm, "commit %d %d\n", atom->lock_core, atom->flush_disable); + drm_atomic_helper_wait_for_fences(dev, state, false); + drm_atomic_helper_wait_for_dependencies(state); + drm_atomic_helper_update_legacy_modeset_state(dev, state); + + if (atom->lock_core) + mutex_lock(&disp->mutex); + + /* Disable head(s). */ + for_each_oldnew_crtc_in_state(state, crtc, old_crtc_state, new_crtc_state, i) { + struct nv50_head_atom *asyh = nv50_head_atom(new_crtc_state); + struct nv50_head *head = nv50_head(crtc); + + NV_ATOMIC(drm, "%s: clr %04x (set %04x)\n", crtc->name, + asyh->clr.mask, asyh->set.mask); + if (old_crtc_state->active && !new_crtc_state->active) + drm_crtc_vblank_off(crtc); + + if (asyh->clr.mask) { + nv50_head_flush_clr(head, asyh, atom->flush_disable); + interlock[NV50_DISP_INTERLOCK_CORE] |= 1; + } + } + + /* Disable plane(s). */ + for_each_new_plane_in_state(state, plane, new_plane_state, i) { + struct nv50_wndw_atom *asyw = nv50_wndw_atom(new_plane_state); + struct nv50_wndw *wndw = nv50_wndw(plane); + + NV_ATOMIC(drm, "%s: clr %02x (set %02x)\n", plane->name, + asyw->clr.mask, asyw->set.mask); + if (!asyw->clr.mask) + continue; + + nv50_wndw_flush_clr(wndw, interlock, atom->flush_disable, asyw); + } + + /* Disable output path(s). */ + list_for_each_entry(outp, &atom->outp, head) { + const struct drm_encoder_helper_funcs *help; + struct drm_encoder *encoder; + + encoder = outp->encoder; + help = encoder->helper_private; + + NV_ATOMIC(drm, "%s: clr %02x (set %02x)\n", encoder->name, + outp->clr.mask, outp->set.mask); + + if (outp->clr.mask) { + help->disable(encoder); + interlock[NV50_DISP_INTERLOCK_CORE] |= 1; + if (outp->flush_disable) { + nv50_disp_atomic_commit_core(drm, interlock); + memset(interlock, 0x00, sizeof(interlock)); + } + } + } + + /* Flush disable. */ + if (interlock[NV50_DISP_INTERLOCK_CORE]) { + if (atom->flush_disable) { + for_each_new_plane_in_state(state, plane, new_plane_state, i) { + struct nv50_wndw *wndw = nv50_wndw(plane); + if (interlock[wndw->interlock.type] & wndw->interlock.data) { + if (wndw->func->update) + wndw->func->update(wndw, interlock); + } + } + + nv50_disp_atomic_commit_core(drm, interlock); + memset(interlock, 0x00, sizeof(interlock)); + } + } + + /* Update output path(s). */ + list_for_each_entry_safe(outp, outt, &atom->outp, head) { + const struct drm_encoder_helper_funcs *help; + struct drm_encoder *encoder; + + encoder = outp->encoder; + help = encoder->helper_private; + + NV_ATOMIC(drm, "%s: set %02x (clr %02x)\n", encoder->name, + outp->set.mask, outp->clr.mask); + + if (outp->set.mask) { + help->enable(encoder); + interlock[NV50_DISP_INTERLOCK_CORE] = 1; + } + + list_del(&outp->head); + kfree(outp); + } + + /* Update head(s). */ + for_each_oldnew_crtc_in_state(state, crtc, old_crtc_state, new_crtc_state, i) { + struct nv50_head_atom *asyh = nv50_head_atom(new_crtc_state); + struct nv50_head *head = nv50_head(crtc); + + NV_ATOMIC(drm, "%s: set %04x (clr %04x)\n", crtc->name, + asyh->set.mask, asyh->clr.mask); + + if (asyh->set.mask) { + nv50_head_flush_set(head, asyh); + interlock[NV50_DISP_INTERLOCK_CORE] = 1; + } + + if (new_crtc_state->active) { + if (!old_crtc_state->active) + drm_crtc_vblank_on(crtc); + if (new_crtc_state->event) + drm_crtc_vblank_get(crtc); + } + } + + /* Update plane(s). */ + for_each_new_plane_in_state(state, plane, new_plane_state, i) { + struct nv50_wndw_atom *asyw = nv50_wndw_atom(new_plane_state); + struct nv50_wndw *wndw = nv50_wndw(plane); + + NV_ATOMIC(drm, "%s: set %02x (clr %02x)\n", plane->name, + asyw->set.mask, asyw->clr.mask); + if ( !asyw->set.mask && + (!asyw->clr.mask || atom->flush_disable)) + continue; + + nv50_wndw_flush_set(wndw, interlock, asyw); + } + + /* Flush update. */ + for_each_new_plane_in_state(state, plane, new_plane_state, i) { + struct nv50_wndw *wndw = nv50_wndw(plane); + if (interlock[wndw->interlock.type] & wndw->interlock.data) { + if (wndw->func->update) + wndw->func->update(wndw, interlock); + } + } + + if (interlock[NV50_DISP_INTERLOCK_CORE]) { + if (interlock[NV50_DISP_INTERLOCK_BASE] || + !atom->state.legacy_cursor_update) + nv50_disp_atomic_commit_core(drm, interlock); + else + disp->core->func->update(disp->core, interlock, false); + } + + if (atom->lock_core) + mutex_unlock(&disp->mutex); + + /* Wait for HW to signal completion. */ + for_each_new_plane_in_state(state, plane, new_plane_state, i) { + struct nv50_wndw_atom *asyw = nv50_wndw_atom(new_plane_state); + struct nv50_wndw *wndw = nv50_wndw(plane); + int ret = nv50_wndw_wait_armed(wndw, asyw); + if (ret) + NV_ERROR(drm, "%s: timeout\n", plane->name); + } + + for_each_new_crtc_in_state(state, crtc, new_crtc_state, i) { + if (new_crtc_state->event) { + unsigned long flags; + /* Get correct count/ts if racing with vblank irq */ + if (new_crtc_state->active) + drm_crtc_accurate_vblank_count(crtc); + spin_lock_irqsave(&crtc->dev->event_lock, flags); + drm_crtc_send_vblank_event(crtc, new_crtc_state->event); + spin_unlock_irqrestore(&crtc->dev->event_lock, flags); + + new_crtc_state->event = NULL; + if (new_crtc_state->active) + drm_crtc_vblank_put(crtc); + } + } + + drm_atomic_helper_commit_hw_done(state); + drm_atomic_helper_cleanup_planes(dev, state); + drm_atomic_helper_commit_cleanup_done(state); + drm_atomic_state_put(state); +} + +static void +nv50_disp_atomic_commit_work(struct work_struct *work) +{ + struct drm_atomic_state *state = + container_of(work, typeof(*state), commit_work); + nv50_disp_atomic_commit_tail(state); +} + +static int +nv50_disp_atomic_commit(struct drm_device *dev, + struct drm_atomic_state *state, bool nonblock) +{ + struct nouveau_drm *drm = nouveau_drm(dev); + struct drm_plane_state *new_plane_state; + struct drm_plane *plane; + struct drm_crtc *crtc; + bool active = false; + int ret, i; + + ret = pm_runtime_get_sync(dev->dev); + if (ret < 0 && ret != -EACCES) + return ret; + + ret = drm_atomic_helper_setup_commit(state, nonblock); + if (ret) + goto done; + + INIT_WORK(&state->commit_work, nv50_disp_atomic_commit_work); + + ret = drm_atomic_helper_prepare_planes(dev, state); + if (ret) + goto done; + + if (!nonblock) { + ret = drm_atomic_helper_wait_for_fences(dev, state, true); + if (ret) + goto err_cleanup; + } + + ret = drm_atomic_helper_swap_state(state, true); + if (ret) + goto err_cleanup; + + for_each_new_plane_in_state(state, plane, new_plane_state, i) { + struct nv50_wndw_atom *asyw = nv50_wndw_atom(new_plane_state); + struct nv50_wndw *wndw = nv50_wndw(plane); + + if (asyw->set.image) + nv50_wndw_ntfy_enable(wndw, asyw); + } + + drm_atomic_state_get(state); + + if (nonblock) + queue_work(system_unbound_wq, &state->commit_work); + else + nv50_disp_atomic_commit_tail(state); + + drm_for_each_crtc(crtc, dev) { + if (crtc->state->enable) { + if (!drm->have_disp_power_ref) { + drm->have_disp_power_ref = true; + return 0; + } + active = true; + break; + } + } + + if (!active && drm->have_disp_power_ref) { + pm_runtime_put_autosuspend(dev->dev); + drm->have_disp_power_ref = false; + } + +err_cleanup: + if (ret) + drm_atomic_helper_cleanup_planes(dev, state); +done: + pm_runtime_put_autosuspend(dev->dev); + return ret; +} + +static struct nv50_outp_atom * +nv50_disp_outp_atomic_add(struct nv50_atom *atom, struct drm_encoder *encoder) +{ + struct nv50_outp_atom *outp; + + list_for_each_entry(outp, &atom->outp, head) { + if (outp->encoder == encoder) + return outp; + } + + outp = kzalloc(sizeof(*outp), GFP_KERNEL); + if (!outp) + return ERR_PTR(-ENOMEM); + + list_add(&outp->head, &atom->outp); + outp->encoder = encoder; + return outp; +} + +static int +nv50_disp_outp_atomic_check_clr(struct nv50_atom *atom, + struct drm_connector_state *old_connector_state) +{ + struct drm_encoder *encoder = old_connector_state->best_encoder; + struct drm_crtc_state *old_crtc_state, *new_crtc_state; + struct drm_crtc *crtc; + struct nv50_outp_atom *outp; + + if (!(crtc = old_connector_state->crtc)) + return 0; + + old_crtc_state = drm_atomic_get_old_crtc_state(&atom->state, crtc); + new_crtc_state = drm_atomic_get_new_crtc_state(&atom->state, crtc); + if (old_crtc_state->active && drm_atomic_crtc_needs_modeset(new_crtc_state)) { + outp = nv50_disp_outp_atomic_add(atom, encoder); + if (IS_ERR(outp)) + return PTR_ERR(outp); + + if (outp->encoder->encoder_type == DRM_MODE_ENCODER_DPMST) { + outp->flush_disable = true; + atom->flush_disable = true; + } + outp->clr.ctrl = true; + atom->lock_core = true; + } + + return 0; +} + +static int +nv50_disp_outp_atomic_check_set(struct nv50_atom *atom, + struct drm_connector_state *connector_state) +{ + struct drm_encoder *encoder = connector_state->best_encoder; + struct drm_crtc_state *new_crtc_state; + struct drm_crtc *crtc; + struct nv50_outp_atom *outp; + + if (!(crtc = connector_state->crtc)) + return 0; + + new_crtc_state = drm_atomic_get_new_crtc_state(&atom->state, crtc); + if (new_crtc_state->active && drm_atomic_crtc_needs_modeset(new_crtc_state)) { + outp = nv50_disp_outp_atomic_add(atom, encoder); + if (IS_ERR(outp)) + return PTR_ERR(outp); + + outp->set.ctrl = true; + atom->lock_core = true; + } + + return 0; +} + +static int +nv50_disp_atomic_check(struct drm_device *dev, struct drm_atomic_state *state) +{ + struct nv50_atom *atom = nv50_atom(state); + struct drm_connector_state *old_connector_state, *new_connector_state; + struct drm_connector *connector; + struct drm_crtc_state *new_crtc_state; + struct drm_crtc *crtc; + int ret, i; + + /* We need to handle colour management on a per-plane basis. */ + for_each_new_crtc_in_state(state, crtc, new_crtc_state, i) { + if (new_crtc_state->color_mgmt_changed) { + ret = drm_atomic_add_affected_planes(state, crtc); + if (ret) + return ret; + } + } + + ret = drm_atomic_helper_check(dev, state); + if (ret) + return ret; + + for_each_oldnew_connector_in_state(state, connector, old_connector_state, new_connector_state, i) { + ret = nv50_disp_outp_atomic_check_clr(atom, old_connector_state); + if (ret) + return ret; + + ret = nv50_disp_outp_atomic_check_set(atom, new_connector_state); + if (ret) + return ret; + } + + return 0; +} + +static void +nv50_disp_atomic_state_clear(struct drm_atomic_state *state) +{ + struct nv50_atom *atom = nv50_atom(state); + struct nv50_outp_atom *outp, *outt; + + list_for_each_entry_safe(outp, outt, &atom->outp, head) { + list_del(&outp->head); + kfree(outp); + } + + drm_atomic_state_default_clear(state); +} + +static void +nv50_disp_atomic_state_free(struct drm_atomic_state *state) +{ + struct nv50_atom *atom = nv50_atom(state); + drm_atomic_state_default_release(&atom->state); + kfree(atom); +} + +static struct drm_atomic_state * +nv50_disp_atomic_state_alloc(struct drm_device *dev) +{ + struct nv50_atom *atom; + if (!(atom = kzalloc(sizeof(*atom), GFP_KERNEL)) || + drm_atomic_state_init(dev, &atom->state) < 0) { + kfree(atom); + return NULL; + } + INIT_LIST_HEAD(&atom->outp); + return &atom->state; +} + +static const struct drm_mode_config_funcs +nv50_disp_func = { + .fb_create = nouveau_user_framebuffer_create, + .output_poll_changed = drm_fb_helper_output_poll_changed, + .atomic_check = nv50_disp_atomic_check, + .atomic_commit = nv50_disp_atomic_commit, + .atomic_state_alloc = nv50_disp_atomic_state_alloc, + .atomic_state_clear = nv50_disp_atomic_state_clear, + .atomic_state_free = nv50_disp_atomic_state_free, +}; + +/****************************************************************************** + * Init + *****************************************************************************/ + +void +nv50_display_fini(struct drm_device *dev) +{ + struct nouveau_encoder *nv_encoder; + struct drm_encoder *encoder; + struct drm_plane *plane; + + drm_for_each_plane(plane, dev) { + struct nv50_wndw *wndw = nv50_wndw(plane); + if (plane->funcs != &nv50_wndw) + continue; + nv50_wndw_fini(wndw); + } + + list_for_each_entry(encoder, &dev->mode_config.encoder_list, head) { + if (encoder->encoder_type != DRM_MODE_ENCODER_DPMST) { + nv_encoder = nouveau_encoder(encoder); + nv50_mstm_fini(nv_encoder->dp.mstm); + } + } +} + +int +nv50_display_init(struct drm_device *dev) +{ + struct nv50_core *core = nv50_disp(dev)->core; + struct drm_encoder *encoder; + struct drm_plane *plane; + + core->func->init(core); + + list_for_each_entry(encoder, &dev->mode_config.encoder_list, head) { + if (encoder->encoder_type != DRM_MODE_ENCODER_DPMST) { + struct nouveau_encoder *nv_encoder = + nouveau_encoder(encoder); + nv50_mstm_init(nv_encoder->dp.mstm); + } + } + + drm_for_each_plane(plane, dev) { + struct nv50_wndw *wndw = nv50_wndw(plane); + if (plane->funcs != &nv50_wndw) + continue; + nv50_wndw_init(wndw); + } + + return 0; +} + +void +nv50_display_destroy(struct drm_device *dev) +{ + struct nv50_disp *disp = nv50_disp(dev); + + nv50_core_del(&disp->core); + + nouveau_bo_unmap(disp->sync); + if (disp->sync) + nouveau_bo_unpin(disp->sync); + nouveau_bo_ref(NULL, &disp->sync); + + nouveau_display(dev)->priv = NULL; + kfree(disp); +} + +MODULE_PARM_DESC(atomic, "Expose atomic ioctl (default: disabled)"); +static int nouveau_atomic = 0; +module_param_named(atomic, nouveau_atomic, int, 0400); + +int +nv50_display_create(struct drm_device *dev) +{ + struct nvif_device *device = &nouveau_drm(dev)->client.device; + struct nouveau_drm *drm = nouveau_drm(dev); + struct dcb_table *dcb = &drm->vbios.dcb; + struct drm_connector *connector, *tmp; + struct nv50_disp *disp; + struct dcb_output *dcbe; + int crtcs, ret, i; + + disp = kzalloc(sizeof(*disp), GFP_KERNEL); + if (!disp) + return -ENOMEM; + + mutex_init(&disp->mutex); + + nouveau_display(dev)->priv = disp; + nouveau_display(dev)->dtor = nv50_display_destroy; + nouveau_display(dev)->init = nv50_display_init; + nouveau_display(dev)->fini = nv50_display_fini; + disp->disp = &nouveau_display(dev)->disp; + dev->mode_config.funcs = &nv50_disp_func; + dev->driver->driver_features |= DRIVER_PREFER_XBGR_30BPP; + if (nouveau_atomic) + dev->driver->driver_features |= DRIVER_ATOMIC; + + /* small shared memory area we use for notifiers and semaphores */ + ret = nouveau_bo_new(&drm->client, 4096, 0x1000, TTM_PL_FLAG_VRAM, + 0, 0x0000, NULL, NULL, &disp->sync); + if (!ret) { + ret = nouveau_bo_pin(disp->sync, TTM_PL_FLAG_VRAM, true); + if (!ret) { + ret = nouveau_bo_map(disp->sync); + if (ret) + nouveau_bo_unpin(disp->sync); + } + if (ret) + nouveau_bo_ref(NULL, &disp->sync); + } + + if (ret) + goto out; + + /* allocate master evo channel */ + ret = nv50_core_new(drm, &disp->core); + if (ret) + goto out; + + /* create crtc objects to represent the hw heads */ + if (disp->disp->object.oclass >= GV100_DISP) + crtcs = nvif_rd32(&device->object, 0x610060) & 0xff; + else + if (disp->disp->object.oclass >= GF110_DISP) + crtcs = nvif_rd32(&device->object, 0x612004) & 0xf; + else + crtcs = 0x3; + + for (i = 0; i < fls(crtcs); i++) { + if (!(crtcs & (1 << i))) + continue; + ret = nv50_head_create(dev, i); + if (ret) + goto out; + } + + /* create encoder/connector objects based on VBIOS DCB table */ + for (i = 0, dcbe = &dcb->entry[0]; i < dcb->entries; i++, dcbe++) { + connector = nouveau_connector_create(dev, dcbe->connector); + if (IS_ERR(connector)) + continue; + + if (dcbe->location == DCB_LOC_ON_CHIP) { + switch (dcbe->type) { + case DCB_OUTPUT_TMDS: + case DCB_OUTPUT_LVDS: + case DCB_OUTPUT_DP: + ret = nv50_sor_create(connector, dcbe); + break; + case DCB_OUTPUT_ANALOG: + ret = nv50_dac_create(connector, dcbe); + break; + default: + ret = -ENODEV; + break; + } + } else { + ret = nv50_pior_create(connector, dcbe); + } + + if (ret) { + NV_WARN(drm, "failed to create encoder %d/%d/%d: %d\n", + dcbe->location, dcbe->type, + ffs(dcbe->or) - 1, ret); + ret = 0; + } + } + + /* cull any connectors we created that don't have an encoder */ + list_for_each_entry_safe(connector, tmp, &dev->mode_config.connector_list, head) { + if (connector->encoder_ids[0]) + continue; + + NV_WARN(drm, "%s has no encoders, removing\n", + connector->name); + connector->funcs->destroy(connector); + } + +out: + if (ret) + nv50_display_destroy(dev); + return ret; +} diff --git a/drivers/gpu/drm/nouveau/dispnv50/disp.h b/drivers/gpu/drm/nouveau/dispnv50/disp.h new file mode 100644 index 000000000000..e48c5eb35b49 --- /dev/null +++ b/drivers/gpu/drm/nouveau/dispnv50/disp.h @@ -0,0 +1,89 @@ +#ifndef __NV50_KMS_H__ +#define __NV50_KMS_H__ +#include <nvif/mem.h> + +#include "nouveau_display.h" + +struct nv50_disp { + struct nvif_disp *disp; + struct nv50_core *core; + +#define NV50_DISP_SYNC(c, o) ((c) * 0x040 + (o)) +#define NV50_DISP_CORE_NTFY NV50_DISP_SYNC(0 , 0x00) +#define NV50_DISP_WNDW_SEM0(c) NV50_DISP_SYNC(1 + (c), 0x00) +#define NV50_DISP_WNDW_SEM1(c) NV50_DISP_SYNC(1 + (c), 0x10) +#define NV50_DISP_WNDW_NTFY(c) NV50_DISP_SYNC(1 + (c), 0x20) +#define NV50_DISP_BASE_SEM0(c) NV50_DISP_WNDW_SEM0(0 + (c)) +#define NV50_DISP_BASE_SEM1(c) NV50_DISP_WNDW_SEM1(0 + (c)) +#define NV50_DISP_BASE_NTFY(c) NV50_DISP_WNDW_NTFY(0 + (c)) +#define NV50_DISP_OVLY_SEM0(c) NV50_DISP_WNDW_SEM0(4 + (c)) +#define NV50_DISP_OVLY_SEM1(c) NV50_DISP_WNDW_SEM1(4 + (c)) +#define NV50_DISP_OVLY_NTFY(c) NV50_DISP_WNDW_NTFY(4 + (c)) + struct nouveau_bo *sync; + + struct mutex mutex; +}; + +static inline struct nv50_disp * +nv50_disp(struct drm_device *dev) +{ + return nouveau_display(dev)->priv; +} + +struct nv50_disp_interlock { + enum nv50_disp_interlock_type { + NV50_DISP_INTERLOCK_CORE = 0, + NV50_DISP_INTERLOCK_CURS, + NV50_DISP_INTERLOCK_BASE, + NV50_DISP_INTERLOCK_OVLY, + NV50_DISP_INTERLOCK_WNDW, + NV50_DISP_INTERLOCK_WIMM, + NV50_DISP_INTERLOCK__SIZE + } type; + u32 data; +}; + +void corec37d_ntfy_init(struct nouveau_bo *, u32); + +struct nv50_chan { + struct nvif_object user; + struct nvif_device *device; +}; + +struct nv50_dmac { + struct nv50_chan base; + + struct nvif_mem push; + u32 *ptr; + + struct nvif_object sync; + struct nvif_object vram; + + /* Protects against concurrent pushbuf access to this channel, lock is + * grabbed by evo_wait (if the pushbuf reservation is successful) and + * dropped again by evo_kick. */ + struct mutex lock; +}; + +int nv50_dmac_create(struct nvif_device *device, struct nvif_object *disp, + const s32 *oclass, u8 head, void *data, u32 size, + u64 syncbuf, struct nv50_dmac *dmac); +void nv50_dmac_destroy(struct nv50_dmac *); + +u32 *evo_wait(struct nv50_dmac *, int nr); +void evo_kick(u32 *, struct nv50_dmac *); + +#define evo_mthd(p, m, s) do { \ + const u32 _m = (m), _s = (s); \ + if (drm_debug & DRM_UT_KMS) \ + pr_err("%04x %d %s\n", _m, _s, __func__); \ + *((p)++) = ((_s << 18) | _m); \ +} while(0) + +#define evo_data(p, d) do { \ + const u32 _d = (d); \ + if (drm_debug & DRM_UT_KMS) \ + pr_err("\t%08x\n", _d); \ + *((p)++) = _d; \ +} while(0) +#endif diff --git a/drivers/gpu/drm/nouveau/dispnv50/head.c b/drivers/gpu/drm/nouveau/dispnv50/head.c new file mode 100644 index 000000000000..4f57e5379796 --- /dev/null +++ b/drivers/gpu/drm/nouveau/dispnv50/head.c @@ -0,0 +1,511 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "head.h" +#include "base.h" +#include "core.h" +#include "curs.h" +#include "ovly.h" + +#include <nvif/class.h> + +#include <drm/drm_atomic_helper.h> +#include <drm/drm_crtc_helper.h> +#include "nouveau_connector.h" +void +nv50_head_flush_clr(struct nv50_head *head, + struct nv50_head_atom *asyh, bool flush) +{ + union nv50_head_atom_mask clr = { + .mask = asyh->clr.mask & ~(flush ? 0 : asyh->set.mask), + }; + if (clr.olut) head->func->olut_clr(head); + if (clr.core) head->func->core_clr(head); + if (clr.curs) head->func->curs_clr(head); +} + +void +nv50_head_flush_set(struct nv50_head *head, struct nv50_head_atom *asyh) +{ + if (asyh->set.view ) head->func->view (head, asyh); + if (asyh->set.mode ) head->func->mode (head, asyh); + if (asyh->set.core ) head->func->core_set(head, asyh); + if (asyh->set.olut ) { + asyh->olut.offset = nv50_lut_load(&head->olut, + asyh->olut.mode <= 1, + asyh->olut.buffer, + asyh->state.gamma_lut); + head->func->olut_set(head, asyh); + } + if (asyh->set.curs ) head->func->curs_set(head, asyh); + if (asyh->set.base ) head->func->base (head, asyh); + if (asyh->set.ovly ) head->func->ovly (head, asyh); + if (asyh->set.dither ) head->func->dither (head, asyh); + if (asyh->set.procamp) head->func->procamp (head, asyh); + if (asyh->set.or ) head->func->or (head, asyh); +} + +static void +nv50_head_atomic_check_procamp(struct nv50_head_atom *armh, + struct nv50_head_atom *asyh, + struct nouveau_conn_atom *asyc) +{ + const int vib = asyc->procamp.color_vibrance - 100; + const int hue = asyc->procamp.vibrant_hue - 90; + const int adj = (vib > 0) ? 50 : 0; + asyh->procamp.sat.cos = ((vib * 2047 + adj) / 100) & 0xfff; + asyh->procamp.sat.sin = ((hue * 2047) / 100) & 0xfff; + asyh->set.procamp = true; +} + +static void +nv50_head_atomic_check_dither(struct nv50_head_atom *armh, + struct nv50_head_atom *asyh, + struct nouveau_conn_atom *asyc) +{ + struct drm_connector *connector = asyc->state.connector; + u32 mode = 0x00; + + if (asyc->dither.mode == DITHERING_MODE_AUTO) { + if (asyh->base.depth > connector->display_info.bpc * 3) + mode = DITHERING_MODE_DYNAMIC2X2; + } else { + mode = asyc->dither.mode; + } + + if (asyc->dither.depth == DITHERING_DEPTH_AUTO) { + if (connector->display_info.bpc >= 8) + mode |= DITHERING_DEPTH_8BPC; + } else { + mode |= asyc->dither.depth; + } + + asyh->dither.enable = mode; + asyh->dither.bits = mode >> 1; + asyh->dither.mode = mode >> 3; + asyh->set.dither = true; +} + +static void +nv50_head_atomic_check_view(struct nv50_head_atom *armh, + struct nv50_head_atom *asyh, + struct nouveau_conn_atom *asyc) +{ + struct drm_connector *connector = asyc->state.connector; + struct drm_display_mode *omode = &asyh->state.adjusted_mode; + struct drm_display_mode *umode = &asyh->state.mode; + int mode = asyc->scaler.mode; + struct edid *edid; + int umode_vdisplay, omode_hdisplay, omode_vdisplay; + + if (connector->edid_blob_ptr) + edid = (struct edid *)connector->edid_blob_ptr->data; + else + edid = NULL; + + if (!asyc->scaler.full) { + if (mode == DRM_MODE_SCALE_NONE) + omode = umode; + } else { + /* Non-EDID LVDS/eDP mode. */ + mode = DRM_MODE_SCALE_FULLSCREEN; + } + + /* For the user-specified mode, we must ignore doublescan and + * the like, but honor frame packing. + */ + umode_vdisplay = umode->vdisplay; + if ((umode->flags & DRM_MODE_FLAG_3D_MASK) == DRM_MODE_FLAG_3D_FRAME_PACKING) + umode_vdisplay += umode->vtotal; + asyh->view.iW = umode->hdisplay; + asyh->view.iH = umode_vdisplay; + /* For the output mode, we can just use the stock helper. */ + drm_mode_get_hv_timing(omode, &omode_hdisplay, &omode_vdisplay); + asyh->view.oW = omode_hdisplay; + asyh->view.oH = omode_vdisplay; + + /* Add overscan compensation if necessary, will keep the aspect + * ratio the same as the backend mode unless overridden by the + * user setting both hborder and vborder properties. + */ + if ((asyc->scaler.underscan.mode == UNDERSCAN_ON || + (asyc->scaler.underscan.mode == UNDERSCAN_AUTO && + drm_detect_hdmi_monitor(edid)))) { + u32 bX = asyc->scaler.underscan.hborder; + u32 bY = asyc->scaler.underscan.vborder; + u32 r = (asyh->view.oH << 19) / asyh->view.oW; + + if (bX) { + asyh->view.oW -= (bX * 2); + if (bY) asyh->view.oH -= (bY * 2); + else asyh->view.oH = ((asyh->view.oW * r) + (r / 2)) >> 19; + } else { + asyh->view.oW -= (asyh->view.oW >> 4) + 32; + if (bY) asyh->view.oH -= (bY * 2); + else asyh->view.oH = ((asyh->view.oW * r) + (r / 2)) >> 19; + } + } + + /* Handle CENTER/ASPECT scaling, taking into account the areas + * removed already for overscan compensation. + */ + switch (mode) { + case DRM_MODE_SCALE_CENTER: + asyh->view.oW = min((u16)umode->hdisplay, asyh->view.oW); + asyh->view.oH = min((u16)umode_vdisplay, asyh->view.oH); + /* fall-through */ + case DRM_MODE_SCALE_ASPECT: + if (asyh->view.oH < asyh->view.oW) { + u32 r = (asyh->view.iW << 19) / asyh->view.iH; + asyh->view.oW = ((asyh->view.oH * r) + (r / 2)) >> 19; + } else { + u32 r = (asyh->view.iH << 19) / asyh->view.iW; + asyh->view.oH = ((asyh->view.oW * r) + (r / 2)) >> 19; + } + break; + default: + break; + } + + asyh->set.view = true; +} + +static int +nv50_head_atomic_check_lut(struct nv50_head *head, + struct nv50_head_atom *asyh) +{ + struct nv50_disp *disp = nv50_disp(head->base.base.dev); + struct drm_property_blob *olut = asyh->state.gamma_lut; + + /* Determine whether core output LUT should be enabled. */ + if (olut) { + /* Check if any window(s) have stolen the core output LUT + * to as an input LUT for legacy gamma + I8 colour format. + */ + if (asyh->wndw.olut) { + /* If any window has stolen the core output LUT, + * all of them must. + */ + if (asyh->wndw.olut != asyh->wndw.mask) + return -EINVAL; + olut = NULL; + } + } + + if (!olut) { + asyh->olut.handle = 0; + return 0; + } + + asyh->olut.handle = disp->core->chan.vram.handle; + asyh->olut.buffer = !asyh->olut.buffer; + head->func->olut(head, asyh); + return 0; +} + +static void +nv50_head_atomic_check_mode(struct nv50_head *head, struct nv50_head_atom *asyh) +{ + struct drm_display_mode *mode = &asyh->state.adjusted_mode; + struct nv50_head_mode *m = &asyh->mode; + u32 blankus; + + drm_mode_set_crtcinfo(mode, CRTC_INTERLACE_HALVE_V | CRTC_STEREO_DOUBLE); + + /* + * DRM modes are defined in terms of a repeating interval + * starting with the active display area. The hardware modes + * are defined in terms of a repeating interval starting one + * unit (pixel or line) into the sync pulse. So, add bias. + */ + + m->h.active = mode->crtc_htotal; + m->h.synce = mode->crtc_hsync_end - mode->crtc_hsync_start - 1; + m->h.blanke = mode->crtc_hblank_end - mode->crtc_hsync_start - 1; + m->h.blanks = m->h.blanke + mode->crtc_hdisplay; + + m->v.active = mode->crtc_vtotal; + m->v.synce = mode->crtc_vsync_end - mode->crtc_vsync_start - 1; + m->v.blanke = mode->crtc_vblank_end - mode->crtc_vsync_start - 1; + m->v.blanks = m->v.blanke + mode->crtc_vdisplay; + + /*XXX: Safe underestimate, even "0" works */ + blankus = (m->v.active - mode->crtc_vdisplay - 2) * m->h.active; + blankus *= 1000; + blankus /= mode->crtc_clock; + m->v.blankus = blankus; + + if (mode->flags & DRM_MODE_FLAG_INTERLACE) { + m->v.blank2e = m->v.active + m->v.blanke; + m->v.blank2s = m->v.blank2e + mode->crtc_vdisplay; + m->v.active = (m->v.active * 2) + 1; + m->interlace = true; + } else { + m->v.blank2e = 0; + m->v.blank2s = 1; + m->interlace = false; + } + m->clock = mode->crtc_clock; + + asyh->or.nhsync = !!(mode->flags & DRM_MODE_FLAG_NHSYNC); + asyh->or.nvsync = !!(mode->flags & DRM_MODE_FLAG_NVSYNC); + asyh->set.or = head->func->or != NULL; + asyh->set.mode = true; +} + +static int +nv50_head_atomic_check(struct drm_crtc *crtc, struct drm_crtc_state *state) +{ + struct nouveau_drm *drm = nouveau_drm(crtc->dev); + struct nv50_head *head = nv50_head(crtc); + struct nv50_head_atom *armh = nv50_head_atom(crtc->state); + struct nv50_head_atom *asyh = nv50_head_atom(state); + struct nouveau_conn_atom *asyc = NULL; + struct drm_connector_state *conns; + struct drm_connector *conn; + int i; + + NV_ATOMIC(drm, "%s atomic_check %d\n", crtc->name, asyh->state.active); + if (asyh->state.active) { + for_each_new_connector_in_state(asyh->state.state, conn, conns, i) { + if (conns->crtc == crtc) { + asyc = nouveau_conn_atom(conns); + break; + } + } + + if (armh->state.active) { + if (asyc) { + if (asyh->state.mode_changed) + asyc->set.scaler = true; + if (armh->base.depth != asyh->base.depth) + asyc->set.dither = true; + } + } else { + if (asyc) + asyc->set.mask = ~0; + asyh->set.mask = ~0; + asyh->set.or = head->func->or != NULL; + } + + if (asyh->state.mode_changed) + nv50_head_atomic_check_mode(head, asyh); + + if (asyh->state.color_mgmt_changed || + memcmp(&armh->wndw, &asyh->wndw, sizeof(asyh->wndw))) { + int ret = nv50_head_atomic_check_lut(head, asyh); + if (ret) + return ret; + + asyh->olut.visible = asyh->olut.handle != 0; + } + + if (asyc) { + if (asyc->set.scaler) + nv50_head_atomic_check_view(armh, asyh, asyc); + if (asyc->set.dither) + nv50_head_atomic_check_dither(armh, asyh, asyc); + if (asyc->set.procamp) + nv50_head_atomic_check_procamp(armh, asyh, asyc); + } + + if (head->func->core_calc) { + head->func->core_calc(head, asyh); + if (!asyh->core.visible) + asyh->olut.visible = false; + } + + asyh->set.base = armh->base.cpp != asyh->base.cpp; + asyh->set.ovly = armh->ovly.cpp != asyh->ovly.cpp; + } else { + asyh->olut.visible = false; + asyh->core.visible = false; + asyh->curs.visible = false; + asyh->base.cpp = 0; + asyh->ovly.cpp = 0; + } + + if (!drm_atomic_crtc_needs_modeset(&asyh->state)) { + if (asyh->core.visible) { + if (memcmp(&armh->core, &asyh->core, sizeof(asyh->core))) + asyh->set.core = true; + } else + if (armh->core.visible) { + asyh->clr.core = true; + } + + if (asyh->curs.visible) { + if (memcmp(&armh->curs, &asyh->curs, sizeof(asyh->curs))) + asyh->set.curs = true; + } else + if (armh->curs.visible) { + asyh->clr.curs = true; + } + + if (asyh->olut.visible) { + if (memcmp(&armh->olut, &asyh->olut, sizeof(asyh->olut))) + asyh->set.olut = true; + } else + if (armh->olut.visible) { + asyh->clr.olut = true; + } + } else { + asyh->clr.olut = armh->olut.visible; + asyh->clr.core = armh->core.visible; + asyh->clr.curs = armh->curs.visible; + asyh->set.olut = asyh->olut.visible; + asyh->set.core = asyh->core.visible; + asyh->set.curs = asyh->curs.visible; + } + + if (asyh->clr.mask || asyh->set.mask) + nv50_atom(asyh->state.state)->lock_core = true; + return 0; +} + +static const struct drm_crtc_helper_funcs +nv50_head_help = { + .atomic_check = nv50_head_atomic_check, +}; + +static void +nv50_head_atomic_destroy_state(struct drm_crtc *crtc, + struct drm_crtc_state *state) +{ + struct nv50_head_atom *asyh = nv50_head_atom(state); + __drm_atomic_helper_crtc_destroy_state(&asyh->state); + kfree(asyh); +} + +static struct drm_crtc_state * +nv50_head_atomic_duplicate_state(struct drm_crtc *crtc) +{ + struct nv50_head_atom *armh = nv50_head_atom(crtc->state); + struct nv50_head_atom *asyh; + if (!(asyh = kmalloc(sizeof(*asyh), GFP_KERNEL))) + return NULL; + __drm_atomic_helper_crtc_duplicate_state(crtc, &asyh->state); + asyh->wndw = armh->wndw; + asyh->view = armh->view; + asyh->mode = armh->mode; + asyh->olut = armh->olut; + asyh->core = armh->core; + asyh->curs = armh->curs; + asyh->base = armh->base; + asyh->ovly = armh->ovly; + asyh->dither = armh->dither; + asyh->procamp = armh->procamp; + asyh->clr.mask = 0; + asyh->set.mask = 0; + return &asyh->state; +} + +static void +__drm_atomic_helper_crtc_reset(struct drm_crtc *crtc, + struct drm_crtc_state *state) +{ + if (crtc->state) + crtc->funcs->atomic_destroy_state(crtc, crtc->state); + crtc->state = state; + crtc->state->crtc = crtc; +} + +static void +nv50_head_reset(struct drm_crtc *crtc) +{ + struct nv50_head_atom *asyh; + + if (WARN_ON(!(asyh = kzalloc(sizeof(*asyh), GFP_KERNEL)))) + return; + + __drm_atomic_helper_crtc_reset(crtc, &asyh->state); +} + +static void +nv50_head_destroy(struct drm_crtc *crtc) +{ + struct nv50_head *head = nv50_head(crtc); + nv50_lut_fini(&head->olut); + drm_crtc_cleanup(crtc); + kfree(head); +} + +static const struct drm_crtc_funcs +nv50_head_func = { + .reset = nv50_head_reset, + .gamma_set = drm_atomic_helper_legacy_gamma_set, + .destroy = nv50_head_destroy, + .set_config = drm_atomic_helper_set_config, + .page_flip = drm_atomic_helper_page_flip, + .atomic_duplicate_state = nv50_head_atomic_duplicate_state, + .atomic_destroy_state = nv50_head_atomic_destroy_state, +}; + +int +nv50_head_create(struct drm_device *dev, int index) +{ + struct nouveau_drm *drm = nouveau_drm(dev); + struct nv50_disp *disp = nv50_disp(dev); + struct nv50_head *head; + struct nv50_wndw *curs, *wndw; + struct drm_crtc *crtc; + int ret; + + head = kzalloc(sizeof(*head), GFP_KERNEL); + if (!head) + return -ENOMEM; + + head->func = disp->core->func->head; + head->base.index = index; + + if (disp->disp->object.oclass < GV100_DISP) { + ret = nv50_ovly_new(drm, head->base.index, &wndw); + ret = nv50_base_new(drm, head->base.index, &wndw); + } else { + ret = nv50_wndw_new(drm, DRM_PLANE_TYPE_OVERLAY, + head->base.index * 2 + 1, &wndw); + ret = nv50_wndw_new(drm, DRM_PLANE_TYPE_PRIMARY, + head->base.index * 2 + 0, &wndw); + } + if (ret == 0) + ret = nv50_curs_new(drm, head->base.index, &curs); + if (ret) { + kfree(head); + return ret; + } + + crtc = &head->base.base; + drm_crtc_init_with_planes(dev, crtc, &wndw->plane, &curs->plane, + &nv50_head_func, "head-%d", head->base.index); + drm_crtc_helper_add(crtc, &nv50_head_help); + drm_mode_crtc_set_gamma_size(crtc, 256); + + if (head->func->olut_set) { + ret = nv50_lut_init(disp, &drm->client.mmu, &head->olut); + if (ret) + goto out; + } + +out: + if (ret) + nv50_head_destroy(crtc); + return ret; +} diff --git a/drivers/gpu/drm/nouveau/dispnv50/head.h b/drivers/gpu/drm/nouveau/dispnv50/head.h new file mode 100644 index 000000000000..37b3248c6dae --- /dev/null +++ b/drivers/gpu/drm/nouveau/dispnv50/head.h @@ -0,0 +1,78 @@ +#ifndef __NV50_KMS_HEAD_H__ +#define __NV50_KMS_HEAD_H__ +#define nv50_head(c) container_of((c), struct nv50_head, base.base) +#include "disp.h" +#include "atom.h" +#include "lut.h" + +#include "nouveau_crtc.h" + +struct nv50_head { + const struct nv50_head_func *func; + struct nouveau_crtc base; + struct nv50_lut olut; +}; + +int nv50_head_create(struct drm_device *, int index); +void nv50_head_flush_set(struct nv50_head *, struct nv50_head_atom *); +void nv50_head_flush_clr(struct nv50_head *, struct nv50_head_atom *, bool y); + +struct nv50_head_func { + void (*view)(struct nv50_head *, struct nv50_head_atom *); + void (*mode)(struct nv50_head *, struct nv50_head_atom *); + void (*olut)(struct nv50_head *, struct nv50_head_atom *); + void (*olut_set)(struct nv50_head *, struct nv50_head_atom *); + void (*olut_clr)(struct nv50_head *); + void (*core_calc)(struct nv50_head *, struct nv50_head_atom *); + void (*core_set)(struct nv50_head *, struct nv50_head_atom *); + void (*core_clr)(struct nv50_head *); + int (*curs_layout)(struct nv50_head *, struct nv50_wndw_atom *, + struct nv50_head_atom *); + int (*curs_format)(struct nv50_head *, struct nv50_wndw_atom *, + struct nv50_head_atom *); + void (*curs_set)(struct nv50_head *, struct nv50_head_atom *); + void (*curs_clr)(struct nv50_head *); + void (*base)(struct nv50_head *, struct nv50_head_atom *); + void (*ovly)(struct nv50_head *, struct nv50_head_atom *); + void (*dither)(struct nv50_head *, struct nv50_head_atom *); + void (*procamp)(struct nv50_head *, struct nv50_head_atom *); + void (*or)(struct nv50_head *, struct nv50_head_atom *); +}; + +extern const struct nv50_head_func head507d; +void head507d_view(struct nv50_head *, struct nv50_head_atom *); +void head507d_mode(struct nv50_head *, struct nv50_head_atom *); +void head507d_olut(struct nv50_head *, struct nv50_head_atom *); +void head507d_core_calc(struct nv50_head *, struct nv50_head_atom *); +void head507d_core_clr(struct nv50_head *); +int head507d_curs_layout(struct nv50_head *, struct nv50_wndw_atom *, + struct nv50_head_atom *); +int head507d_curs_format(struct nv50_head *, struct nv50_wndw_atom *, + struct nv50_head_atom *); +void head507d_base(struct nv50_head *, struct nv50_head_atom *); +void head507d_ovly(struct nv50_head *, struct nv50_head_atom *); +void head507d_dither(struct nv50_head *, struct nv50_head_atom *); +void head507d_procamp(struct nv50_head *, struct nv50_head_atom *); + +extern const struct nv50_head_func head827d; + +extern const struct nv50_head_func head907d; +void head907d_view(struct nv50_head *, struct nv50_head_atom *); +void head907d_mode(struct nv50_head *, struct nv50_head_atom *); +void head907d_olut(struct nv50_head *, struct nv50_head_atom *); +void head907d_olut_set(struct nv50_head *, struct nv50_head_atom *); +void head907d_olut_clr(struct nv50_head *); +void head907d_core_set(struct nv50_head *, struct nv50_head_atom *); +void head907d_core_clr(struct nv50_head *); +void head907d_curs_set(struct nv50_head *, struct nv50_head_atom *); +void head907d_curs_clr(struct nv50_head *); +void head907d_ovly(struct nv50_head *, struct nv50_head_atom *); +void head907d_procamp(struct nv50_head *, struct nv50_head_atom *); +void head907d_or(struct nv50_head *, struct nv50_head_atom *); + +extern const struct nv50_head_func head917d; +int head917d_curs_layout(struct nv50_head *, struct nv50_wndw_atom *, + struct nv50_head_atom *); + +extern const struct nv50_head_func headc37d; +#endif diff --git a/drivers/gpu/drm/nouveau/dispnv50/head507d.c b/drivers/gpu/drm/nouveau/dispnv50/head507d.c new file mode 100644 index 000000000000..51bc5996fd37 --- /dev/null +++ b/drivers/gpu/drm/nouveau/dispnv50/head507d.c @@ -0,0 +1,325 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "head.h" +#include "core.h" + +void +head507d_procamp(struct nv50_head *head, struct nv50_head_atom *asyh) +{ + struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->core->chan; + u32 *push; + if ((push = evo_wait(core, 2))) { + evo_mthd(push, 0x08a8 + (head->base.index * 0x400), 1); + evo_data(push, asyh->procamp.sat.sin << 20 | + asyh->procamp.sat.cos << 8); + evo_kick(push, core); + } +} + +void +head507d_dither(struct nv50_head *head, struct nv50_head_atom *asyh) +{ + struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->core->chan; + u32 *push; + if ((push = evo_wait(core, 2))) { + evo_mthd(push, 0x08a0 + (head->base.index * 0x0400), 1); + evo_data(push, asyh->dither.mode << 3 | + asyh->dither.bits << 1 | + asyh->dither.enable); + evo_kick(push, core); + } +} + +void +head507d_ovly(struct nv50_head *head, struct nv50_head_atom *asyh) +{ + struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->core->chan; + u32 bounds = 0; + u32 *push; + + if (asyh->ovly.cpp) { + switch (asyh->ovly.cpp) { + case 4: bounds |= 0x00000300; break; + case 2: bounds |= 0x00000100; break; + default: + WARN_ON(1); + break; + } + bounds |= 0x00000001; + } else { + bounds |= 0x00000100; + } + + if ((push = evo_wait(core, 2))) { + evo_mthd(push, 0x0904 + head->base.index * 0x400, 1); + evo_data(push, bounds); + evo_kick(push, core); + } +} + +void +head507d_base(struct nv50_head *head, struct nv50_head_atom *asyh) +{ + struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->core->chan; + u32 bounds = 0; + u32 *push; + + if (asyh->base.cpp) { + switch (asyh->base.cpp) { + case 8: bounds |= 0x00000500; break; + case 4: bounds |= 0x00000300; break; + case 2: bounds |= 0x00000100; break; + case 1: bounds |= 0x00000000; break; + default: + WARN_ON(1); + break; + } + bounds |= 0x00000001; + } + + if ((push = evo_wait(core, 2))) { + evo_mthd(push, 0x0900 + head->base.index * 0x400, 1); + evo_data(push, bounds); + evo_kick(push, core); + } +} + +static void +head507d_curs_clr(struct nv50_head *head) +{ + struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->core->chan; + u32 *push; + if ((push = evo_wait(core, 2))) { + evo_mthd(push, 0x0880 + head->base.index * 0x400, 1); + evo_data(push, 0x05000000); + evo_kick(push, core); + } +} + +static void +head507d_curs_set(struct nv50_head *head, struct nv50_head_atom *asyh) +{ + struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->core->chan; + u32 *push; + if ((push = evo_wait(core, 3))) { + evo_mthd(push, 0x0880 + head->base.index * 0x400, 2); + evo_data(push, 0x80000000 | asyh->curs.layout << 26 | + asyh->curs.format << 24); + evo_data(push, asyh->curs.offset >> 8); + evo_kick(push, core); + } +} + +int +head507d_curs_format(struct nv50_head *head, struct nv50_wndw_atom *asyw, + struct nv50_head_atom *asyh) +{ + switch (asyw->image.format) { + case 0xcf: asyh->curs.format = 1; break; + default: + WARN_ON(1); + return -EINVAL; + } + return 0; +} + +int +head507d_curs_layout(struct nv50_head *head, struct nv50_wndw_atom *asyw, + struct nv50_head_atom *asyh) +{ + switch (asyw->image.w) { + case 32: asyh->curs.layout = 0; break; + case 64: asyh->curs.layout = 1; break; + default: + return -EINVAL; + } + return 0; +} + +void +head507d_core_clr(struct nv50_head *head) +{ + struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->core->chan; + u32 *push; + if ((push = evo_wait(core, 2))) { + evo_mthd(push, 0x0874 + head->base.index * 0x400, 1); + evo_data(push, 0x00000000); + evo_kick(push, core); + } +} + +static void +head507d_core_set(struct nv50_head *head, struct nv50_head_atom *asyh) +{ + struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->core->chan; + u32 *push; + if ((push = evo_wait(core, 9))) { + evo_mthd(push, 0x0860 + head->base.index * 0x400, 1); + evo_data(push, asyh->core.offset >> 8); + evo_mthd(push, 0x0868 + head->base.index * 0x400, 4); + evo_data(push, asyh->core.h << 16 | asyh->core.w); + evo_data(push, asyh->core.layout << 20 | + (asyh->core.pitch >> 8) << 8 | + asyh->core.blocks << 8 | + asyh->core.blockh); + evo_data(push, asyh->core.kind << 16 | + asyh->core.format << 8); + evo_data(push, asyh->core.handle); + evo_mthd(push, 0x08c0 + head->base.index * 0x400, 1); + evo_data(push, asyh->core.y << 16 | asyh->core.x); + evo_kick(push, core); + + /* EVO will complain with INVALID_STATE if we have an + * active cursor and (re)specify HeadSetContextDmaIso + * without also updating HeadSetOffsetCursor. + */ + asyh->set.curs = asyh->curs.visible; + asyh->set.olut = asyh->olut.handle != 0; + } +} + +void +head507d_core_calc(struct nv50_head *head, struct nv50_head_atom *asyh) +{ + struct nv50_disp *disp = nv50_disp(head->base.base.dev); + if ((asyh->core.visible = (asyh->base.cpp != 0))) { + asyh->core.x = asyh->base.x; + asyh->core.y = asyh->base.y; + asyh->core.w = asyh->base.w; + asyh->core.h = asyh->base.h; + } else + if ((asyh->core.visible = (asyh->ovly.cpp != 0)) || + (asyh->core.visible = asyh->curs.visible)) { + /*XXX: We need to either find some way of having the + * primary base layer appear black, while still + * being able to display the other layers, or we + * need to allocate a dummy black surface here. + */ + asyh->core.x = 0; + asyh->core.y = 0; + asyh->core.w = asyh->state.mode.hdisplay; + asyh->core.h = asyh->state.mode.vdisplay; + } + asyh->core.handle = disp->core->chan.vram.handle; + asyh->core.offset = 0; + asyh->core.format = 0xcf; + asyh->core.kind = 0; + asyh->core.layout = 1; + asyh->core.blockh = 0; + asyh->core.blocks = 0; + asyh->core.pitch = ALIGN(asyh->core.w, 64) * 4; +} + +static void +head507d_olut_clr(struct nv50_head *head) +{ + struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->core->chan; + u32 *push; + if ((push = evo_wait(core, 2))) { + evo_mthd(push, 0x0840 + (head->base.index * 0x400), 1); + evo_data(push, 0x00000000); + evo_kick(push, core); + } +} + +static void +head507d_olut_set(struct nv50_head *head, struct nv50_head_atom *asyh) +{ + struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->core->chan; + u32 *push; + if ((push = evo_wait(core, 3))) { + evo_mthd(push, 0x0840 + (head->base.index * 0x400), 2); + evo_data(push, 0x80000000 | asyh->olut.mode << 30); + evo_data(push, asyh->olut.offset >> 8); + evo_kick(push, core); + } +} + +void +head507d_olut(struct nv50_head *head, struct nv50_head_atom *asyh) +{ + if (asyh->base.cpp == 1) + asyh->olut.mode = 0; + else + asyh->olut.mode = 1; +} + +void +head507d_mode(struct nv50_head *head, struct nv50_head_atom *asyh) +{ + struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->core->chan; + struct nv50_head_mode *m = &asyh->mode; + u32 *push; + if ((push = evo_wait(core, 13))) { + evo_mthd(push, 0x0804 + (head->base.index * 0x400), 2); + evo_data(push, 0x00800000 | m->clock); + evo_data(push, m->interlace ? 0x00000002 : 0x00000000); + evo_mthd(push, 0x0810 + (head->base.index * 0x400), 7); + evo_data(push, 0x00000000); + evo_data(push, m->v.active << 16 | m->h.active ); + evo_data(push, m->v.synce << 16 | m->h.synce ); + evo_data(push, m->v.blanke << 16 | m->h.blanke ); + evo_data(push, m->v.blanks << 16 | m->h.blanks ); + evo_data(push, m->v.blank2e << 16 | m->v.blank2s); + evo_data(push, asyh->mode.v.blankus); + evo_mthd(push, 0x082c + (head->base.index * 0x400), 1); + evo_data(push, 0x00000000); + evo_kick(push, core); + } +} + +void +head507d_view(struct nv50_head *head, struct nv50_head_atom *asyh) +{ + struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->core->chan; + u32 *push; + if ((push = evo_wait(core, 7))) { + evo_mthd(push, 0x08a4 + (head->base.index * 0x400), 1); + evo_data(push, 0x00000000); + evo_mthd(push, 0x08c8 + (head->base.index * 0x400), 1); + evo_data(push, asyh->view.iH << 16 | asyh->view.iW); + evo_mthd(push, 0x08d8 + (head->base.index * 0x400), 2); + evo_data(push, asyh->view.oH << 16 | asyh->view.oW); + evo_data(push, asyh->view.oH << 16 | asyh->view.oW); + evo_kick(push, core); + } +} + +const struct nv50_head_func +head507d = { + .view = head507d_view, + .mode = head507d_mode, + .olut = head507d_olut, + .olut_set = head507d_olut_set, + .olut_clr = head507d_olut_clr, + .core_calc = head507d_core_calc, + .core_set = head507d_core_set, + .core_clr = head507d_core_clr, + .curs_layout = head507d_curs_layout, + .curs_format = head507d_curs_format, + .curs_set = head507d_curs_set, + .curs_clr = head507d_curs_clr, + .base = head507d_base, + .ovly = head507d_ovly, + .dither = head507d_dither, + .procamp = head507d_procamp, +}; diff --git a/drivers/gpu/drm/nouveau/dispnv50/head827d.c b/drivers/gpu/drm/nouveau/dispnv50/head827d.c new file mode 100644 index 000000000000..af5e7bd5978b --- /dev/null +++ b/drivers/gpu/drm/nouveau/dispnv50/head827d.c @@ -0,0 +1,124 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "head.h" +#include "core.h" + +static void +head827d_curs_clr(struct nv50_head *head) +{ + struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->core->chan; + u32 *push; + if ((push = evo_wait(core, 4))) { + evo_mthd(push, 0x0880 + head->base.index * 0x400, 1); + evo_data(push, 0x05000000); + evo_mthd(push, 0x089c + head->base.index * 0x400, 1); + evo_data(push, 0x00000000); + evo_kick(push, core); + } +} + +static void +head827d_curs_set(struct nv50_head *head, struct nv50_head_atom *asyh) +{ + struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->core->chan; + u32 *push; + if ((push = evo_wait(core, 5))) { + evo_mthd(push, 0x0880 + head->base.index * 0x400, 2); + evo_data(push, 0x80000000 | asyh->curs.layout << 26 | + asyh->curs.format << 24); + evo_data(push, asyh->curs.offset >> 8); + evo_mthd(push, 0x089c + head->base.index * 0x400, 1); + evo_data(push, asyh->curs.handle); + evo_kick(push, core); + } +} + +static void +head827d_core_set(struct nv50_head *head, struct nv50_head_atom *asyh) +{ + struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->core->chan; + u32 *push; + if ((push = evo_wait(core, 9))) { + evo_mthd(push, 0x0860 + head->base.index * 0x400, 1); + evo_data(push, asyh->core.offset >> 8); + evo_mthd(push, 0x0868 + head->base.index * 0x400, 4); + evo_data(push, asyh->core.h << 16 | asyh->core.w); + evo_data(push, asyh->core.layout << 20 | + (asyh->core.pitch >> 8) << 8 | + asyh->core.blocks << 8 | + asyh->core.blockh); + evo_data(push, asyh->core.format << 8); + evo_data(push, asyh->core.handle); + evo_mthd(push, 0x08c0 + head->base.index * 0x400, 1); + evo_data(push, asyh->core.y << 16 | asyh->core.x); + evo_kick(push, core); + } +} + +static void +head827d_olut_clr(struct nv50_head *head) +{ + struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->core->chan; + u32 *push; + if ((push = evo_wait(core, 4))) { + evo_mthd(push, 0x0840 + (head->base.index * 0x400), 1); + evo_data(push, 0x00000000); + evo_mthd(push, 0x085c + (head->base.index * 0x400), 1); + evo_data(push, 0x00000000); + evo_kick(push, core); + } +} + +static void +head827d_olut_set(struct nv50_head *head, struct nv50_head_atom *asyh) +{ + struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->core->chan; + u32 *push; + if ((push = evo_wait(core, 5))) { + evo_mthd(push, 0x0840 + (head->base.index * 0x400), 2); + evo_data(push, 0x80000000 | asyh->olut.mode << 30); + evo_data(push, asyh->olut.offset >> 8); + evo_mthd(push, 0x085c + (head->base.index * 0x400), 1); + evo_data(push, asyh->olut.handle); + evo_kick(push, core); + } +} + +const struct nv50_head_func +head827d = { + .view = head507d_view, + .mode = head507d_mode, + .olut = head507d_olut, + .olut_set = head827d_olut_set, + .olut_clr = head827d_olut_clr, + .core_calc = head507d_core_calc, + .core_set = head827d_core_set, + .core_clr = head507d_core_clr, + .curs_layout = head507d_curs_layout, + .curs_format = head507d_curs_format, + .curs_set = head827d_curs_set, + .curs_clr = head827d_curs_clr, + .base = head507d_base, + .ovly = head507d_ovly, + .dither = head507d_dither, + .procamp = head507d_procamp, +}; diff --git a/drivers/gpu/drm/nouveau/dispnv50/head907d.c b/drivers/gpu/drm/nouveau/dispnv50/head907d.c new file mode 100644 index 000000000000..633907163eb1 --- /dev/null +++ b/drivers/gpu/drm/nouveau/dispnv50/head907d.c @@ -0,0 +1,284 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "head.h" +#include "core.h" + +void +head907d_or(struct nv50_head *head, struct nv50_head_atom *asyh) +{ + struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->core->chan; + u32 *push; + if ((push = evo_wait(core, 3))) { + evo_mthd(push, 0x0404 + (head->base.index * 0x300), 2); + evo_data(push, 0x00000001 | asyh->or.depth << 6 | + asyh->or.nvsync << 4 | + asyh->or.nhsync << 3); + evo_data(push, 0x31ec6000 | head->base.index << 25 | + asyh->mode.interlace); + evo_kick(push, core); + } +} + +void +head907d_procamp(struct nv50_head *head, struct nv50_head_atom *asyh) +{ + struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->core->chan; + u32 *push; + if ((push = evo_wait(core, 2))) { + evo_mthd(push, 0x0498 + (head->base.index * 0x300), 1); + evo_data(push, asyh->procamp.sat.sin << 20 | + asyh->procamp.sat.cos << 8); + evo_kick(push, core); + } +} + +static void +head907d_dither(struct nv50_head *head, struct nv50_head_atom *asyh) +{ + struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->core->chan; + u32 *push; + if ((push = evo_wait(core, 2))) { + evo_mthd(push, 0x0490 + (head->base.index * 0x0300), 1); + evo_data(push, asyh->dither.mode << 3 | + asyh->dither.bits << 1 | + asyh->dither.enable); + evo_kick(push, core); + } +} + +void +head907d_ovly(struct nv50_head *head, struct nv50_head_atom *asyh) +{ + struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->core->chan; + u32 bounds = 0; + u32 *push; + + if (asyh->ovly.cpp) { + switch (asyh->ovly.cpp) { + case 8: bounds |= 0x00000500; break; + case 4: bounds |= 0x00000300; break; + case 2: bounds |= 0x00000100; break; + default: + WARN_ON(1); + break; + } + bounds |= 0x00000001; + } else { + bounds |= 0x00000100; + } + + if ((push = evo_wait(core, 2))) { + evo_mthd(push, 0x04d4 + head->base.index * 0x300, 1); + evo_data(push, bounds); + evo_kick(push, core); + } +} + +static void +head907d_base(struct nv50_head *head, struct nv50_head_atom *asyh) +{ + struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->core->chan; + u32 bounds = 0; + u32 *push; + + if (asyh->base.cpp) { + switch (asyh->base.cpp) { + case 8: bounds |= 0x00000500; break; + case 4: bounds |= 0x00000300; break; + case 2: bounds |= 0x00000100; break; + case 1: bounds |= 0x00000000; break; + default: + WARN_ON(1); + break; + } + bounds |= 0x00000001; + } + + if ((push = evo_wait(core, 2))) { + evo_mthd(push, 0x04d0 + head->base.index * 0x300, 1); + evo_data(push, bounds); + evo_kick(push, core); + } +} + +void +head907d_curs_clr(struct nv50_head *head) +{ + struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->core->chan; + u32 *push; + if ((push = evo_wait(core, 4))) { + evo_mthd(push, 0x0480 + head->base.index * 0x300, 1); + evo_data(push, 0x05000000); + evo_mthd(push, 0x048c + head->base.index * 0x300, 1); + evo_data(push, 0x00000000); + evo_kick(push, core); + } +} + +void +head907d_curs_set(struct nv50_head *head, struct nv50_head_atom *asyh) +{ + struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->core->chan; + u32 *push; + if ((push = evo_wait(core, 5))) { + evo_mthd(push, 0x0480 + head->base.index * 0x300, 2); + evo_data(push, 0x80000000 | asyh->curs.layout << 26 | + asyh->curs.format << 24); + evo_data(push, asyh->curs.offset >> 8); + evo_mthd(push, 0x048c + head->base.index * 0x300, 1); + evo_data(push, asyh->curs.handle); + evo_kick(push, core); + } +} + +void +head907d_core_clr(struct nv50_head *head) +{ + struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->core->chan; + u32 *push; + if ((push = evo_wait(core, 2))) { + evo_mthd(push, 0x0474 + head->base.index * 0x300, 1); + evo_data(push, 0x00000000); + evo_kick(push, core); + } +} + +void +head907d_core_set(struct nv50_head *head, struct nv50_head_atom *asyh) +{ + struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->core->chan; + u32 *push; + if ((push = evo_wait(core, 9))) { + evo_mthd(push, 0x0460 + head->base.index * 0x300, 1); + evo_data(push, asyh->core.offset >> 8); + evo_mthd(push, 0x0468 + head->base.index * 0x300, 4); + evo_data(push, asyh->core.h << 16 | asyh->core.w); + evo_data(push, asyh->core.layout << 24 | + (asyh->core.pitch >> 8) << 8 | + asyh->core.blocks << 8 | + asyh->core.blockh); + evo_data(push, asyh->core.format << 8); + evo_data(push, asyh->core.handle); + evo_mthd(push, 0x04b0 + head->base.index * 0x300, 1); + evo_data(push, asyh->core.y << 16 | asyh->core.x); + evo_kick(push, core); + } +} + +void +head907d_olut_clr(struct nv50_head *head) +{ + struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->core->chan; + u32 *push; + if ((push = evo_wait(core, 4))) { + evo_mthd(push, 0x0448 + (head->base.index * 0x300), 1); + evo_data(push, 0x00000000); + evo_mthd(push, 0x045c + (head->base.index * 0x300), 1); + evo_data(push, 0x00000000); + evo_kick(push, core); + } +} + +void +head907d_olut_set(struct nv50_head *head, struct nv50_head_atom *asyh) +{ + struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->core->chan; + u32 *push; + if ((push = evo_wait(core, 5))) { + evo_mthd(push, 0x0448 + (head->base.index * 0x300), 2); + evo_data(push, 0x80000000 | asyh->olut.mode << 24); + evo_data(push, asyh->olut.offset >> 8); + evo_mthd(push, 0x045c + (head->base.index * 0x300), 1); + evo_data(push, asyh->olut.handle); + evo_kick(push, core); + } +} + +void +head907d_olut(struct nv50_head *head, struct nv50_head_atom *asyh) +{ + asyh->olut.mode = 7; +} + +void +head907d_mode(struct nv50_head *head, struct nv50_head_atom *asyh) +{ + struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->core->chan; + struct nv50_head_mode *m = &asyh->mode; + u32 *push; + if ((push = evo_wait(core, 14))) { + evo_mthd(push, 0x0410 + (head->base.index * 0x300), 6); + evo_data(push, 0x00000000); + evo_data(push, m->v.active << 16 | m->h.active ); + evo_data(push, m->v.synce << 16 | m->h.synce ); + evo_data(push, m->v.blanke << 16 | m->h.blanke ); + evo_data(push, m->v.blanks << 16 | m->h.blanks ); + evo_data(push, m->v.blank2e << 16 | m->v.blank2s); + evo_mthd(push, 0x042c + (head->base.index * 0x300), 2); + evo_data(push, 0x00000000); /* ??? */ + evo_data(push, 0xffffff00); + evo_mthd(push, 0x0450 + (head->base.index * 0x300), 3); + evo_data(push, m->clock * 1000); + evo_data(push, 0x00200000); /* ??? */ + evo_data(push, m->clock * 1000); + evo_kick(push, core); + } +} + +void +head907d_view(struct nv50_head *head, struct nv50_head_atom *asyh) +{ + struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->core->chan; + u32 *push; + if ((push = evo_wait(core, 8))) { + evo_mthd(push, 0x0494 + (head->base.index * 0x300), 1); + evo_data(push, 0x00000000); + evo_mthd(push, 0x04b8 + (head->base.index * 0x300), 1); + evo_data(push, asyh->view.iH << 16 | asyh->view.iW); + evo_mthd(push, 0x04c0 + (head->base.index * 0x300), 3); + evo_data(push, asyh->view.oH << 16 | asyh->view.oW); + evo_data(push, asyh->view.oH << 16 | asyh->view.oW); + evo_data(push, asyh->view.oH << 16 | asyh->view.oW); + evo_kick(push, core); + } +} + +const struct nv50_head_func +head907d = { + .view = head907d_view, + .mode = head907d_mode, + .olut = head907d_olut, + .olut_set = head907d_olut_set, + .olut_clr = head907d_olut_clr, + .core_calc = head507d_core_calc, + .core_set = head907d_core_set, + .core_clr = head907d_core_clr, + .curs_layout = head507d_curs_layout, + .curs_format = head507d_curs_format, + .curs_set = head907d_curs_set, + .curs_clr = head907d_curs_clr, + .base = head907d_base, + .ovly = head907d_ovly, + .dither = head907d_dither, + .procamp = head907d_procamp, + .or = head907d_or, +}; diff --git a/drivers/gpu/drm/nouveau/dispnv50/head917d.c b/drivers/gpu/drm/nouveau/dispnv50/head917d.c new file mode 100644 index 000000000000..303df8459ca8 --- /dev/null +++ b/drivers/gpu/drm/nouveau/dispnv50/head917d.c @@ -0,0 +1,100 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "head.h" +#include "core.h" + +static void +head917d_dither(struct nv50_head *head, struct nv50_head_atom *asyh) +{ + struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->core->chan; + u32 *push; + if ((push = evo_wait(core, 2))) { + evo_mthd(push, 0x04a0 + (head->base.index * 0x0300), 1); + evo_data(push, asyh->dither.mode << 3 | + asyh->dither.bits << 1 | + asyh->dither.enable); + evo_kick(push, core); + } +} + +static void +head917d_base(struct nv50_head *head, struct nv50_head_atom *asyh) +{ + struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->core->chan; + u32 bounds = 0; + u32 *push; + + if (asyh->base.cpp) { + switch (asyh->base.cpp) { + case 8: bounds |= 0x00000500; break; + case 4: bounds |= 0x00000300; break; + case 2: bounds |= 0x00000100; break; + case 1: bounds |= 0x00000000; break; + default: + WARN_ON(1); + break; + } + bounds |= 0x00020001; + } + + if ((push = evo_wait(core, 2))) { + evo_mthd(push, 0x04d0 + head->base.index * 0x300, 1); + evo_data(push, bounds); + evo_kick(push, core); + } +} + +int +head917d_curs_layout(struct nv50_head *head, struct nv50_wndw_atom *asyw, + struct nv50_head_atom *asyh) +{ + switch (asyw->state.fb->width) { + case 32: asyh->curs.layout = 0; break; + case 64: asyh->curs.layout = 1; break; + case 128: asyh->curs.layout = 2; break; + case 256: asyh->curs.layout = 3; break; + default: + return -EINVAL; + } + return 0; +} + +const struct nv50_head_func +head917d = { + .view = head907d_view, + .mode = head907d_mode, + .olut = head907d_olut, + .olut_set = head907d_olut_set, + .olut_clr = head907d_olut_clr, + .core_calc = head507d_core_calc, + .core_set = head907d_core_set, + .core_clr = head907d_core_clr, + .curs_layout = head917d_curs_layout, + .curs_format = head507d_curs_format, + .curs_set = head907d_curs_set, + .curs_clr = head907d_curs_clr, + .base = head917d_base, + .ovly = head907d_ovly, + .dither = head917d_dither, + .procamp = head907d_procamp, + .or = head907d_or, +}; diff --git a/drivers/gpu/drm/nouveau/dispnv50/headc37d.c b/drivers/gpu/drm/nouveau/dispnv50/headc37d.c new file mode 100644 index 000000000000..989c14083066 --- /dev/null +++ b/drivers/gpu/drm/nouveau/dispnv50/headc37d.c @@ -0,0 +1,212 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "head.h" +#include "atom.h" +#include "core.h" + +static void +headc37d_or(struct nv50_head *head, struct nv50_head_atom *asyh) +{ + struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->core->chan; + u32 *push; + if ((push = evo_wait(core, 2))) { + /*XXX: This is a dirty hack until OR depth handling is + * improved later for deep colour etc. + */ + switch (asyh->or.depth) { + case 6: asyh->or.depth = 5; break; + case 5: asyh->or.depth = 4; break; + case 2: asyh->or.depth = 1; break; + case 0: asyh->or.depth = 4; break; + default: + WARN_ON(1); + break; + } + + evo_mthd(push, 0x2004 + (head->base.index * 0x400), 1); + evo_data(push, 0x00000001 | + asyh->or.depth << 4 | + asyh->or.nvsync << 3 | + asyh->or.nhsync << 2); + evo_kick(push, core); + } +} + +static void +headc37d_procamp(struct nv50_head *head, struct nv50_head_atom *asyh) +{ + struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->core->chan; + u32 *push; + if ((push = evo_wait(core, 2))) { + evo_mthd(push, 0x2000 + (head->base.index * 0x400), 1); + evo_data(push, 0x80000000 | + asyh->procamp.sat.sin << 16 | + asyh->procamp.sat.cos << 4); + evo_kick(push, core); + } +} + +static void +headc37d_dither(struct nv50_head *head, struct nv50_head_atom *asyh) +{ + struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->core->chan; + u32 *push; + if ((push = evo_wait(core, 2))) { + evo_mthd(push, 0x2018 + (head->base.index * 0x0400), 1); + evo_data(push, asyh->dither.mode << 8 | + asyh->dither.bits << 4 | + asyh->dither.enable); + evo_kick(push, core); + } +} + +static void +headc37d_curs_clr(struct nv50_head *head) +{ + struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->core->chan; + u32 *push; + if ((push = evo_wait(core, 4))) { + evo_mthd(push, 0x209c + head->base.index * 0x400, 1); + evo_data(push, 0x000000cf); + evo_mthd(push, 0x2088 + head->base.index * 0x400, 1); + evo_data(push, 0x00000000); + evo_kick(push, core); + } +} + +static void +headc37d_curs_set(struct nv50_head *head, struct nv50_head_atom *asyh) +{ + struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->core->chan; + u32 *push; + if ((push = evo_wait(core, 7))) { + evo_mthd(push, 0x209c + head->base.index * 0x400, 2); + evo_data(push, 0x80000000 | + asyh->curs.layout << 8 | + asyh->curs.format << 0); + evo_data(push, 0x000072ff); + evo_mthd(push, 0x2088 + head->base.index * 0x400, 1); + evo_data(push, asyh->curs.handle); + evo_mthd(push, 0x2090 + head->base.index * 0x400, 1); + evo_data(push, asyh->curs.offset >> 8); + evo_kick(push, core); + } +} + +static int +headc37d_curs_format(struct nv50_head *head, struct nv50_wndw_atom *asyw, + struct nv50_head_atom *asyh) +{ + asyh->curs.format = asyw->image.format; + return 0; +} + +static void +headc37d_olut_clr(struct nv50_head *head) +{ + struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->core->chan; + u32 *push; + if ((push = evo_wait(core, 2))) { + evo_mthd(push, 0x20ac + (head->base.index * 0x400), 1); + evo_data(push, 0x00000000); + evo_kick(push, core); + } +} + +static void +headc37d_olut_set(struct nv50_head *head, struct nv50_head_atom *asyh) +{ + struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->core->chan; + u32 *push; + if ((push = evo_wait(core, 4))) { + evo_mthd(push, 0x20a4 + (head->base.index * 0x400), 3); + evo_data(push, asyh->olut.output_mode << 8 | + asyh->olut.range << 4 | + asyh->olut.size); + evo_data(push, asyh->olut.offset >> 8); + evo_data(push, asyh->olut.handle); + evo_kick(push, core); + } +} + +static void +headc37d_olut(struct nv50_head *head, struct nv50_head_atom *asyh) +{ + asyh->olut.mode = 2; + asyh->olut.size = 0; + asyh->olut.range = 0; + asyh->olut.output_mode = 1; +} + +static void +headc37d_mode(struct nv50_head *head, struct nv50_head_atom *asyh) +{ + struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->core->chan; + struct nv50_head_mode *m = &asyh->mode; + u32 *push; + if ((push = evo_wait(core, 12))) { + evo_mthd(push, 0x2064 + (head->base.index * 0x400), 5); + evo_data(push, (m->v.active << 16) | m->h.active ); + evo_data(push, (m->v.synce << 16) | m->h.synce ); + evo_data(push, (m->v.blanke << 16) | m->h.blanke ); + evo_data(push, (m->v.blanks << 16) | m->h.blanks ); + evo_data(push, (m->v.blank2e << 16) | m->v.blank2s); + evo_mthd(push, 0x200c + (head->base.index * 0x400), 1); + evo_data(push, m->clock * 1000); + evo_mthd(push, 0x2028 + (head->base.index * 0x400), 1); + evo_data(push, m->clock * 1000); + /*XXX: HEAD_USAGE_BOUNDS, doesn't belong here. */ + evo_mthd(push, 0x2030 + (head->base.index * 0x400), 1); + evo_data(push, 0x00000124); + evo_kick(push, core); + } +} + +static void +headc37d_view(struct nv50_head *head, struct nv50_head_atom *asyh) +{ + struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->core->chan; + u32 *push; + if ((push = evo_wait(core, 4))) { + evo_mthd(push, 0x204c + (head->base.index * 0x400), 1); + evo_data(push, (asyh->view.iH << 16) | asyh->view.iW); + evo_mthd(push, 0x2058 + (head->base.index * 0x400), 1); + evo_data(push, (asyh->view.oH << 16) | asyh->view.oW); + evo_kick(push, core); + } +} + +const struct nv50_head_func +headc37d = { + .view = headc37d_view, + .mode = headc37d_mode, + .olut = headc37d_olut, + .olut_set = headc37d_olut_set, + .olut_clr = headc37d_olut_clr, + .curs_layout = head917d_curs_layout, + .curs_format = headc37d_curs_format, + .curs_set = headc37d_curs_set, + .curs_clr = headc37d_curs_clr, + .dither = headc37d_dither, + .procamp = headc37d_procamp, + .or = headc37d_or, +}; diff --git a/drivers/gpu/drm/nouveau/dispnv50/lut.c b/drivers/gpu/drm/nouveau/dispnv50/lut.c new file mode 100644 index 000000000000..a6b96ae2a22f --- /dev/null +++ b/drivers/gpu/drm/nouveau/dispnv50/lut.c @@ -0,0 +1,95 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "lut.h" +#include "disp.h" + +#include <drm/drm_color_mgmt.h> +#include <drm/drm_mode.h> +#include <drm/drm_property.h> + +#include <nvif/class.h> + +u32 +nv50_lut_load(struct nv50_lut *lut, bool legacy, int buffer, + struct drm_property_blob *blob) +{ + struct drm_color_lut *in = (struct drm_color_lut *)blob->data; + void __iomem *mem = lut->mem[buffer].object.map.ptr; + const int size = blob->length / sizeof(*in); + int bits, shift, i; + u16 zero, r, g, b; + u32 addr = lut->mem[buffer].addr; + + /* This can't happen.. But it shuts the compiler up. */ + if (WARN_ON(size != 256)) + return 0; + + if (legacy) { + bits = 11; + shift = 3; + zero = 0x0000; + } else { + bits = 14; + shift = 0; + zero = 0x6000; + } + + for (i = 0; i < size; i++) { + r = (drm_color_lut_extract(in[i]. red, bits) + zero) << shift; + g = (drm_color_lut_extract(in[i].green, bits) + zero) << shift; + b = (drm_color_lut_extract(in[i]. blue, bits) + zero) << shift; + writew(r, mem + (i * 0x08) + 0); + writew(g, mem + (i * 0x08) + 2); + writew(b, mem + (i * 0x08) + 4); + } + + /* INTERPOLATE modes require a "next" entry to interpolate with, + * so we replicate the last entry to deal with this for now. + */ + writew(r, mem + (i * 0x08) + 0); + writew(g, mem + (i * 0x08) + 2); + writew(b, mem + (i * 0x08) + 4); + return addr; +} + +void +nv50_lut_fini(struct nv50_lut *lut) +{ + int i; + for (i = 0; i < ARRAY_SIZE(lut->mem); i++) + nvif_mem_fini(&lut->mem[i]); +} + +int +nv50_lut_init(struct nv50_disp *disp, struct nvif_mmu *mmu, + struct nv50_lut *lut) +{ + const u32 size = disp->disp->object.oclass < GF110_DISP ? 257 : 1025; + int i; + for (i = 0; i < ARRAY_SIZE(lut->mem); i++) { + int ret = nvif_mem_init_map(mmu, NVIF_MEM_VRAM, size * 8, + &lut->mem[i]); + if (ret) + return ret; + } + return 0; +} diff --git a/drivers/gpu/drm/nouveau/dispnv50/lut.h b/drivers/gpu/drm/nouveau/dispnv50/lut.h new file mode 100644 index 000000000000..6d7b8352e4cb --- /dev/null +++ b/drivers/gpu/drm/nouveau/dispnv50/lut.h @@ -0,0 +1,15 @@ +#ifndef __NV50_KMS_LUT_H__ +#define __NV50_KMS_LUT_H__ +#include <nvif/mem.h> +struct drm_property_blob; +struct nv50_disp; + +struct nv50_lut { + struct nvif_mem mem[2]; +}; + +int nv50_lut_init(struct nv50_disp *, struct nvif_mmu *, struct nv50_lut *); +void nv50_lut_fini(struct nv50_lut *); +u32 nv50_lut_load(struct nv50_lut *, bool legacy, int buffer, + struct drm_property_blob *); +#endif diff --git a/drivers/gpu/drm/nouveau/dispnv50/oimm.c b/drivers/gpu/drm/nouveau/dispnv50/oimm.c new file mode 100644 index 000000000000..2a2841d344c8 --- /dev/null +++ b/drivers/gpu/drm/nouveau/dispnv50/oimm.c @@ -0,0 +1,51 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "oimm.h" + +#include <nvif/class.h> + +int +nv50_oimm_init(struct nouveau_drm *drm, struct nv50_wndw *wndw) +{ + static const struct { + s32 oclass; + int version; + int (*init)(struct nouveau_drm *, s32, struct nv50_wndw *); + } oimms[] = { + { GK104_DISP_OVERLAY, 0, oimm507b_init }, + { GF110_DISP_OVERLAY, 0, oimm507b_init }, + { GT214_DISP_OVERLAY, 0, oimm507b_init }, + { G82_DISP_OVERLAY, 0, oimm507b_init }, + { NV50_DISP_OVERLAY, 0, oimm507b_init }, + {} + }; + struct nv50_disp *disp = nv50_disp(drm->dev); + int cid; + + cid = nvif_mclass(&disp->disp->object, oimms); + if (cid < 0) { + NV_ERROR(drm, "No supported overlay immediate class\n"); + return cid; + } + + return oimms[cid].init(drm, oimms[cid].oclass, wndw); +} diff --git a/drivers/gpu/drm/nouveau/dispnv50/oimm.h b/drivers/gpu/drm/nouveau/dispnv50/oimm.h new file mode 100644 index 000000000000..6fa51f101e94 --- /dev/null +++ b/drivers/gpu/drm/nouveau/dispnv50/oimm.h @@ -0,0 +1,8 @@ +#ifndef __NV50_KMS_OIMM_H__ +#define __NV50_KMS_OIMM_H__ +#include "wndw.h" + +int oimm507b_init(struct nouveau_drm *, s32, struct nv50_wndw *); + +int nv50_oimm_init(struct nouveau_drm *, struct nv50_wndw *); +#endif diff --git a/drivers/gpu/drm/nouveau/dispnv50/oimm507b.c b/drivers/gpu/drm/nouveau/dispnv50/oimm507b.c new file mode 100644 index 000000000000..2ee404b3e19f --- /dev/null +++ b/drivers/gpu/drm/nouveau/dispnv50/oimm507b.c @@ -0,0 +1,52 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "oimm.h" + +#include <nvif/cl507b.h> + +static int +oimm507b_init_(const struct nv50_wimm_func *func, struct nouveau_drm *drm, + s32 oclass, struct nv50_wndw *wndw) +{ + struct nv50_disp_overlay_v0 args = { + .head = wndw->id, + }; + struct nv50_disp *disp = nv50_disp(drm->dev); + int ret; + + ret = nvif_object_init(&disp->disp->object, 0, oclass, &args, + sizeof(args), &wndw->wimm.base.user); + if (ret) { + NV_ERROR(drm, "oimm%04x allocation failed: %d\n", oclass, ret); + return ret; + } + + nvif_object_map(&wndw->wimm.base.user, NULL, 0); + wndw->immd = func; + return 0; +} + +int +oimm507b_init(struct nouveau_drm *drm, s32 oclass, struct nv50_wndw *wndw) +{ + return oimm507b_init_(&curs507a, drm, oclass, wndw); +} diff --git a/drivers/gpu/drm/nouveau/dispnv50/ovly.c b/drivers/gpu/drm/nouveau/dispnv50/ovly.c new file mode 100644 index 000000000000..90c246d47604 --- /dev/null +++ b/drivers/gpu/drm/nouveau/dispnv50/ovly.c @@ -0,0 +1,57 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "ovly.h" +#include "oimm.h" + +#include <nvif/class.h> + +int +nv50_ovly_new(struct nouveau_drm *drm, int head, struct nv50_wndw **pwndw) +{ + static const struct { + s32 oclass; + int version; + int (*new)(struct nouveau_drm *, int, s32, struct nv50_wndw **); + } ovlys[] = { + { GK104_DISP_OVERLAY_CONTROL_DMA, 0, ovly917e_new }, + { GF110_DISP_OVERLAY_CONTROL_DMA, 0, ovly907e_new }, + { GT214_DISP_OVERLAY_CHANNEL_DMA, 0, ovly827e_new }, + { GT200_DISP_OVERLAY_CHANNEL_DMA, 0, ovly827e_new }, + { G82_DISP_OVERLAY_CHANNEL_DMA, 0, ovly827e_new }, + { NV50_DISP_OVERLAY_CHANNEL_DMA, 0, ovly507e_new }, + {} + }; + struct nv50_disp *disp = nv50_disp(drm->dev); + int cid, ret; + + cid = nvif_mclass(&disp->disp->object, ovlys); + if (cid < 0) { + NV_ERROR(drm, "No supported overlay class\n"); + return cid; + } + + ret = ovlys[cid].new(drm, head, ovlys[cid].oclass, pwndw); + if (ret) + return ret; + + return nv50_oimm_init(drm, *pwndw); +} diff --git a/drivers/gpu/drm/nouveau/dispnv50/ovly.h b/drivers/gpu/drm/nouveau/dispnv50/ovly.h new file mode 100644 index 000000000000..4869d52d1786 --- /dev/null +++ b/drivers/gpu/drm/nouveau/dispnv50/ovly.h @@ -0,0 +1,30 @@ +#ifndef __NV50_KMS_OVLY_H__ +#define __NV50_KMS_OVLY_H__ +#include "wndw.h" + +int ovly507e_new(struct nouveau_drm *, int, s32, struct nv50_wndw **); +int ovly507e_new_(const struct nv50_wndw_func *, const u32 *format, + struct nouveau_drm *, int head, s32 oclass, + u32 interlock_data, struct nv50_wndw **); +int ovly507e_acquire(struct nv50_wndw *, struct nv50_wndw_atom *, + struct nv50_head_atom *); +void ovly507e_release(struct nv50_wndw *, struct nv50_wndw_atom *, + struct nv50_head_atom *); +void ovly507e_ntfy_set(struct nv50_wndw *, struct nv50_wndw_atom *); +void ovly507e_ntfy_clr(struct nv50_wndw *); +void ovly507e_image_clr(struct nv50_wndw *); +void ovly507e_scale_set(struct nv50_wndw *, struct nv50_wndw_atom *); +void ovly507e_update(struct nv50_wndw *, u32 *); + +extern const u32 ovly827e_format[]; +void ovly827e_ntfy_reset(struct nouveau_bo *, u32); +int ovly827e_ntfy_wait_begun(struct nouveau_bo *, u32, struct nvif_device *); + +extern const struct nv50_wndw_func ovly907e; + +int ovly827e_new(struct nouveau_drm *, int, s32, struct nv50_wndw **); +int ovly907e_new(struct nouveau_drm *, int, s32, struct nv50_wndw **); +int ovly917e_new(struct nouveau_drm *, int, s32, struct nv50_wndw **); + +int nv50_ovly_new(struct nouveau_drm *, int head, struct nv50_wndw **); +#endif diff --git a/drivers/gpu/drm/nouveau/dispnv50/ovly507e.c b/drivers/gpu/drm/nouveau/dispnv50/ovly507e.c new file mode 100644 index 000000000000..cc417664f823 --- /dev/null +++ b/drivers/gpu/drm/nouveau/dispnv50/ovly507e.c @@ -0,0 +1,217 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "ovly.h" +#include "atom.h" + +#include <drm/drm_atomic_helper.h> +#include <drm/drm_plane_helper.h> + +#include <nvif/cl507e.h> +#include <nvif/event.h> + +void +ovly507e_update(struct nv50_wndw *wndw, u32 *interlock) +{ + u32 *push; + if ((push = evo_wait(&wndw->wndw, 2))) { + evo_mthd(push, 0x0080, 1); + evo_data(push, interlock[NV50_DISP_INTERLOCK_CORE]); + evo_kick(push, &wndw->wndw); + } +} + +void +ovly507e_scale_set(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw) +{ + u32 *push; + if ((push = evo_wait(&wndw->wndw, 4))) { + evo_mthd(push, 0x00e0, 3); + evo_data(push, asyw->scale.sy << 16 | asyw->scale.sx); + evo_data(push, asyw->scale.sh << 16 | asyw->scale.sw); + evo_data(push, asyw->scale.dw); + evo_kick(push, &wndw->wndw); + } +} + +void +ovly507e_image_clr(struct nv50_wndw *wndw) +{ + u32 *push; + if ((push = evo_wait(&wndw->wndw, 4))) { + evo_mthd(push, 0x0084, 1); + evo_data(push, 0x00000000); + evo_mthd(push, 0x00c0, 1); + evo_data(push, 0x00000000); + evo_kick(push, &wndw->wndw); + } +} + +static void +ovly507e_image_set(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw) +{ + u32 *push; + if ((push = evo_wait(&wndw->wndw, 12))) { + evo_mthd(push, 0x0084, 1); + evo_data(push, asyw->image.interval << 4); + evo_mthd(push, 0x00c0, 1); + evo_data(push, asyw->image.handle[0]); + evo_mthd(push, 0x0100, 1); + evo_data(push, 0x00000002); + evo_mthd(push, 0x0800, 1); + evo_data(push, asyw->image.offset[0] >> 8); + evo_mthd(push, 0x0808, 3); + evo_data(push, asyw->image.h << 16 | asyw->image.w); + evo_data(push, asyw->image.layout << 20 | + (asyw->image.pitch[0] >> 8) << 8 | + asyw->image.blocks[0] << 8 | + asyw->image.blockh); + evo_data(push, asyw->image.kind << 16 | + asyw->image.format << 8 | + asyw->image.colorspace); + evo_kick(push, &wndw->wndw); + } +} + +void +ovly507e_ntfy_clr(struct nv50_wndw *wndw) +{ + u32 *push; + if ((push = evo_wait(&wndw->wndw, 2))) { + evo_mthd(push, 0x00a4, 1); + evo_data(push, 0x00000000); + evo_kick(push, &wndw->wndw); + } +} + +void +ovly507e_ntfy_set(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw) +{ + u32 *push; + if ((push = evo_wait(&wndw->wndw, 3))) { + evo_mthd(push, 0x00a0, 2); + evo_data(push, asyw->ntfy.awaken << 30 | asyw->ntfy.offset); + evo_data(push, asyw->ntfy.handle); + evo_kick(push, &wndw->wndw); + } +} + +void +ovly507e_release(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw, + struct nv50_head_atom *asyh) +{ + asyh->ovly.cpp = 0; +} + +int +ovly507e_acquire(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw, + struct nv50_head_atom *asyh) +{ + const struct drm_framebuffer *fb = asyw->state.fb; + int ret; + + ret = drm_atomic_helper_check_plane_state(&asyw->state, &asyh->state, + DRM_PLANE_HELPER_NO_SCALING, + DRM_PLANE_HELPER_NO_SCALING, + true, true); + if (ret) + return ret; + + asyh->ovly.cpp = fb->format->cpp[0]; + return 0; +} + +#include "nouveau_bo.h" + +static const struct nv50_wndw_func +ovly507e = { + .acquire = ovly507e_acquire, + .release = ovly507e_release, + .ntfy_set = ovly507e_ntfy_set, + .ntfy_clr = ovly507e_ntfy_clr, + .ntfy_reset = base507c_ntfy_reset, + .ntfy_wait_begun = base507c_ntfy_wait_begun, + .image_set = ovly507e_image_set, + .image_clr = ovly507e_image_clr, + .scale_set = ovly507e_scale_set, + .update = ovly507e_update, +}; + +static const u32 +ovly507e_format[] = { + DRM_FORMAT_YUYV, + DRM_FORMAT_UYVY, + DRM_FORMAT_XRGB8888, + DRM_FORMAT_ARGB8888, + DRM_FORMAT_XRGB1555, + DRM_FORMAT_ARGB1555, + 0 +}; + +int +ovly507e_new_(const struct nv50_wndw_func *func, const u32 *format, + struct nouveau_drm *drm, int head, s32 oclass, u32 interlock_data, + struct nv50_wndw **pwndw) +{ + struct nv50_disp_overlay_channel_dma_v0 args = { + .head = head, + }; + struct nv50_disp *disp = nv50_disp(drm->dev); + struct nv50_wndw *wndw; + int ret; + + ret = nv50_wndw_new_(func, drm->dev, DRM_PLANE_TYPE_OVERLAY, + "ovly", head, format, BIT(head), + NV50_DISP_INTERLOCK_OVLY, interlock_data, + &wndw); + if (*pwndw = wndw, ret) + return ret; + + ret = nv50_dmac_create(&drm->client.device, &disp->disp->object, + &oclass, 0, &args, sizeof(args), + disp->sync->bo.offset, &wndw->wndw); + if (ret) { + NV_ERROR(drm, "ovly%04x allocation failed: %d\n", oclass, ret); + return ret; + } + + ret = nvif_notify_init(&wndw->wndw.base.user, wndw->notify.func, false, + NV50_DISP_OVERLAY_CHANNEL_DMA_V0_NTFY_UEVENT, + &(struct nvif_notify_uevent_req) {}, + sizeof(struct nvif_notify_uevent_req), + sizeof(struct nvif_notify_uevent_rep), + &wndw->notify); + if (ret) + return ret; + + wndw->ntfy = NV50_DISP_OVLY_NTFY(wndw->id); + wndw->sema = NV50_DISP_OVLY_SEM0(wndw->id); + wndw->data = 0x00000000; + return 0; +} + +int +ovly507e_new(struct nouveau_drm *drm, int head, s32 oclass, + struct nv50_wndw **pwndw) +{ + return ovly507e_new_(&ovly507e, ovly507e_format, drm, head, oclass, + 0x00000004 << (head * 8), pwndw); +} diff --git a/drivers/gpu/drm/nouveau/dispnv50/ovly827e.c b/drivers/gpu/drm/nouveau/dispnv50/ovly827e.c new file mode 100644 index 000000000000..aaa9fe5a4fc8 --- /dev/null +++ b/drivers/gpu/drm/nouveau/dispnv50/ovly827e.c @@ -0,0 +1,107 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "ovly.h" +#include "atom.h" + +#include <nouveau_bo.h> + +static void +ovly827e_image_set(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw) +{ + u32 *push; + if ((push = evo_wait(&wndw->wndw, 12))) { + evo_mthd(push, 0x0084, 1); + evo_data(push, asyw->image.interval << 4); + evo_mthd(push, 0x00c0, 1); + evo_data(push, asyw->image.handle[0]); + evo_mthd(push, 0x0100, 1); + evo_data(push, 0x00000002); + evo_mthd(push, 0x0800, 1); + evo_data(push, asyw->image.offset[0] >> 8); + evo_mthd(push, 0x0808, 3); + evo_data(push, asyw->image.h << 16 | asyw->image.w); + evo_data(push, asyw->image.layout << 20 | + (asyw->image.pitch[0] >> 8) << 8 | + asyw->image.blocks[0] << 8 | + asyw->image.blockh); + evo_data(push, asyw->image.format << 8 | + asyw->image.colorspace); + evo_kick(push, &wndw->wndw); + } +} + +int +ovly827e_ntfy_wait_begun(struct nouveau_bo *bo, u32 offset, + struct nvif_device *device) +{ + s64 time = nvif_msec(device, 2000ULL, + u32 data = nouveau_bo_rd32(bo, offset / 4 + 3); + if ((data & 0xffff0000) == 0xffff0000) + break; + usleep_range(1, 2); + ); + return time < 0 ? time : 0; +} + +void +ovly827e_ntfy_reset(struct nouveau_bo *bo, u32 offset) +{ + nouveau_bo_wr32(bo, offset / 4 + 0, 0x00000000); + nouveau_bo_wr32(bo, offset / 4 + 1, 0x00000000); + nouveau_bo_wr32(bo, offset / 4 + 2, 0x00000000); + nouveau_bo_wr32(bo, offset / 4 + 3, 0x80000000); +} + +static const struct nv50_wndw_func +ovly827e = { + .acquire = ovly507e_acquire, + .release = ovly507e_release, + .ntfy_set = ovly507e_ntfy_set, + .ntfy_clr = ovly507e_ntfy_clr, + .ntfy_reset = ovly827e_ntfy_reset, + .ntfy_wait_begun = ovly827e_ntfy_wait_begun, + .image_set = ovly827e_image_set, + .image_clr = ovly507e_image_clr, + .scale_set = ovly507e_scale_set, + .update = ovly507e_update, +}; + +const u32 +ovly827e_format[] = { + DRM_FORMAT_YUYV, + DRM_FORMAT_UYVY, + DRM_FORMAT_XRGB8888, + DRM_FORMAT_ARGB8888, + DRM_FORMAT_XRGB1555, + DRM_FORMAT_ARGB1555, + DRM_FORMAT_XBGR2101010, + DRM_FORMAT_ABGR2101010, + 0 +}; + +int +ovly827e_new(struct nouveau_drm *drm, int head, s32 oclass, + struct nv50_wndw **pwndw) +{ + return ovly507e_new_(&ovly827e, ovly827e_format, drm, head, oclass, + 0x00000004 << (head * 8), pwndw); +} diff --git a/drivers/gpu/drm/nouveau/dispnv50/ovly907e.c b/drivers/gpu/drm/nouveau/dispnv50/ovly907e.c new file mode 100644 index 000000000000..a3ce53046015 --- /dev/null +++ b/drivers/gpu/drm/nouveau/dispnv50/ovly907e.c @@ -0,0 +1,70 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "ovly.h" +#include "atom.h" + +static void +ovly907e_image_set(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw) +{ + u32 *push; + if ((push = evo_wait(&wndw->wndw, 12))) { + evo_mthd(push, 0x0084, 1); + evo_data(push, asyw->image.interval << 4); + evo_mthd(push, 0x00c0, 1); + evo_data(push, asyw->image.handle[0]); + evo_mthd(push, 0x0100, 1); + evo_data(push, 0x00000002); + evo_mthd(push, 0x0400, 1); + evo_data(push, asyw->image.offset[0] >> 8); + evo_mthd(push, 0x0408, 3); + evo_data(push, asyw->image.h << 16 | asyw->image.w); + evo_data(push, asyw->image.layout << 24 | + (asyw->image.pitch[0] >> 8) << 8 | + asyw->image.blocks[0] << 8 | + asyw->image.blockh); + evo_data(push, asyw->image.format << 8 | + asyw->image.colorspace); + evo_kick(push, &wndw->wndw); + } +} + +const struct nv50_wndw_func +ovly907e = { + .acquire = ovly507e_acquire, + .release = ovly507e_release, + .ntfy_set = ovly507e_ntfy_set, + .ntfy_clr = ovly507e_ntfy_clr, + .ntfy_reset = ovly827e_ntfy_reset, + .ntfy_wait_begun = ovly827e_ntfy_wait_begun, + .image_set = ovly907e_image_set, + .image_clr = ovly507e_image_clr, + .scale_set = ovly507e_scale_set, + .update = ovly507e_update, +}; + +int +ovly907e_new(struct nouveau_drm *drm, int head, s32 oclass, + struct nv50_wndw **pwndw) +{ + return ovly507e_new_(&ovly907e, ovly827e_format, drm, head, oclass, + 0x00000004 << (head * 4), pwndw); +} diff --git a/drivers/gpu/drm/nouveau/dispnv50/ovly917e.c b/drivers/gpu/drm/nouveau/dispnv50/ovly917e.c new file mode 100644 index 000000000000..505fa7e78523 --- /dev/null +++ b/drivers/gpu/drm/nouveau/dispnv50/ovly917e.c @@ -0,0 +1,45 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "ovly.h" + +static const u32 +ovly917e_format[] = { + DRM_FORMAT_YUYV, + DRM_FORMAT_UYVY, + DRM_FORMAT_XRGB8888, + DRM_FORMAT_ARGB8888, + DRM_FORMAT_XRGB1555, + DRM_FORMAT_ARGB1555, + DRM_FORMAT_XBGR2101010, + DRM_FORMAT_ABGR2101010, + DRM_FORMAT_XRGB2101010, + DRM_FORMAT_ARGB2101010, + 0 +}; + +int +ovly917e_new(struct nouveau_drm *drm, int head, s32 oclass, + struct nv50_wndw **pwndw) +{ + return ovly507e_new_(&ovly907e, ovly917e_format, drm, head, oclass, + 0x00000004 << (head * 4), pwndw); +} diff --git a/drivers/gpu/drm/nouveau/dispnv50/pior507d.c b/drivers/gpu/drm/nouveau/dispnv50/pior507d.c new file mode 100644 index 000000000000..d2bac6a341dc --- /dev/null +++ b/drivers/gpu/drm/nouveau/dispnv50/pior507d.c @@ -0,0 +1,44 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "core.h" + +static void +pior507d_ctrl(struct nv50_core *core, int or, u32 ctrl, + struct nv50_head_atom *asyh) +{ + u32 *push; + if ((push = evo_wait(&core->chan, 2))) { + if (asyh) { + ctrl |= asyh->or.depth << 16; + ctrl |= asyh->or.nvsync << 13; + ctrl |= asyh->or.nhsync << 12; + } + evo_mthd(push, 0x0700 + (or * 0x040), 1); + evo_data(push, ctrl); + evo_kick(push, &core->chan); + } +} + +const struct nv50_outp_func +pior507d = { + .ctrl = pior507d_ctrl, +}; diff --git a/drivers/gpu/drm/nouveau/dispnv50/sor507d.c b/drivers/gpu/drm/nouveau/dispnv50/sor507d.c new file mode 100644 index 000000000000..5222fe6a9b21 --- /dev/null +++ b/drivers/gpu/drm/nouveau/dispnv50/sor507d.c @@ -0,0 +1,44 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "core.h" + +static void +sor507d_ctrl(struct nv50_core *core, int or, u32 ctrl, + struct nv50_head_atom *asyh) +{ + u32 *push; + if ((push = evo_wait(&core->chan, 2))) { + if (asyh) { + ctrl |= asyh->or.depth << 16; + ctrl |= asyh->or.nvsync << 13; + ctrl |= asyh->or.nhsync << 12; + } + evo_mthd(push, 0x0600 + (or * 0x40), 1); + evo_data(push, ctrl); + evo_kick(push, &core->chan); + } +} + +const struct nv50_outp_func +sor507d = { + .ctrl = sor507d_ctrl, +}; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/coregt200.c b/drivers/gpu/drm/nouveau/dispnv50/sor907d.c index b234547708fc..b0314ec11fb3 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/coregt200.c +++ b/drivers/gpu/drm/nouveau/dispnv50/sor907d.c @@ -1,5 +1,5 @@ /* - * Copyright 2015 Red Hat Inc. + * Copyright 2018 Red Hat Inc. * * Permission is hereby granted, free of charge, to any person obtaining a * copy of this software and associated documentation files (the "Software"), @@ -18,21 +18,24 @@ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR * OTHER DEALINGS IN THE SOFTWARE. - * - * Authors: Ben Skeggs <bskeggs@redhat.com> */ -#include "dmacnv50.h" -#include "rootnv50.h" +#include "core.h" #include <nvif/class.h> -const struct nv50_disp_dmac_oclass -gt200_disp_core_oclass = { - .base.oclass = GT200_DISP_CORE_CHANNEL_DMA, - .base.minver = 0, - .base.maxver = 0, - .ctor = nv50_disp_core_new, - .func = &nv50_disp_core_func, - .mthd = &g84_disp_core_chan_mthd, - .chid = 0, +static void +sor907d_ctrl(struct nv50_core *core, int or, u32 ctrl, + struct nv50_head_atom *asyh) +{ + u32 *push; + if ((push = evo_wait(&core->chan, 2))) { + evo_mthd(push, 0x0200 + (or * 0x20), 1); + evo_data(push, ctrl); + evo_kick(push, &core->chan); + } +} + +const struct nv50_outp_func +sor907d = { + .ctrl = sor907d_ctrl, }; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/cursgk104.c b/drivers/gpu/drm/nouveau/dispnv50/sorc37d.c index 2a99db4bf8f8..dff059241c5d 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/cursgk104.c +++ b/drivers/gpu/drm/nouveau/dispnv50/sorc37d.c @@ -1,5 +1,5 @@ /* - * Copyright 2012 Red Hat Inc. + * Copyright 2018 Red Hat Inc. * * Permission is hereby granted, free of charge, to any person obtaining a * copy of this software and associated documentation files (the "Software"), @@ -18,20 +18,22 @@ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR * OTHER DEALINGS IN THE SOFTWARE. - * - * Authors: Ben Skeggs */ -#include "channv50.h" -#include "rootnv50.h" +#include "core.h" -#include <nvif/class.h> +static void +sorc37d_ctrl(struct nv50_core *core, int or, u32 ctrl, + struct nv50_head_atom *asyh) +{ + u32 *push; + if ((push = evo_wait(&core->chan, 2))) { + evo_mthd(push, 0x0300 + (or * 0x20), 1); + evo_data(push, ctrl); + evo_kick(push, &core->chan); + } +} -const struct nv50_disp_pioc_oclass -gk104_disp_curs_oclass = { - .base.oclass = GK104_DISP_CURSOR, - .base.minver = 0, - .base.maxver = 0, - .ctor = nv50_disp_curs_new, - .func = &gf119_disp_pioc_func, - .chid = { 13, 13 }, +const struct nv50_outp_func +sorc37d = { + .ctrl = sorc37d_ctrl, }; diff --git a/drivers/gpu/drm/nouveau/dispnv50/wimm.c b/drivers/gpu/drm/nouveau/dispnv50/wimm.c new file mode 100644 index 000000000000..fc36e0696407 --- /dev/null +++ b/drivers/gpu/drm/nouveau/dispnv50/wimm.c @@ -0,0 +1,47 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "wimm.h" + +#include <nvif/class.h> + +int +nv50_wimm_init(struct nouveau_drm *drm, struct nv50_wndw *wndw) +{ + struct { + s32 oclass; + int version; + int (*init)(struct nouveau_drm *, s32, struct nv50_wndw *); + } wimms[] = { + { GV100_DISP_WINDOW_IMM_CHANNEL_DMA, 0, wimmc37b_init }, + {} + }; + struct nv50_disp *disp = nv50_disp(drm->dev); + int cid; + + cid = nvif_mclass(&disp->disp->object, wimms); + if (cid < 0) { + NV_ERROR(drm, "No supported window immediate class\n"); + return cid; + } + + return wimms[cid].init(drm, wimms[cid].oclass, wndw); +} diff --git a/drivers/gpu/drm/nouveau/dispnv50/wimm.h b/drivers/gpu/drm/nouveau/dispnv50/wimm.h new file mode 100644 index 000000000000..363052309be9 --- /dev/null +++ b/drivers/gpu/drm/nouveau/dispnv50/wimm.h @@ -0,0 +1,8 @@ +#ifndef __NV50_KMS_WIMM_H__ +#define __NV50_KMS_WIMM_H__ +#include "wndw.h" + +int nv50_wimm_init(struct nouveau_drm *drm, struct nv50_wndw *); + +int wimmc37b_init(struct nouveau_drm *, s32, struct nv50_wndw *); +#endif diff --git a/drivers/gpu/drm/nouveau/dispnv50/wimmc37b.c b/drivers/gpu/drm/nouveau/dispnv50/wimmc37b.c new file mode 100644 index 000000000000..9103b8494279 --- /dev/null +++ b/drivers/gpu/drm/nouveau/dispnv50/wimmc37b.c @@ -0,0 +1,86 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "wimm.h" +#include "atom.h" +#include "wndw.h" + +#include <nvif/clc37b.h> + +static void +wimmc37b_update(struct nv50_wndw *wndw, u32 *interlock) +{ + u32 *push; + if ((push = evo_wait(&wndw->wimm, 2))) { + evo_mthd(push, 0x0200, 1); + if (interlock[NV50_DISP_INTERLOCK_WNDW] & wndw->interlock.data) + evo_data(push, 0x00000003); + else + evo_data(push, 0x00000001); + evo_kick(push, &wndw->wimm); + } +} + +static void +wimmc37b_point(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw) +{ + u32 *push; + if ((push = evo_wait(&wndw->wimm, 2))) { + evo_mthd(push, 0x0208, 1); + evo_data(push, asyw->point.y << 16 | asyw->point.x); + evo_kick(push, &wndw->wimm); + } +} + +static const struct nv50_wimm_func +wimmc37b = { + .point = wimmc37b_point, + .update = wimmc37b_update, +}; + +static int +wimmc37b_init_(const struct nv50_wimm_func *func, struct nouveau_drm *drm, + s32 oclass, struct nv50_wndw *wndw) +{ + struct nvc37b_window_imm_channel_dma_v0 args = { + .pushbuf = 0xb0007b00 | wndw->id, + .index = wndw->id, + }; + struct nv50_disp *disp = nv50_disp(drm->dev); + int ret; + + ret = nv50_dmac_create(&drm->client.device, &disp->disp->object, + &oclass, 0, &args, sizeof(args), 0, + &wndw->wimm); + if (ret) { + NV_ERROR(drm, "wimm%04x allocation failed: %d\n", oclass, ret); + return ret; + } + + wndw->immd = func; + return 0; +} + +int +wimmc37b_init(struct nouveau_drm *drm, s32 oclass, struct nv50_wndw *wndw) +{ + return wimmc37b_init_(&wimmc37b, drm, oclass, wndw); +} diff --git a/drivers/gpu/drm/nouveau/dispnv50/wndw.c b/drivers/gpu/drm/nouveau/dispnv50/wndw.c new file mode 100644 index 000000000000..224963b533a6 --- /dev/null +++ b/drivers/gpu/drm/nouveau/dispnv50/wndw.c @@ -0,0 +1,641 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "wndw.h" +#include "wimm.h" + +#include <nvif/class.h> +#include <nvif/cl0002.h> + +#include <drm/drm_atomic_helper.h> +#include "nouveau_bo.h" + +static void +nv50_wndw_ctxdma_del(struct nv50_wndw_ctxdma *ctxdma) +{ + nvif_object_fini(&ctxdma->object); + list_del(&ctxdma->head); + kfree(ctxdma); +} + +static struct nv50_wndw_ctxdma * +nv50_wndw_ctxdma_new(struct nv50_wndw *wndw, struct nouveau_framebuffer *fb) +{ + struct nouveau_drm *drm = nouveau_drm(fb->base.dev); + struct nv50_wndw_ctxdma *ctxdma; + const u8 kind = fb->nvbo->kind; + const u32 handle = 0xfb000000 | kind; + struct { + struct nv_dma_v0 base; + union { + struct nv50_dma_v0 nv50; + struct gf100_dma_v0 gf100; + struct gf119_dma_v0 gf119; + }; + } args = {}; + u32 argc = sizeof(args.base); + int ret; + + list_for_each_entry(ctxdma, &wndw->ctxdma.list, head) { + if (ctxdma->object.handle == handle) + return ctxdma; + } + + if (!(ctxdma = kzalloc(sizeof(*ctxdma), GFP_KERNEL))) + return ERR_PTR(-ENOMEM); + list_add(&ctxdma->head, &wndw->ctxdma.list); + + args.base.target = NV_DMA_V0_TARGET_VRAM; + args.base.access = NV_DMA_V0_ACCESS_RDWR; + args.base.start = 0; + args.base.limit = drm->client.device.info.ram_user - 1; + + if (drm->client.device.info.chipset < 0x80) { + args.nv50.part = NV50_DMA_V0_PART_256; + argc += sizeof(args.nv50); + } else + if (drm->client.device.info.chipset < 0xc0) { + args.nv50.part = NV50_DMA_V0_PART_256; + args.nv50.kind = kind; + argc += sizeof(args.nv50); + } else + if (drm->client.device.info.chipset < 0xd0) { + args.gf100.kind = kind; + argc += sizeof(args.gf100); + } else { + args.gf119.page = GF119_DMA_V0_PAGE_LP; + args.gf119.kind = kind; + argc += sizeof(args.gf119); + } + + ret = nvif_object_init(wndw->ctxdma.parent, handle, NV_DMA_IN_MEMORY, + &args, argc, &ctxdma->object); + if (ret) { + nv50_wndw_ctxdma_del(ctxdma); + return ERR_PTR(ret); + } + + return ctxdma; +} + +int +nv50_wndw_wait_armed(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw) +{ + struct nv50_disp *disp = nv50_disp(wndw->plane.dev); + if (asyw->set.ntfy) { + return wndw->func->ntfy_wait_begun(disp->sync, + asyw->ntfy.offset, + wndw->wndw.base.device); + } + return 0; +} + +void +nv50_wndw_flush_clr(struct nv50_wndw *wndw, u32 *interlock, bool flush, + struct nv50_wndw_atom *asyw) +{ + union nv50_wndw_atom_mask clr = { + .mask = asyw->clr.mask & ~(flush ? 0 : asyw->set.mask), + }; + if (clr.sema ) wndw->func-> sema_clr(wndw); + if (clr.ntfy ) wndw->func-> ntfy_clr(wndw); + if (clr.xlut ) wndw->func-> xlut_clr(wndw); + if (clr.image) wndw->func->image_clr(wndw); + + interlock[wndw->interlock.type] |= wndw->interlock.data; +} + +void +nv50_wndw_flush_set(struct nv50_wndw *wndw, u32 *interlock, + struct nv50_wndw_atom *asyw) +{ + if (interlock) { + asyw->image.mode = 0; + asyw->image.interval = 1; + } + + if (asyw->set.sema ) wndw->func->sema_set (wndw, asyw); + if (asyw->set.ntfy ) wndw->func->ntfy_set (wndw, asyw); + if (asyw->set.image) wndw->func->image_set(wndw, asyw); + + if (asyw->set.xlut ) { + if (asyw->ilut) { + asyw->xlut.i.offset = + nv50_lut_load(&wndw->ilut, + asyw->xlut.i.mode <= 1, + asyw->xlut.i.buffer, + asyw->ilut); + } + wndw->func->xlut_set(wndw, asyw); + } + + if (asyw->set.scale) wndw->func->scale_set(wndw, asyw); + if (asyw->set.point) { + if (asyw->set.point = false, asyw->set.mask) + interlock[wndw->interlock.type] |= wndw->interlock.data; + interlock[NV50_DISP_INTERLOCK_WIMM] |= wndw->interlock.data; + + wndw->immd->point(wndw, asyw); + wndw->immd->update(wndw, interlock); + } else { + interlock[wndw->interlock.type] |= wndw->interlock.data; + } +} + +void +nv50_wndw_ntfy_enable(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw) +{ + struct nv50_disp *disp = nv50_disp(wndw->plane.dev); + + asyw->ntfy.handle = wndw->wndw.sync.handle; + asyw->ntfy.offset = wndw->ntfy; + asyw->ntfy.awaken = false; + asyw->set.ntfy = true; + + wndw->func->ntfy_reset(disp->sync, wndw->ntfy); + wndw->ntfy ^= 0x10; +} + +static void +nv50_wndw_atomic_check_release(struct nv50_wndw *wndw, + struct nv50_wndw_atom *asyw, + struct nv50_head_atom *asyh) +{ + struct nouveau_drm *drm = nouveau_drm(wndw->plane.dev); + NV_ATOMIC(drm, "%s release\n", wndw->plane.name); + wndw->func->release(wndw, asyw, asyh); + asyw->ntfy.handle = 0; + asyw->sema.handle = 0; +} + +static int +nv50_wndw_atomic_check_acquire_yuv(struct nv50_wndw_atom *asyw) +{ + switch (asyw->state.fb->format->format) { + case DRM_FORMAT_YUYV: asyw->image.format = 0x28; break; + case DRM_FORMAT_UYVY: asyw->image.format = 0x29; break; + default: + WARN_ON(1); + return -EINVAL; + } + asyw->image.colorspace = 1; + return 0; +} + +static int +nv50_wndw_atomic_check_acquire_rgb(struct nv50_wndw_atom *asyw) +{ + switch (asyw->state.fb->format->format) { + case DRM_FORMAT_C8 : asyw->image.format = 0x1e; break; + case DRM_FORMAT_XRGB8888 : + case DRM_FORMAT_ARGB8888 : asyw->image.format = 0xcf; break; + case DRM_FORMAT_RGB565 : asyw->image.format = 0xe8; break; + case DRM_FORMAT_XRGB1555 : + case DRM_FORMAT_ARGB1555 : asyw->image.format = 0xe9; break; + case DRM_FORMAT_XBGR2101010: + case DRM_FORMAT_ABGR2101010: asyw->image.format = 0xd1; break; + case DRM_FORMAT_XBGR8888 : + case DRM_FORMAT_ABGR8888 : asyw->image.format = 0xd5; break; + case DRM_FORMAT_XRGB2101010: + case DRM_FORMAT_ARGB2101010: asyw->image.format = 0xdf; break; + default: + return -EINVAL; + } + asyw->image.colorspace = 0; + return 0; +} + +static int +nv50_wndw_atomic_check_acquire(struct nv50_wndw *wndw, bool modeset, + struct nv50_wndw_atom *armw, + struct nv50_wndw_atom *asyw, + struct nv50_head_atom *asyh) +{ + struct nouveau_framebuffer *fb = nouveau_framebuffer(asyw->state.fb); + struct nouveau_drm *drm = nouveau_drm(wndw->plane.dev); + int ret; + + NV_ATOMIC(drm, "%s acquire\n", wndw->plane.name); + + if (asyw->state.fb != armw->state.fb || !armw->visible || modeset) { + asyw->image.w = fb->base.width; + asyw->image.h = fb->base.height; + asyw->image.kind = fb->nvbo->kind; + + ret = nv50_wndw_atomic_check_acquire_rgb(asyw); + if (ret) { + ret = nv50_wndw_atomic_check_acquire_yuv(asyw); + if (ret) + return ret; + } + + if (asyw->image.kind) { + asyw->image.layout = 0; + if (drm->client.device.info.chipset >= 0xc0) + asyw->image.blockh = fb->nvbo->mode >> 4; + else + asyw->image.blockh = fb->nvbo->mode; + asyw->image.blocks[0] = fb->base.pitches[0] / 64; + asyw->image.pitch[0] = 0; + } else { + asyw->image.layout = 1; + asyw->image.blockh = 0; + asyw->image.blocks[0] = 0; + asyw->image.pitch[0] = fb->base.pitches[0]; + } + + if (!(asyh->state.pageflip_flags & DRM_MODE_PAGE_FLIP_ASYNC)) + asyw->image.interval = 1; + else + asyw->image.interval = 0; + asyw->image.mode = asyw->image.interval ? 0 : 1; + asyw->set.image = wndw->func->image_set != NULL; + } + + if (wndw->func->scale_set) { + asyw->scale.sx = asyw->state.src_x >> 16; + asyw->scale.sy = asyw->state.src_y >> 16; + asyw->scale.sw = asyw->state.src_w >> 16; + asyw->scale.sh = asyw->state.src_h >> 16; + asyw->scale.dw = asyw->state.crtc_w; + asyw->scale.dh = asyw->state.crtc_h; + if (memcmp(&armw->scale, &asyw->scale, sizeof(asyw->scale))) + asyw->set.scale = true; + } + + if (wndw->immd) { + asyw->point.x = asyw->state.crtc_x; + asyw->point.y = asyw->state.crtc_y; + if (memcmp(&armw->point, &asyw->point, sizeof(asyw->point))) + asyw->set.point = true; + } + + return wndw->func->acquire(wndw, asyw, asyh); +} + +static void +nv50_wndw_atomic_check_lut(struct nv50_wndw *wndw, + struct nv50_wndw_atom *armw, + struct nv50_wndw_atom *asyw, + struct nv50_head_atom *asyh) +{ + struct drm_property_blob *ilut = asyh->state.degamma_lut; + + /* I8 format without an input LUT makes no sense, and the + * HW error-checks for this. + * + * In order to handle legacy gamma, when there's no input + * LUT we need to steal the output LUT and use it instead. + */ + if (!ilut && asyw->state.fb->format->format == DRM_FORMAT_C8) { + /* This should be an error, but there's legacy clients + * that do a modeset before providing a gamma table. + * + * We keep the window disabled to avoid angering HW. + */ + if (!(ilut = asyh->state.gamma_lut)) { + asyw->visible = false; + return; + } + + if (wndw->func->ilut) + asyh->wndw.olut |= BIT(wndw->id); + } else { + asyh->wndw.olut &= ~BIT(wndw->id); + } + + /* Recalculate LUT state. */ + memset(&asyw->xlut, 0x00, sizeof(asyw->xlut)); + if ((asyw->ilut = wndw->func->ilut ? ilut : NULL)) { + wndw->func->ilut(wndw, asyw); + asyw->xlut.handle = wndw->wndw.vram.handle; + asyw->xlut.i.buffer = !asyw->xlut.i.buffer; + asyw->set.xlut = true; + } + + /* Handle setting base SET_OUTPUT_LUT_LO_ENABLE_USE_CORE_LUT. */ + if (wndw->func->olut_core && + (!armw->visible || (armw->xlut.handle && !asyw->xlut.handle))) + asyw->set.xlut = true; + + /* Can't do an immediate flip while changing the LUT. */ + asyh->state.pageflip_flags &= ~DRM_MODE_PAGE_FLIP_ASYNC; +} + +static int +nv50_wndw_atomic_check(struct drm_plane *plane, struct drm_plane_state *state) +{ + struct nouveau_drm *drm = nouveau_drm(plane->dev); + struct nv50_wndw *wndw = nv50_wndw(plane); + struct nv50_wndw_atom *armw = nv50_wndw_atom(wndw->plane.state); + struct nv50_wndw_atom *asyw = nv50_wndw_atom(state); + struct nv50_head_atom *harm = NULL, *asyh = NULL; + bool modeset = false; + int ret; + + NV_ATOMIC(drm, "%s atomic_check\n", plane->name); + + /* Fetch the assembly state for the head the window will belong to, + * and determine whether the window will be visible. + */ + if (asyw->state.crtc) { + asyh = nv50_head_atom_get(asyw->state.state, asyw->state.crtc); + if (IS_ERR(asyh)) + return PTR_ERR(asyh); + modeset = drm_atomic_crtc_needs_modeset(&asyh->state); + asyw->visible = asyh->state.active; + } else { + asyw->visible = false; + } + + /* Fetch assembly state for the head the window used to belong to. */ + if (armw->state.crtc) { + harm = nv50_head_atom_get(asyw->state.state, armw->state.crtc); + if (IS_ERR(harm)) + return PTR_ERR(harm); + } + + /* LUT configuration can potentially cause the window to be disabled. */ + if (asyw->visible && wndw->func->xlut_set && + (!armw->visible || + asyh->state.color_mgmt_changed || + asyw->state.fb->format->format != + armw->state.fb->format->format)) + nv50_wndw_atomic_check_lut(wndw, armw, asyw, asyh); + + /* Calculate new window state. */ + if (asyw->visible) { + ret = nv50_wndw_atomic_check_acquire(wndw, modeset, + armw, asyw, asyh); + if (ret) + return ret; + + asyh->wndw.mask |= BIT(wndw->id); + } else + if (armw->visible) { + nv50_wndw_atomic_check_release(wndw, asyw, harm); + harm->wndw.mask &= ~BIT(wndw->id); + } else { + return 0; + } + + /* Aside from the obvious case where the window is actively being + * disabled, we might also need to temporarily disable the window + * when performing certain modeset operations. + */ + if (!asyw->visible || modeset) { + asyw->clr.ntfy = armw->ntfy.handle != 0; + asyw->clr.sema = armw->sema.handle != 0; + asyw->clr.xlut = armw->xlut.handle != 0; + if (wndw->func->image_clr) + asyw->clr.image = armw->image.handle[0] != 0; + } + + return 0; +} + +static void +nv50_wndw_cleanup_fb(struct drm_plane *plane, struct drm_plane_state *old_state) +{ + struct nouveau_framebuffer *fb = nouveau_framebuffer(old_state->fb); + struct nouveau_drm *drm = nouveau_drm(plane->dev); + + NV_ATOMIC(drm, "%s cleanup: %p\n", plane->name, old_state->fb); + if (!old_state->fb) + return; + + nouveau_bo_unpin(fb->nvbo); +} + +static int +nv50_wndw_prepare_fb(struct drm_plane *plane, struct drm_plane_state *state) +{ + struct nouveau_framebuffer *fb = nouveau_framebuffer(state->fb); + struct nouveau_drm *drm = nouveau_drm(plane->dev); + struct nv50_wndw *wndw = nv50_wndw(plane); + struct nv50_wndw_atom *asyw = nv50_wndw_atom(state); + struct nv50_head_atom *asyh; + struct nv50_wndw_ctxdma *ctxdma; + int ret; + + NV_ATOMIC(drm, "%s prepare: %p\n", plane->name, state->fb); + if (!asyw->state.fb) + return 0; + + ret = nouveau_bo_pin(fb->nvbo, TTM_PL_FLAG_VRAM, true); + if (ret) + return ret; + + ctxdma = nv50_wndw_ctxdma_new(wndw, fb); + if (IS_ERR(ctxdma)) { + nouveau_bo_unpin(fb->nvbo); + return PTR_ERR(ctxdma); + } + + asyw->state.fence = reservation_object_get_excl_rcu(fb->nvbo->bo.resv); + asyw->image.handle[0] = ctxdma->object.handle; + asyw->image.offset[0] = fb->nvbo->bo.offset; + + if (wndw->func->prepare) { + asyh = nv50_head_atom_get(asyw->state.state, asyw->state.crtc); + if (IS_ERR(asyh)) + return PTR_ERR(asyh); + + wndw->func->prepare(wndw, asyh, asyw); + } + + return 0; +} + +static const struct drm_plane_helper_funcs +nv50_wndw_helper = { + .prepare_fb = nv50_wndw_prepare_fb, + .cleanup_fb = nv50_wndw_cleanup_fb, + .atomic_check = nv50_wndw_atomic_check, +}; + +static void +nv50_wndw_atomic_destroy_state(struct drm_plane *plane, + struct drm_plane_state *state) +{ + struct nv50_wndw_atom *asyw = nv50_wndw_atom(state); + __drm_atomic_helper_plane_destroy_state(&asyw->state); + kfree(asyw); +} + +static struct drm_plane_state * +nv50_wndw_atomic_duplicate_state(struct drm_plane *plane) +{ + struct nv50_wndw_atom *armw = nv50_wndw_atom(plane->state); + struct nv50_wndw_atom *asyw; + if (!(asyw = kmalloc(sizeof(*asyw), GFP_KERNEL))) + return NULL; + __drm_atomic_helper_plane_duplicate_state(plane, &asyw->state); + asyw->sema = armw->sema; + asyw->ntfy = armw->ntfy; + asyw->ilut = NULL; + asyw->xlut = armw->xlut; + asyw->image = armw->image; + asyw->point = armw->point; + asyw->clr.mask = 0; + asyw->set.mask = 0; + return &asyw->state; +} + +static void +nv50_wndw_reset(struct drm_plane *plane) +{ + struct nv50_wndw_atom *asyw; + + if (WARN_ON(!(asyw = kzalloc(sizeof(*asyw), GFP_KERNEL)))) + return; + + if (plane->state) + plane->funcs->atomic_destroy_state(plane, plane->state); + plane->state = &asyw->state; + plane->state->plane = plane; + plane->state->rotation = DRM_MODE_ROTATE_0; +} + +static void +nv50_wndw_destroy(struct drm_plane *plane) +{ + struct nv50_wndw *wndw = nv50_wndw(plane); + struct nv50_wndw_ctxdma *ctxdma, *ctxtmp; + + list_for_each_entry_safe(ctxdma, ctxtmp, &wndw->ctxdma.list, head) { + nv50_wndw_ctxdma_del(ctxdma); + } + + nvif_notify_fini(&wndw->notify); + nv50_dmac_destroy(&wndw->wimm); + nv50_dmac_destroy(&wndw->wndw); + + nv50_lut_fini(&wndw->ilut); + + drm_plane_cleanup(&wndw->plane); + kfree(wndw); +} + +const struct drm_plane_funcs +nv50_wndw = { + .update_plane = drm_atomic_helper_update_plane, + .disable_plane = drm_atomic_helper_disable_plane, + .destroy = nv50_wndw_destroy, + .reset = nv50_wndw_reset, + .atomic_duplicate_state = nv50_wndw_atomic_duplicate_state, + .atomic_destroy_state = nv50_wndw_atomic_destroy_state, +}; + +static int +nv50_wndw_notify(struct nvif_notify *notify) +{ + return NVIF_NOTIFY_KEEP; +} + +void +nv50_wndw_fini(struct nv50_wndw *wndw) +{ + nvif_notify_put(&wndw->notify); +} + +void +nv50_wndw_init(struct nv50_wndw *wndw) +{ + nvif_notify_get(&wndw->notify); +} + +int +nv50_wndw_new_(const struct nv50_wndw_func *func, struct drm_device *dev, + enum drm_plane_type type, const char *name, int index, + const u32 *format, u32 heads, + enum nv50_disp_interlock_type interlock_type, u32 interlock_data, + struct nv50_wndw **pwndw) +{ + struct nouveau_drm *drm = nouveau_drm(dev); + struct nvif_mmu *mmu = &drm->client.mmu; + struct nv50_disp *disp = nv50_disp(dev); + struct nv50_wndw *wndw; + int nformat; + int ret; + + if (!(wndw = *pwndw = kzalloc(sizeof(*wndw), GFP_KERNEL))) + return -ENOMEM; + wndw->func = func; + wndw->id = index; + wndw->interlock.type = interlock_type; + wndw->interlock.data = interlock_data; + wndw->ctxdma.parent = &wndw->wndw.base.user; + + wndw->ctxdma.parent = &wndw->wndw.base.user; + INIT_LIST_HEAD(&wndw->ctxdma.list); + + for (nformat = 0; format[nformat]; nformat++); + + ret = drm_universal_plane_init(dev, &wndw->plane, heads, &nv50_wndw, + format, nformat, NULL, + type, "%s-%d", name, index); + if (ret) { + kfree(*pwndw); + *pwndw = NULL; + return ret; + } + + drm_plane_helper_add(&wndw->plane, &nv50_wndw_helper); + + if (wndw->func->ilut) { + ret = nv50_lut_init(disp, mmu, &wndw->ilut); + if (ret) + return ret; + } + + wndw->notify.func = nv50_wndw_notify; + return 0; +} + +int +nv50_wndw_new(struct nouveau_drm *drm, enum drm_plane_type type, int index, + struct nv50_wndw **pwndw) +{ + struct { + s32 oclass; + int version; + int (*new)(struct nouveau_drm *, enum drm_plane_type, + int, s32, struct nv50_wndw **); + } wndws[] = { + { GV100_DISP_WINDOW_CHANNEL_DMA, 0, wndwc37e_new }, + {} + }; + struct nv50_disp *disp = nv50_disp(drm->dev); + int cid, ret; + + cid = nvif_mclass(&disp->disp->object, wndws); + if (cid < 0) { + NV_ERROR(drm, "No supported window class\n"); + return cid; + } + + ret = wndws[cid].new(drm, type, index, wndws[cid].oclass, pwndw); + if (ret) + return ret; + + return nv50_wimm_init(drm, *pwndw); +} diff --git a/drivers/gpu/drm/nouveau/dispnv50/wndw.h b/drivers/gpu/drm/nouveau/dispnv50/wndw.h new file mode 100644 index 000000000000..b0b6428034b0 --- /dev/null +++ b/drivers/gpu/drm/nouveau/dispnv50/wndw.h @@ -0,0 +1,96 @@ +#ifndef __NV50_KMS_WNDW_H__ +#define __NV50_KMS_WNDW_H__ +#define nv50_wndw(p) container_of((p), struct nv50_wndw, plane) +#include "disp.h" +#include "atom.h" +#include "lut.h" + +#include <nvif/notify.h> + +struct nv50_wndw_ctxdma { + struct list_head head; + struct nvif_object object; +}; + +struct nv50_wndw { + const struct nv50_wndw_func *func; + const struct nv50_wimm_func *immd; + int id; + struct nv50_disp_interlock interlock; + + struct { + struct nvif_object *parent; + struct list_head list; + } ctxdma; + + struct drm_plane plane; + + struct nv50_lut ilut; + + struct nv50_dmac wndw; + struct nv50_dmac wimm; + + struct nvif_notify notify; + u16 ntfy; + u16 sema; + u32 data; +}; + +int nv50_wndw_new_(const struct nv50_wndw_func *, struct drm_device *, + enum drm_plane_type, const char *name, int index, + const u32 *format, enum nv50_disp_interlock_type, + u32 interlock_data, u32 heads, struct nv50_wndw **); +void nv50_wndw_init(struct nv50_wndw *); +void nv50_wndw_fini(struct nv50_wndw *); +void nv50_wndw_flush_set(struct nv50_wndw *, u32 *interlock, + struct nv50_wndw_atom *); +void nv50_wndw_flush_clr(struct nv50_wndw *, u32 *interlock, bool flush, + struct nv50_wndw_atom *); +void nv50_wndw_ntfy_enable(struct nv50_wndw *, struct nv50_wndw_atom *); +int nv50_wndw_wait_armed(struct nv50_wndw *, struct nv50_wndw_atom *); + +struct nv50_wndw_func { + int (*acquire)(struct nv50_wndw *, struct nv50_wndw_atom *asyw, + struct nv50_head_atom *asyh); + void (*release)(struct nv50_wndw *, struct nv50_wndw_atom *asyw, + struct nv50_head_atom *asyh); + void (*prepare)(struct nv50_wndw *, struct nv50_head_atom *asyh, + struct nv50_wndw_atom *asyw); + + void (*sema_set)(struct nv50_wndw *, struct nv50_wndw_atom *); + void (*sema_clr)(struct nv50_wndw *); + void (*ntfy_reset)(struct nouveau_bo *, u32 offset); + void (*ntfy_set)(struct nv50_wndw *, struct nv50_wndw_atom *); + void (*ntfy_clr)(struct nv50_wndw *); + int (*ntfy_wait_begun)(struct nouveau_bo *, u32 offset, + struct nvif_device *); + void (*ilut)(struct nv50_wndw *, struct nv50_wndw_atom *); + bool olut_core; + void (*xlut_set)(struct nv50_wndw *, struct nv50_wndw_atom *); + void (*xlut_clr)(struct nv50_wndw *); + void (*image_set)(struct nv50_wndw *, struct nv50_wndw_atom *); + void (*image_clr)(struct nv50_wndw *); + void (*scale_set)(struct nv50_wndw *, struct nv50_wndw_atom *); + + void (*update)(struct nv50_wndw *, u32 *interlock); +}; + +extern const struct drm_plane_funcs nv50_wndw; + +void base507c_ntfy_reset(struct nouveau_bo *, u32); +int base507c_ntfy_wait_begun(struct nouveau_bo *, u32, struct nvif_device *); + +struct nv50_wimm_func { + void (*point)(struct nv50_wndw *, struct nv50_wndw_atom *); + + void (*update)(struct nv50_wndw *, u32 *interlock); +}; + +extern const struct nv50_wimm_func curs507a; + +int wndwc37e_new(struct nouveau_drm *, enum drm_plane_type, int, s32, + struct nv50_wndw **); + +int nv50_wndw_new(struct nouveau_drm *, enum drm_plane_type, int index, + struct nv50_wndw **); +#endif diff --git a/drivers/gpu/drm/nouveau/dispnv50/wndwc37e.c b/drivers/gpu/drm/nouveau/dispnv50/wndwc37e.c new file mode 100644 index 000000000000..44afb0f069a5 --- /dev/null +++ b/drivers/gpu/drm/nouveau/dispnv50/wndwc37e.c @@ -0,0 +1,278 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "wndw.h" +#include "atom.h" + +#include <drm/drm_atomic_helper.h> +#include <drm/drm_plane_helper.h> +#include <nouveau_bo.h> + +#include <nvif/clc37e.h> + +static void +wndwc37e_ilut_clr(struct nv50_wndw *wndw) +{ + u32 *push; + if ((push = evo_wait(&wndw->wndw, 2))) { + evo_mthd(push, 0x02b8, 1); + evo_data(push, 0x00000000); + evo_kick(push, &wndw->wndw); + } +} + +static void +wndwc37e_ilut_set(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw) +{ + u32 *push; + if ((push = evo_wait(&wndw->wndw, 4))) { + evo_mthd(push, 0x02b0, 3); + evo_data(push, asyw->xlut.i.output_mode << 8 | + asyw->xlut.i.range << 4 | + asyw->xlut.i.size); + evo_data(push, asyw->xlut.i.offset >> 8); + evo_data(push, asyw->xlut.handle); + evo_kick(push, &wndw->wndw); + } +} + +static void +wndwc37e_ilut(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw) +{ + asyw->xlut.i.mode = 2; + asyw->xlut.i.size = 0; + asyw->xlut.i.range = 0; + asyw->xlut.i.output_mode = 1; +} + +static void +wndwc37e_image_clr(struct nv50_wndw *wndw) +{ + u32 *push; + if ((push = evo_wait(&wndw->wndw, 4))) { + evo_mthd(push, 0x0308, 1); + evo_data(push, 0x00000000); + evo_mthd(push, 0x0240, 1); + evo_data(push, 0x00000000); + evo_kick(push, &wndw->wndw); + } +} + +static void +wndwc37e_image_set(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw) +{ + u32 *push; + + if (!(push = evo_wait(&wndw->wndw, 25))) + return; + + evo_mthd(push, 0x0308, 1); + evo_data(push, asyw->image.mode << 4 | asyw->image.interval); + evo_mthd(push, 0x0224, 4); + evo_data(push, asyw->image.h << 16 | asyw->image.w); + evo_data(push, asyw->image.layout << 4 | asyw->image.blockh); + evo_data(push, asyw->image.colorspace << 8 | asyw->image.format); + evo_data(push, asyw->image.blocks[0] | (asyw->image.pitch[0] >> 6)); + evo_mthd(push, 0x0240, 1); + evo_data(push, asyw->image.handle[0]); + evo_mthd(push, 0x0260, 1); + evo_data(push, asyw->image.offset[0] >> 8); + evo_mthd(push, 0x0290, 1); + evo_data(push, (asyw->state.src_y >> 16) << 16 | + (asyw->state.src_x >> 16)); + evo_mthd(push, 0x0298, 1); + evo_data(push, (asyw->state.src_h >> 16) << 16 | + (asyw->state.src_w >> 16)); + evo_mthd(push, 0x02a4, 1); + evo_data(push, asyw->state.crtc_h << 16 | + asyw->state.crtc_w); + + /*XXX: Composition-related stuff. Need to implement properly. */ + evo_mthd(push, 0x02ec, 1); + evo_data(push, (2 - (wndw->id & 1)) << 4); + evo_mthd(push, 0x02f4, 5); + evo_data(push, 0x00000011); + evo_data(push, 0xffff0000); + evo_data(push, 0xffff0000); + evo_data(push, 0xffff0000); + evo_data(push, 0xffff0000); + evo_kick(push, &wndw->wndw); +} + +static void +wndwc37e_ntfy_clr(struct nv50_wndw *wndw) +{ + u32 *push; + if ((push = evo_wait(&wndw->wndw, 2))) { + evo_mthd(push, 0x021c, 1); + evo_data(push, 0x00000000); + evo_kick(push, &wndw->wndw); + } +} + +static void +wndwc37e_ntfy_set(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw) +{ + u32 *push; + if ((push = evo_wait(&wndw->wndw, 3))) { + evo_mthd(push, 0x021c, 2); + evo_data(push, asyw->ntfy.handle); + evo_data(push, asyw->ntfy.offset | asyw->ntfy.awaken); + evo_kick(push, &wndw->wndw); + } +} + +static void +wndwc37e_sema_clr(struct nv50_wndw *wndw) +{ + u32 *push; + if ((push = evo_wait(&wndw->wndw, 2))) { + evo_mthd(push, 0x0218, 1); + evo_data(push, 0x00000000); + evo_kick(push, &wndw->wndw); + } +} + +static void +wndwc37e_sema_set(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw) +{ + u32 *push; + if ((push = evo_wait(&wndw->wndw, 5))) { + evo_mthd(push, 0x020c, 4); + evo_data(push, asyw->sema.offset); + evo_data(push, asyw->sema.acquire); + evo_data(push, asyw->sema.release); + evo_data(push, asyw->sema.handle); + evo_kick(push, &wndw->wndw); + } +} + +static void +wndwc37e_update(struct nv50_wndw *wndw, u32 *interlock) +{ + u32 *push; + if ((push = evo_wait(&wndw->wndw, 5))) { + evo_mthd(push, 0x0370, 2); + evo_data(push, interlock[NV50_DISP_INTERLOCK_CURS] << 1 | + interlock[NV50_DISP_INTERLOCK_CORE]); + evo_data(push, interlock[NV50_DISP_INTERLOCK_WNDW]); + evo_mthd(push, 0x0200, 1); + if (interlock[NV50_DISP_INTERLOCK_WIMM] & wndw->interlock.data) + evo_data(push, 0x00001001); + else + evo_data(push, 0x00000001); + evo_kick(push, &wndw->wndw); + } +} + +static void +wndwc37e_release(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw, + struct nv50_head_atom *asyh) +{ +} + +static int +wndwc37e_acquire(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw, + struct nv50_head_atom *asyh) +{ + return drm_atomic_helper_check_plane_state(&asyw->state, &asyh->state, + DRM_PLANE_HELPER_NO_SCALING, + DRM_PLANE_HELPER_NO_SCALING, + true, true); +} + +static const u32 +wndwc37e_format[] = { + DRM_FORMAT_C8, + DRM_FORMAT_YUYV, + DRM_FORMAT_UYVY, + DRM_FORMAT_XRGB8888, + DRM_FORMAT_ARGB8888, + DRM_FORMAT_RGB565, + DRM_FORMAT_XRGB1555, + DRM_FORMAT_ARGB1555, + DRM_FORMAT_XBGR2101010, + DRM_FORMAT_ABGR2101010, + DRM_FORMAT_XBGR8888, + DRM_FORMAT_ABGR8888, + DRM_FORMAT_XRGB2101010, + DRM_FORMAT_ARGB2101010, + 0 +}; + +static const struct nv50_wndw_func +wndwc37e = { + .acquire = wndwc37e_acquire, + .release = wndwc37e_release, + .sema_set = wndwc37e_sema_set, + .sema_clr = wndwc37e_sema_clr, + .ntfy_set = wndwc37e_ntfy_set, + .ntfy_clr = wndwc37e_ntfy_clr, + .ntfy_reset = corec37d_ntfy_init, + .ntfy_wait_begun = base507c_ntfy_wait_begun, + .ilut = wndwc37e_ilut, + .xlut_set = wndwc37e_ilut_set, + .xlut_clr = wndwc37e_ilut_clr, + .image_set = wndwc37e_image_set, + .image_clr = wndwc37e_image_clr, + .update = wndwc37e_update, +}; + +static int +wndwc37e_new_(const struct nv50_wndw_func *func, struct nouveau_drm *drm, + enum drm_plane_type type, int index, s32 oclass, u32 heads, + struct nv50_wndw **pwndw) +{ + struct nvc37e_window_channel_dma_v0 args = { + .pushbuf = 0xb0007e00 | index, + .index = index, + }; + struct nv50_disp *disp = nv50_disp(drm->dev); + struct nv50_wndw *wndw; + int ret; + + ret = nv50_wndw_new_(func, drm->dev, type, "wndw", index, + wndwc37e_format, heads, NV50_DISP_INTERLOCK_WNDW, + BIT(index), &wndw); + if (*pwndw = wndw, ret) + return ret; + + ret = nv50_dmac_create(&drm->client.device, &disp->disp->object, + &oclass, 0, &args, sizeof(args), + disp->sync->bo.offset, &wndw->wndw); + if (ret) { + NV_ERROR(drm, "qndw%04x allocation failed: %d\n", oclass, ret); + return ret; + } + + wndw->ntfy = NV50_DISP_WNDW_NTFY(wndw->id); + wndw->sema = NV50_DISP_WNDW_SEM0(wndw->id); + wndw->data = 0x00000000; + return 0; +} + +int +wndwc37e_new(struct nouveau_drm *drm, enum drm_plane_type type, int index, + s32 oclass, struct nv50_wndw **pwndw) +{ + return wndwc37e_new_(&wndwc37e, drm, type, index, oclass, + BIT(index >> 1), pwndw); +} diff --git a/drivers/gpu/drm/nouveau/include/nvif/cl0080.h b/drivers/gpu/drm/nouveau/include/nvif/cl0080.h index 2740278d226b..4f5233107f5f 100644 --- a/drivers/gpu/drm/nouveau/include/nvif/cl0080.h +++ b/drivers/gpu/drm/nouveau/include/nvif/cl0080.h @@ -31,6 +31,7 @@ struct nv_device_info_v0 { #define NV_DEVICE_INFO_V0_KEPLER 0x08 #define NV_DEVICE_INFO_V0_MAXWELL 0x09 #define NV_DEVICE_INFO_V0_PASCAL 0x0a +#define NV_DEVICE_INFO_V0_VOLTA 0x0b __u8 family; __u8 pad06[2]; __u64 ram_size; @@ -39,9 +40,55 @@ struct nv_device_info_v0 { char name[64]; }; +struct nv_device_info_v1 { + __u8 version; + __u8 count; + __u8 pad02[6]; + struct nv_device_info_v1_data { + __u64 mthd; /* NV_DEVICE_INFO_* (see below). */ + __u64 data; + } data[]; +}; + struct nv_device_time_v0 { __u8 version; __u8 pad01[7]; __u64 time; }; + +#define NV_DEVICE_INFO_UNIT (0xffffffffULL << 32) +#define NV_DEVICE_INFO(n) ((n) | (0x00000000ULL << 32)) +#define NV_DEVICE_FIFO(n) ((n) | (0x00000001ULL << 32)) + +/* This will be returned for unsupported queries. */ +#define NV_DEVICE_INFO_INVALID ~0ULL + +/* These return a mask of available engines of particular type. */ +#define NV_DEVICE_INFO_ENGINE_SW NV_DEVICE_INFO(0x00000000) +#define NV_DEVICE_INFO_ENGINE_GR NV_DEVICE_INFO(0x00000001) +#define NV_DEVICE_INFO_ENGINE_MPEG NV_DEVICE_INFO(0x00000002) +#define NV_DEVICE_INFO_ENGINE_ME NV_DEVICE_INFO(0x00000003) +#define NV_DEVICE_INFO_ENGINE_CIPHER NV_DEVICE_INFO(0x00000004) +#define NV_DEVICE_INFO_ENGINE_BSP NV_DEVICE_INFO(0x00000005) +#define NV_DEVICE_INFO_ENGINE_VP NV_DEVICE_INFO(0x00000006) +#define NV_DEVICE_INFO_ENGINE_CE NV_DEVICE_INFO(0x00000007) +#define NV_DEVICE_INFO_ENGINE_SEC NV_DEVICE_INFO(0x00000008) +#define NV_DEVICE_INFO_ENGINE_MSVLD NV_DEVICE_INFO(0x00000009) +#define NV_DEVICE_INFO_ENGINE_MSPDEC NV_DEVICE_INFO(0x0000000a) +#define NV_DEVICE_INFO_ENGINE_MSPPP NV_DEVICE_INFO(0x0000000b) +#define NV_DEVICE_INFO_ENGINE_MSENC NV_DEVICE_INFO(0x0000000c) +#define NV_DEVICE_INFO_ENGINE_VIC NV_DEVICE_INFO(0x0000000d) +#define NV_DEVICE_INFO_ENGINE_SEC2 NV_DEVICE_INFO(0x0000000e) +#define NV_DEVICE_INFO_ENGINE_NVDEC NV_DEVICE_INFO(0x0000000f) +#define NV_DEVICE_INFO_ENGINE_NVENC NV_DEVICE_INFO(0x00000010) + +/* Returns the number of available channels. */ +#define NV_DEVICE_FIFO_CHANNELS NV_DEVICE_FIFO(0x00000000) + +/* Returns a mask of available runlists. */ +#define NV_DEVICE_FIFO_RUNLISTS NV_DEVICE_FIFO(0x00000001) + +/* These return a mask of engines available on a particular runlist. */ +#define NV_DEVICE_FIFO_RUNLIST_ENGINES(n) ((n) + NV_DEVICE_FIFO(0x00000010)) +#define NV_DEVICE_FIFO_RUNLIST_ENGINES__SIZE 64 #endif diff --git a/drivers/gpu/drm/nouveau/include/nvif/cla06f.h b/drivers/gpu/drm/nouveau/include/nvif/cla06f.h index 56f5bd81e480..fbfcffc5feb2 100644 --- a/drivers/gpu/drm/nouveau/include/nvif/cla06f.h +++ b/drivers/gpu/drm/nouveau/include/nvif/cla06f.h @@ -4,25 +4,11 @@ struct kepler_channel_gpfifo_a_v0 { __u8 version; - __u8 pad01[5]; + __u8 pad01[1]; __u16 chid; -#define NVA06F_V0_ENGINE_SW 0x00000001 -#define NVA06F_V0_ENGINE_GR 0x00000002 -#define NVA06F_V0_ENGINE_SEC 0x00000004 -#define NVA06F_V0_ENGINE_MSVLD 0x00000010 -#define NVA06F_V0_ENGINE_MSPDEC 0x00000020 -#define NVA06F_V0_ENGINE_MSPPP 0x00000040 -#define NVA06F_V0_ENGINE_MSENC 0x00000080 -#define NVA06F_V0_ENGINE_VIC 0x00000100 -#define NVA06F_V0_ENGINE_NVDEC 0x00000200 -#define NVA06F_V0_ENGINE_NVENC0 0x00000400 -#define NVA06F_V0_ENGINE_NVENC1 0x00000800 -#define NVA06F_V0_ENGINE_CE0 0x00010000 -#define NVA06F_V0_ENGINE_CE1 0x00020000 -#define NVA06F_V0_ENGINE_CE2 0x00040000 - __u32 engines; __u32 ilength; __u64 ioffset; + __u64 runlist; __u64 vmm; }; diff --git a/drivers/gpu/drm/nouveau/include/nvif/class.h b/drivers/gpu/drm/nouveau/include/nvif/class.h index a7c5bf572788..6db56bd7d67e 100644 --- a/drivers/gpu/drm/nouveau/include/nvif/class.h +++ b/drivers/gpu/drm/nouveau/include/nvif/class.h @@ -52,6 +52,8 @@ #define NV04_DISP /* cl0046.h */ 0x00000046 +#define VOLTA_USERMODE_A 0x0000c361 + #define NV03_CHANNEL_DMA /* cl506b.h */ 0x0000006b #define NV10_CHANNEL_DMA /* cl506b.h */ 0x0000006e #define NV17_CHANNEL_DMA /* cl506b.h */ 0x0000176e @@ -66,6 +68,7 @@ #define KEPLER_CHANNEL_GPFIFO_B /* cla06f.h */ 0x0000a16f #define MAXWELL_CHANNEL_GPFIFO_A /* cla06f.h */ 0x0000b06f #define PASCAL_CHANNEL_GPFIFO_A /* cla06f.h */ 0x0000c06f +#define VOLTA_CHANNEL_GPFIFO_A /* cla06f.h */ 0x0000c36f #define NV50_DISP /* cl5070.h */ 0x00005070 #define G82_DISP /* cl5070.h */ 0x00008270 @@ -79,6 +82,7 @@ #define GM200_DISP /* cl5070.h */ 0x00009570 #define GP100_DISP /* cl5070.h */ 0x00009770 #define GP102_DISP /* cl5070.h */ 0x00009870 +#define GV100_DISP /* cl5070.h */ 0x0000c370 #define NV31_MPEG 0x00003174 #define G82_MPEG 0x00008274 @@ -90,6 +94,7 @@ #define GT214_DISP_CURSOR /* cl507a.h */ 0x0000857a #define GF110_DISP_CURSOR /* cl507a.h */ 0x0000907a #define GK104_DISP_CURSOR /* cl507a.h */ 0x0000917a +#define GV100_DISP_CURSOR /* cl507a.h */ 0x0000c37a #define NV50_DISP_OVERLAY /* cl507b.h */ 0x0000507b #define G82_DISP_OVERLAY /* cl507b.h */ 0x0000827b @@ -97,6 +102,8 @@ #define GF110_DISP_OVERLAY /* cl507b.h */ 0x0000907b #define GK104_DISP_OVERLAY /* cl507b.h */ 0x0000917b +#define GV100_DISP_WINDOW_IMM_CHANNEL_DMA /* clc37b.h */ 0x0000c37b + #define NV50_DISP_BASE_CHANNEL_DMA /* cl507c.h */ 0x0000507c #define G82_DISP_BASE_CHANNEL_DMA /* cl507c.h */ 0x0000827c #define GT200_DISP_BASE_CHANNEL_DMA /* cl507c.h */ 0x0000837c @@ -117,6 +124,7 @@ #define GM200_DISP_CORE_CHANNEL_DMA /* cl507d.h */ 0x0000957d #define GP100_DISP_CORE_CHANNEL_DMA /* cl507d.h */ 0x0000977d #define GP102_DISP_CORE_CHANNEL_DMA /* cl507d.h */ 0x0000987d +#define GV100_DISP_CORE_CHANNEL_DMA /* cl507d.h */ 0x0000c37d #define NV50_DISP_OVERLAY_CHANNEL_DMA /* cl507e.h */ 0x0000507e #define G82_DISP_OVERLAY_CHANNEL_DMA /* cl507e.h */ 0x0000827e @@ -125,6 +133,8 @@ #define GF110_DISP_OVERLAY_CONTROL_DMA /* cl507e.h */ 0x0000907e #define GK104_DISP_OVERLAY_CONTROL_DMA /* cl507e.h */ 0x0000917e +#define GV100_DISP_WINDOW_CHANNEL_DMA /* clc37e.h */ 0x0000c37e + #define NV50_TESLA 0x00005097 #define G82_TESLA 0x00008297 #define GT200_TESLA 0x00008397 @@ -145,6 +155,8 @@ #define PASCAL_A /* cl9097.h */ 0x0000c097 #define PASCAL_B /* cl9097.h */ 0x0000c197 +#define VOLTA_A /* cl9097.h */ 0x0000c397 + #define NV74_BSP 0x000074b0 #define GT212_MSVLD 0x000085b1 @@ -170,6 +182,7 @@ #define MAXWELL_DMA_COPY_A 0x0000b0b5 #define PASCAL_DMA_COPY_A 0x0000c0b5 #define PASCAL_DMA_COPY_B 0x0000c1b5 +#define VOLTA_DMA_COPY_A 0x0000c3b5 #define FERMI_DECOMPRESS 0x000090b8 @@ -183,6 +196,7 @@ #define MAXWELL_COMPUTE_B 0x0000b1c0 #define PASCAL_COMPUTE_A 0x0000c0c0 #define PASCAL_COMPUTE_B 0x0000c1c0 +#define VOLTA_COMPUTE_A 0x0000c3c0 #define NV74_CIPHER 0x000074c1 #endif diff --git a/drivers/gpu/drm/nouveau/include/nvif/clc37b.h b/drivers/gpu/drm/nouveau/include/nvif/clc37b.h new file mode 100644 index 000000000000..89b18189d43b --- /dev/null +++ b/drivers/gpu/drm/nouveau/include/nvif/clc37b.h @@ -0,0 +1,11 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef __NVIF_CLC37B_H__ +#define __NVIF_CLC37B_H__ + +struct nvc37b_window_imm_channel_dma_v0 { + __u8 version; + __u8 index; + __u8 pad02[6]; + __u64 pushbuf; +}; +#endif diff --git a/drivers/gpu/drm/nouveau/include/nvif/clc37e.h b/drivers/gpu/drm/nouveau/include/nvif/clc37e.h new file mode 100644 index 000000000000..899db9e915ef --- /dev/null +++ b/drivers/gpu/drm/nouveau/include/nvif/clc37e.h @@ -0,0 +1,13 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef __NVIF_CLC37E_H__ +#define __NVIF_CLC37E_H__ + +struct nvc37e_window_channel_dma_v0 { + __u8 version; + __u8 index; + __u8 pad02[6]; + __u64 pushbuf; +}; + +#define NVC37E_WINDOW_CHANNEL_DMA_V0_NTFY_UEVENT 0x00 +#endif diff --git a/drivers/gpu/drm/nouveau/include/nvif/device.h b/drivers/gpu/drm/nouveau/include/nvif/device.h index 6edb6266857e..ef839bd1d37e 100644 --- a/drivers/gpu/drm/nouveau/include/nvif/device.h +++ b/drivers/gpu/drm/nouveau/include/nvif/device.h @@ -4,10 +4,18 @@ #include <nvif/object.h> #include <nvif/cl0080.h> +#include <nvif/user.h> struct nvif_device { struct nvif_object object; struct nv_device_info_v0 info; + + struct nvif_fifo_runlist { + u64 engines; + } *runlist; + int runlists; + + struct nvif_user user; }; int nvif_device_init(struct nvif_object *, u32 handle, s32 oclass, void *, u32, @@ -67,6 +75,5 @@ u64 nvif_device_time(struct nvif_device *); #include <engine/fifo.h> #include <engine/gr.h> -#define nvxx_fifo(a) nvxx_device(a)->fifo #define nvxx_gr(a) nvxx_device(a)->gr #endif diff --git a/drivers/gpu/drm/nouveau/include/nvif/disp.h b/drivers/gpu/drm/nouveau/include/nvif/disp.h new file mode 100644 index 000000000000..7c0eda375c01 --- /dev/null +++ b/drivers/gpu/drm/nouveau/include/nvif/disp.h @@ -0,0 +1,12 @@ +#ifndef __NVIF_DISP_H__ +#define __NVIF_DISP_H__ +#include <nvif/object.h> +struct nvif_device; + +struct nvif_disp { + struct nvif_object object; +}; + +int nvif_disp_ctor(struct nvif_device *, s32 oclass, struct nvif_disp *); +void nvif_disp_dtor(struct nvif_disp *); +#endif diff --git a/drivers/gpu/drm/nouveau/include/nvif/fifo.h b/drivers/gpu/drm/nouveau/include/nvif/fifo.h new file mode 100644 index 000000000000..e9468c9f9abf --- /dev/null +++ b/drivers/gpu/drm/nouveau/include/nvif/fifo.h @@ -0,0 +1,18 @@ +#ifndef __NVIF_FIFO_H__ +#define __NVIF_FIFO_H__ +#include <nvif/device.h> + +/* Returns mask of runlists that support a NV_DEVICE_INFO_ENGINE_* type. */ +u64 nvif_fifo_runlist(struct nvif_device *, u64 engine); + +/* CE-supporting runlists (excluding GRCE, if others exist). */ +static inline u64 +nvif_fifo_runlist_ce(struct nvif_device *device) +{ + u64 runmgr = nvif_fifo_runlist(device, NV_DEVICE_INFO_ENGINE_GR); + u64 runmce = nvif_fifo_runlist(device, NV_DEVICE_INFO_ENGINE_CE); + if (runmce && !(runmce &= ~runmgr)) + runmce = runmgr; + return runmce; +} +#endif diff --git a/drivers/gpu/drm/nouveau/include/nvif/mem.h b/drivers/gpu/drm/nouveau/include/nvif/mem.h index b542fe38398e..80ee4ab0f016 100644 --- a/drivers/gpu/drm/nouveau/include/nvif/mem.h +++ b/drivers/gpu/drm/nouveau/include/nvif/mem.h @@ -15,4 +15,6 @@ int nvif_mem_init_type(struct nvif_mmu *mmu, s32 oclass, int type, u8 page, int nvif_mem_init(struct nvif_mmu *mmu, s32 oclass, u8 type, u8 page, u64 size, void *argv, u32 argc, struct nvif_mem *); void nvif_mem_fini(struct nvif_mem *); + +int nvif_mem_init_map(struct nvif_mmu *, u8 type, u64 size, struct nvif_mem *); #endif diff --git a/drivers/gpu/drm/nouveau/include/nvif/mmu.h b/drivers/gpu/drm/nouveau/include/nvif/mmu.h index c8cd5b5b0688..747ecf67e403 100644 --- a/drivers/gpu/drm/nouveau/include/nvif/mmu.h +++ b/drivers/gpu/drm/nouveau/include/nvif/mmu.h @@ -8,6 +8,7 @@ struct nvif_mmu { u8 heap_nr; u8 type_nr; u16 kind_nr; + s32 mem; struct { u64 size; diff --git a/drivers/gpu/drm/nouveau/include/nvif/object.h b/drivers/gpu/drm/nouveau/include/nvif/object.h index a2d5244ff2b7..20754d9e6883 100644 --- a/drivers/gpu/drm/nouveau/include/nvif/object.h +++ b/drivers/gpu/drm/nouveau/include/nvif/object.h @@ -99,6 +99,22 @@ struct nvif_mclass { ret; \ }) +#define nvif_sclass(o,m,u) ({ \ + const typeof(m[0]) *_mclass = (m); \ + s32 _oclass = (u); \ + int _cid; \ + if (_oclass) { \ + for (_cid = 0; _mclass[_cid].oclass; _cid++) { \ + if (_mclass[_cid].oclass == _oclass) \ + break; \ + } \ + _cid = _mclass[_cid].oclass ? _cid : -ENOSYS; \ + } else { \ + _cid = nvif_mclass((o), _mclass); \ + } \ + _cid; \ +}) + /*XXX*/ #include <core/object.h> #define nvxx_object(a) ({ \ diff --git a/drivers/gpu/drm/nouveau/include/nvif/user.h b/drivers/gpu/drm/nouveau/include/nvif/user.h new file mode 100644 index 000000000000..03c11826b693 --- /dev/null +++ b/drivers/gpu/drm/nouveau/include/nvif/user.h @@ -0,0 +1,19 @@ +#ifndef __NVIF_USER_H__ +#define __NVIF_USER_H__ +#include <nvif/object.h> +struct nvif_device; + +struct nvif_user { + const struct nvif_user_func *func; + struct nvif_object object; +}; + +struct nvif_user_func { + void (*doorbell)(struct nvif_user *, u32 token); +}; + +int nvif_user_init(struct nvif_device *); +void nvif_user_fini(struct nvif_device *); + +extern const struct nvif_user_func nvif_userc361; +#endif diff --git a/drivers/gpu/drm/nouveau/include/nvkm/core/device.h b/drivers/gpu/drm/nouveau/include/nvkm/core/device.h index 560265b15ec2..d83d834b7452 100644 --- a/drivers/gpu/drm/nouveau/include/nvkm/core/device.h +++ b/drivers/gpu/drm/nouveau/include/nvkm/core/device.h @@ -22,6 +22,7 @@ enum nvkm_devidx { NVKM_SUBDEV_LTC, NVKM_SUBDEV_MMU, NVKM_SUBDEV_BAR, + NVKM_SUBDEV_FAULT, NVKM_SUBDEV_PMU, NVKM_SUBDEV_VOLT, NVKM_SUBDEV_ICCSENSE, @@ -37,7 +38,10 @@ enum nvkm_devidx { NVKM_ENGINE_CE3, NVKM_ENGINE_CE4, NVKM_ENGINE_CE5, - NVKM_ENGINE_CE_LAST = NVKM_ENGINE_CE5, + NVKM_ENGINE_CE6, + NVKM_ENGINE_CE7, + NVKM_ENGINE_CE8, + NVKM_ENGINE_CE_LAST = NVKM_ENGINE_CE8, NVKM_ENGINE_CIPHER, NVKM_ENGINE_DISP, @@ -109,6 +113,7 @@ struct nvkm_device { NV_E0 = 0xe0, GM100 = 0x110, GP100 = 0x130, + GV100 = 0x140, } card_type; u32 chipset; u8 chiprev; @@ -123,6 +128,7 @@ struct nvkm_device { struct nvkm_bus *bus; struct nvkm_clk *clk; struct nvkm_devinit *devinit; + struct nvkm_fault *fault; struct nvkm_fb *fb; struct nvkm_fuse *fuse; struct nvkm_gpio *gpio; @@ -143,7 +149,7 @@ struct nvkm_device { struct nvkm_volt *volt; struct nvkm_engine *bsp; - struct nvkm_engine *ce[6]; + struct nvkm_engine *ce[9]; struct nvkm_engine *cipher; struct nvkm_disp *disp; struct nvkm_dma *dma; @@ -194,6 +200,7 @@ struct nvkm_device_chip { int (*bus )(struct nvkm_device *, int idx, struct nvkm_bus **); int (*clk )(struct nvkm_device *, int idx, struct nvkm_clk **); int (*devinit )(struct nvkm_device *, int idx, struct nvkm_devinit **); + int (*fault )(struct nvkm_device *, int idx, struct nvkm_fault **); int (*fb )(struct nvkm_device *, int idx, struct nvkm_fb **); int (*fuse )(struct nvkm_device *, int idx, struct nvkm_fuse **); int (*gpio )(struct nvkm_device *, int idx, struct nvkm_gpio **); @@ -214,7 +221,7 @@ struct nvkm_device_chip { int (*volt )(struct nvkm_device *, int idx, struct nvkm_volt **); int (*bsp )(struct nvkm_device *, int idx, struct nvkm_engine **); - int (*ce[6] )(struct nvkm_device *, int idx, struct nvkm_engine **); + int (*ce[9] )(struct nvkm_device *, int idx, struct nvkm_engine **); int (*cipher )(struct nvkm_device *, int idx, struct nvkm_engine **); int (*disp )(struct nvkm_device *, int idx, struct nvkm_disp **); int (*dma )(struct nvkm_device *, int idx, struct nvkm_dma **); diff --git a/drivers/gpu/drm/nouveau/include/nvkm/core/engine.h b/drivers/gpu/drm/nouveau/include/nvkm/core/engine.h index ebf8473a39fe..8a2be5b635e2 100644 --- a/drivers/gpu/drm/nouveau/include/nvkm/core/engine.h +++ b/drivers/gpu/drm/nouveau/include/nvkm/core/engine.h @@ -18,6 +18,7 @@ struct nvkm_engine_func { void *(*dtor)(struct nvkm_engine *); void (*preinit)(struct nvkm_engine *); int (*oneinit)(struct nvkm_engine *); + int (*info)(struct nvkm_engine *, u64 mthd, u64 *data); int (*init)(struct nvkm_engine *); int (*fini)(struct nvkm_engine *, bool suspend); void (*intr)(struct nvkm_engine *); diff --git a/drivers/gpu/drm/nouveau/include/nvkm/core/subdev.h b/drivers/gpu/drm/nouveau/include/nvkm/core/subdev.h index 63df2290177f..85a0777c2ce4 100644 --- a/drivers/gpu/drm/nouveau/include/nvkm/core/subdev.h +++ b/drivers/gpu/drm/nouveau/include/nvkm/core/subdev.h @@ -17,6 +17,7 @@ struct nvkm_subdev_func { void *(*dtor)(struct nvkm_subdev *); int (*preinit)(struct nvkm_subdev *); int (*oneinit)(struct nvkm_subdev *); + int (*info)(struct nvkm_subdev *, u64 mthd, u64 *data); int (*init)(struct nvkm_subdev *); int (*fini)(struct nvkm_subdev *, bool suspend); void (*intr)(struct nvkm_subdev *); @@ -29,6 +30,7 @@ void nvkm_subdev_del(struct nvkm_subdev **); int nvkm_subdev_preinit(struct nvkm_subdev *); int nvkm_subdev_init(struct nvkm_subdev *); int nvkm_subdev_fini(struct nvkm_subdev *, bool suspend); +int nvkm_subdev_info(struct nvkm_subdev *, u64, u64 *); void nvkm_subdev_intr(struct nvkm_subdev *); /* subdev logging */ diff --git a/drivers/gpu/drm/nouveau/include/nvkm/engine/ce.h b/drivers/gpu/drm/nouveau/include/nvkm/engine/ce.h index 553245994450..fc295e1faa19 100644 --- a/drivers/gpu/drm/nouveau/include/nvkm/engine/ce.h +++ b/drivers/gpu/drm/nouveau/include/nvkm/engine/ce.h @@ -10,4 +10,5 @@ int gm107_ce_new(struct nvkm_device *, int, struct nvkm_engine **); int gm200_ce_new(struct nvkm_device *, int, struct nvkm_engine **); int gp100_ce_new(struct nvkm_device *, int, struct nvkm_engine **); int gp102_ce_new(struct nvkm_device *, int, struct nvkm_engine **); +int gv100_ce_new(struct nvkm_device *, int, struct nvkm_engine **); #endif diff --git a/drivers/gpu/drm/nouveau/include/nvkm/engine/disp.h b/drivers/gpu/drm/nouveau/include/nvkm/engine/disp.h index e83193d3ccab..ef7dc0844d26 100644 --- a/drivers/gpu/drm/nouveau/include/nvkm/engine/disp.h +++ b/drivers/gpu/drm/nouveau/include/nvkm/engine/disp.h @@ -35,4 +35,5 @@ int gm107_disp_new(struct nvkm_device *, int, struct nvkm_disp **); int gm200_disp_new(struct nvkm_device *, int, struct nvkm_disp **); int gp100_disp_new(struct nvkm_device *, int, struct nvkm_disp **); int gp102_disp_new(struct nvkm_device *, int, struct nvkm_disp **); +int gv100_disp_new(struct nvkm_device *, int, struct nvkm_disp **); #endif diff --git a/drivers/gpu/drm/nouveau/include/nvkm/engine/dma.h b/drivers/gpu/drm/nouveau/include/nvkm/engine/dma.h index 0f9c1c702ed6..f0c1b2c8c78c 100644 --- a/drivers/gpu/drm/nouveau/include/nvkm/engine/dma.h +++ b/drivers/gpu/drm/nouveau/include/nvkm/engine/dma.h @@ -27,4 +27,5 @@ int nv04_dma_new(struct nvkm_device *, int, struct nvkm_dma **); int nv50_dma_new(struct nvkm_device *, int, struct nvkm_dma **); int gf100_dma_new(struct nvkm_device *, int, struct nvkm_dma **); int gf119_dma_new(struct nvkm_device *, int, struct nvkm_dma **); +int gv100_dma_new(struct nvkm_device *, int, struct nvkm_dma **); #endif diff --git a/drivers/gpu/drm/nouveau/include/nvkm/engine/fifo.h b/drivers/gpu/drm/nouveau/include/nvkm/engine/fifo.h index c17b3a9bf8fb..7e39fbed2519 100644 --- a/drivers/gpu/drm/nouveau/include/nvkm/engine/fifo.h +++ b/drivers/gpu/drm/nouveau/include/nvkm/engine/fifo.h @@ -4,6 +4,7 @@ #include <core/engine.h> #include <core/object.h> #include <core/event.h> +struct nvkm_fault_data; #define NVKM_FIFO_CHID_NR 4096 @@ -45,6 +46,7 @@ struct nvkm_fifo { struct nvkm_event kevent; /* channel killed */ }; +void nvkm_fifo_fault(struct nvkm_fifo *, struct nvkm_fault_data *); void nvkm_fifo_pause(struct nvkm_fifo *, unsigned long *); void nvkm_fifo_start(struct nvkm_fifo *, unsigned long *); @@ -71,4 +73,5 @@ int gm200_fifo_new(struct nvkm_device *, int, struct nvkm_fifo **); int gm20b_fifo_new(struct nvkm_device *, int, struct nvkm_fifo **); int gp100_fifo_new(struct nvkm_device *, int, struct nvkm_fifo **); int gp10b_fifo_new(struct nvkm_device *, int, struct nvkm_fifo **); +int gv100_fifo_new(struct nvkm_device *, int, struct nvkm_fifo **); #endif diff --git a/drivers/gpu/drm/nouveau/include/nvkm/engine/gr.h b/drivers/gpu/drm/nouveau/include/nvkm/engine/gr.h index fb18f105fc43..ba1518ff8b66 100644 --- a/drivers/gpu/drm/nouveau/include/nvkm/engine/gr.h +++ b/drivers/gpu/drm/nouveau/include/nvkm/engine/gr.h @@ -45,6 +45,8 @@ int gm200_gr_new(struct nvkm_device *, int, struct nvkm_gr **); int gm20b_gr_new(struct nvkm_device *, int, struct nvkm_gr **); int gp100_gr_new(struct nvkm_device *, int, struct nvkm_gr **); int gp102_gr_new(struct nvkm_device *, int, struct nvkm_gr **); +int gp104_gr_new(struct nvkm_device *, int, struct nvkm_gr **); int gp107_gr_new(struct nvkm_device *, int, struct nvkm_gr **); int gp10b_gr_new(struct nvkm_device *, int, struct nvkm_gr **); +int gv100_gr_new(struct nvkm_device *, int, struct nvkm_gr **); #endif diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/bios/dp.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/bios/dp.h index df34b41838d6..512e25a41803 100644 --- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/bios/dp.h +++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/bios/dp.h @@ -1,6 +1,10 @@ /* SPDX-License-Identifier: GPL-2.0 */ #ifndef __NVBIOS_DP_H__ #define __NVBIOS_DP_H__ + +u16 +nvbios_dp_table(struct nvkm_bios *bios, u8 *ver, u8 *hdr, u8 *cnt, u8 *len); + struct nvbios_dpout { u16 type; u16 mask; diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/devinit.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/devinit.h index 40558064d589..486e7635c29d 100644 --- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/devinit.h +++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/devinit.h @@ -30,4 +30,5 @@ int mcp89_devinit_new(struct nvkm_device *, int, struct nvkm_devinit **); int gf100_devinit_new(struct nvkm_device *, int, struct nvkm_devinit **); int gm107_devinit_new(struct nvkm_device *, int, struct nvkm_devinit **); int gm200_devinit_new(struct nvkm_device *, int, struct nvkm_devinit **); +int gv100_devinit_new(struct nvkm_device *, int, struct nvkm_devinit **); #endif diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/fault.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/fault.h new file mode 100644 index 000000000000..5a77498fe6a0 --- /dev/null +++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/fault.h @@ -0,0 +1,33 @@ +#ifndef __NVKM_FAULT_H__ +#define __NVKM_FAULT_H__ +#include <core/subdev.h> +#include <core/notify.h> + +struct nvkm_fault { + const struct nvkm_fault_func *func; + struct nvkm_subdev subdev; + + struct nvkm_fault_buffer *buffer[2]; + int buffer_nr; + + struct nvkm_event event; + + struct nvkm_notify nrpfb; +}; + +struct nvkm_fault_data { + u64 addr; + u64 inst; + u64 time; + u8 engine; + u8 valid; + u8 gpc; + u8 hub; + u8 access; + u8 client; + u8 reason; +}; + +int gp100_fault_new(struct nvkm_device *, int, struct nvkm_fault **); +int gv100_fault_new(struct nvkm_device *, int, struct nvkm_fault **); +#endif diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/fb.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/fb.h index 92be0e5269c6..96ccc624ee81 100644 --- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/fb.h +++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/fb.h @@ -83,6 +83,7 @@ int gm20b_fb_new(struct nvkm_device *, int, struct nvkm_fb **); int gp100_fb_new(struct nvkm_device *, int, struct nvkm_fb **); int gp102_fb_new(struct nvkm_device *, int, struct nvkm_fb **); int gp10b_fb_new(struct nvkm_device *, int, struct nvkm_fb **); +int gv100_fb_new(struct nvkm_device *, int, struct nvkm_fb **); #include <subdev/bios.h> #include <subdev/bios/ramcfg.h> diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/ltc.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/ltc.h index 95b611554d53..9db5f8293198 100644 --- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/ltc.h +++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/ltc.h @@ -21,12 +21,14 @@ struct nvkm_ltc { int zbc_max; u32 zbc_color[NVKM_LTC_MAX_ZBC_CNT][4]; u32 zbc_depth[NVKM_LTC_MAX_ZBC_CNT]; + u32 zbc_stencil[NVKM_LTC_MAX_ZBC_CNT]; }; void nvkm_ltc_tags_clear(struct nvkm_device *, u32 first, u32 count); int nvkm_ltc_zbc_color_get(struct nvkm_ltc *, int index, const u32[4]); int nvkm_ltc_zbc_depth_get(struct nvkm_ltc *, int index, const u32); +int nvkm_ltc_zbc_stencil_get(struct nvkm_ltc *, int index, const u32); void nvkm_ltc_invalidate(struct nvkm_ltc *); void nvkm_ltc_flush(struct nvkm_ltc *); @@ -37,4 +39,5 @@ int gk20a_ltc_new(struct nvkm_device *, int, struct nvkm_ltc **); int gm107_ltc_new(struct nvkm_device *, int, struct nvkm_ltc **); int gm200_ltc_new(struct nvkm_device *, int, struct nvkm_ltc **); int gp100_ltc_new(struct nvkm_device *, int, struct nvkm_ltc **); +int gp102_ltc_new(struct nvkm_device *, int, struct nvkm_ltc **); #endif diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/mmu.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/mmu.h index baab93398e54..688595545e21 100644 --- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/mmu.h +++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/mmu.h @@ -129,4 +129,5 @@ int gm200_mmu_new(struct nvkm_device *, int, struct nvkm_mmu **); int gm20b_mmu_new(struct nvkm_device *, int, struct nvkm_mmu **); int gp100_mmu_new(struct nvkm_device *, int, struct nvkm_mmu **); int gp10b_mmu_new(struct nvkm_device *, int, struct nvkm_mmu **); +int gv100_mmu_new(struct nvkm_device *, int, struct nvkm_mmu **); #endif diff --git a/drivers/gpu/drm/nouveau/nouveau_abi16.c b/drivers/gpu/drm/nouveau/nouveau_abi16.c index ece650a0c5f9..e2211bb2cf79 100644 --- a/drivers/gpu/drm/nouveau/nouveau_abi16.c +++ b/drivers/gpu/drm/nouveau/nouveau_abi16.c @@ -23,6 +23,7 @@ #include <nvif/client.h> #include <nvif/driver.h> +#include <nvif/fifo.h> #include <nvif/ioctl.h> #include <nvif/class.h> #include <nvif/cl0002.h> @@ -102,6 +103,7 @@ nouveau_abi16_swclass(struct nouveau_drm *drm) case NV_DEVICE_INFO_V0_KEPLER: case NV_DEVICE_INFO_V0_MAXWELL: case NV_DEVICE_INFO_V0_PASCAL: + case NV_DEVICE_INFO_V0_VOLTA: return NVIF_CLASS_SW_GF100; } @@ -256,6 +258,7 @@ nouveau_abi16_ioctl_channel_alloc(ABI16_IOCTL_ARGS) struct nouveau_abi16 *abi16 = nouveau_abi16_get(file_priv); struct nouveau_abi16_chan *chan; struct nvif_device *device; + u64 engine; int ret; if (unlikely(!abi16)) @@ -268,25 +271,26 @@ nouveau_abi16_ioctl_channel_alloc(ABI16_IOCTL_ARGS) /* hack to allow channel engine type specification on kepler */ if (device->info.family >= NV_DEVICE_INFO_V0_KEPLER) { - if (init->fb_ctxdma_handle != ~0) - init->fb_ctxdma_handle = NVA06F_V0_ENGINE_GR; - else { - init->fb_ctxdma_handle = 0; -#define _(A,B) if (init->tt_ctxdma_handle & (A)) init->fb_ctxdma_handle |= (B) - _(0x01, NVA06F_V0_ENGINE_GR); - _(0x02, NVA06F_V0_ENGINE_MSPDEC); - _(0x04, NVA06F_V0_ENGINE_MSPPP); - _(0x08, NVA06F_V0_ENGINE_MSVLD); - _(0x10, NVA06F_V0_ENGINE_CE0); - _(0x20, NVA06F_V0_ENGINE_CE1); - _(0x40, NVA06F_V0_ENGINE_MSENC); -#undef _ + if (init->fb_ctxdma_handle == ~0) { + switch (init->tt_ctxdma_handle) { + case 0x01: engine = NV_DEVICE_INFO_ENGINE_GR ; break; + case 0x02: engine = NV_DEVICE_INFO_ENGINE_MSPDEC; break; + case 0x04: engine = NV_DEVICE_INFO_ENGINE_MSPPP ; break; + case 0x08: engine = NV_DEVICE_INFO_ENGINE_MSVLD ; break; + case 0x30: engine = NV_DEVICE_INFO_ENGINE_CE ; break; + default: + return nouveau_abi16_put(abi16, -ENOSYS); + } + } else { + engine = NV_DEVICE_INFO_ENGINE_GR; } - /* allow flips to be executed if this is a graphics channel */ + if (engine != NV_DEVICE_INFO_ENGINE_CE) + engine = nvif_fifo_runlist(device, engine); + else + engine = nvif_fifo_runlist_ce(device); + init->fb_ctxdma_handle = engine; init->tt_ctxdma_handle = 0; - if (init->fb_ctxdma_handle == NVA06F_V0_ENGINE_GR) - init->tt_ctxdma_handle = 1; } if (init->fb_ctxdma_handle == ~0 || init->tt_ctxdma_handle == ~0) diff --git a/drivers/gpu/drm/nouveau/nouveau_acpi.c b/drivers/gpu/drm/nouveau/nouveau_acpi.c index 5ffcb6683776..ffb195850314 100644 --- a/drivers/gpu/drm/nouveau/nouveau_acpi.c +++ b/drivers/gpu/drm/nouveau/nouveau_acpi.c @@ -193,7 +193,7 @@ static int nouveau_dsm_power_state(enum vga_switcheroo_client_id id, return nouveau_dsm_set_discrete_state(nouveau_dsm_priv.dhandle, state); } -static int nouveau_dsm_get_client_id(struct pci_dev *pdev) +static enum vga_switcheroo_client_id nouveau_dsm_get_client_id(struct pci_dev *pdev) { /* easy option one - intel vendor ID means Integrated */ if (pdev->vendor == PCI_VENDOR_ID_INTEL) diff --git a/drivers/gpu/drm/nouveau/nouveau_bo.c b/drivers/gpu/drm/nouveau/nouveau_bo.c index 6f402c4f2bdd..7214022dfb91 100644 --- a/drivers/gpu/drm/nouveau/nouveau_bo.c +++ b/drivers/gpu/drm/nouveau/nouveau_bo.c @@ -214,7 +214,6 @@ nouveau_bo_new(struct nouveau_cli *cli, u64 size, int align, INIT_LIST_HEAD(&nvbo->entry); INIT_LIST_HEAD(&nvbo->vma_list); nvbo->bo.bdev = &drm->ttm.bdev; - nvbo->cli = cli; /* This is confusing, and doesn't actually mean we want an uncached * mapping, but is what NOUVEAU_GEM_DOMAIN_COHERENT gets translated @@ -1142,6 +1141,8 @@ nouveau_bo_move_init(struct nouveau_drm *drm) struct ttm_mem_reg *, struct ttm_mem_reg *); int (*init)(struct nouveau_channel *, u32 handle); } _methods[] = { + { "COPY", 4, 0xc3b5, nve0_bo_move_copy, nve0_bo_move_init }, + { "GRCE", 0, 0xc3b5, nve0_bo_move_copy, nvc0_bo_move_init }, { "COPY", 4, 0xc1b5, nve0_bo_move_copy, nve0_bo_move_init }, { "GRCE", 0, 0xc1b5, nve0_bo_move_copy, nvc0_bo_move_init }, { "COPY", 4, 0xc0b5, nve0_bo_move_copy, nve0_bo_move_init }, diff --git a/drivers/gpu/drm/nouveau/nouveau_bo.h b/drivers/gpu/drm/nouveau/nouveau_bo.h index be8e00b49cde..73c48440d4d7 100644 --- a/drivers/gpu/drm/nouveau/nouveau_bo.h +++ b/drivers/gpu/drm/nouveau/nouveau_bo.h @@ -26,8 +26,6 @@ struct nouveau_bo { struct list_head vma_list; - struct nouveau_cli *cli; - unsigned contig:1; unsigned page:5; unsigned kind:8; diff --git a/drivers/gpu/drm/nouveau/nouveau_chan.c b/drivers/gpu/drm/nouveau/nouveau_chan.c index af1116655910..92d3115f96b5 100644 --- a/drivers/gpu/drm/nouveau/nouveau_chan.c +++ b/drivers/gpu/drm/nouveau/nouveau_chan.c @@ -163,12 +163,15 @@ nouveau_channel_prep(struct nouveau_drm *drm, struct nvif_device *device, return ret; } + chan->push.addr = chan->push.vma->addr; + + if (device->info.family >= NV_DEVICE_INFO_V0_FERMI) + return 0; + args.target = NV_DMA_V0_TARGET_VM; args.access = NV_DMA_V0_ACCESS_VM; args.start = 0; args.limit = cli->vmm.vmm.limit - 1; - - chan->push.addr = chan->push.vma->addr; } else if (chan->push.buffer->bo.mem.mem_type == TTM_PL_VRAM) { if (device->info.family == NV_DEVICE_INFO_V0_TNT) { @@ -214,10 +217,11 @@ nouveau_channel_prep(struct nouveau_drm *drm, struct nvif_device *device, static int nouveau_channel_ind(struct nouveau_drm *drm, struct nvif_device *device, - u32 engine, struct nouveau_channel **pchan) + u64 runlist, struct nouveau_channel **pchan) { struct nouveau_cli *cli = (void *)device->object.client; - static const u16 oclasses[] = { PASCAL_CHANNEL_GPFIFO_A, + static const u16 oclasses[] = { VOLTA_CHANNEL_GPFIFO_A, + PASCAL_CHANNEL_GPFIFO_A, MAXWELL_CHANNEL_GPFIFO_A, KEPLER_CHANNEL_GPFIFO_B, KEPLER_CHANNEL_GPFIFO_A, @@ -245,9 +249,9 @@ nouveau_channel_ind(struct nouveau_drm *drm, struct nvif_device *device, do { if (oclass[0] >= KEPLER_CHANNEL_GPFIFO_A) { args.kepler.version = 0; - args.kepler.engines = engine; args.kepler.ilength = 0x02000; args.kepler.ioffset = 0x10000 + chan->push.addr; + args.kepler.runlist = runlist; args.kepler.vmm = nvif_handle(&cli->vmm.vmm.object); size = sizeof(args.kepler); } else @@ -474,3 +478,28 @@ done: cli->base.super = super; return ret; } + +int +nouveau_channels_init(struct nouveau_drm *drm) +{ + struct { + struct nv_device_info_v1 m; + struct { + struct nv_device_info_v1_data channels; + } v; + } args = { + .m.version = 1, + .m.count = sizeof(args.v) / sizeof(args.v.channels), + .v.channels.mthd = NV_DEVICE_FIFO_CHANNELS, + }; + struct nvif_object *device = &drm->client.device.object; + int ret; + + ret = nvif_object_mthd(device, NV_DEVICE_V0_INFO, &args, sizeof(args)); + if (ret || args.v.channels.mthd == NV_DEVICE_INFO_INVALID) + return -ENODEV; + + drm->chan.nr = args.v.channels.data; + drm->chan.context_base = dma_fence_context_alloc(drm->chan.nr); + return 0; +} diff --git a/drivers/gpu/drm/nouveau/nouveau_chan.h b/drivers/gpu/drm/nouveau/nouveau_chan.h index 14607c16a2bd..64454c2ebd90 100644 --- a/drivers/gpu/drm/nouveau/nouveau_chan.h +++ b/drivers/gpu/drm/nouveau/nouveau_chan.h @@ -45,6 +45,7 @@ struct nouveau_channel { atomic_t killed; }; +int nouveau_channels_init(struct nouveau_drm *); int nouveau_channel_new(struct nouveau_drm *, struct nvif_device *, u32 arg0, u32 arg1, struct nouveau_channel **); diff --git a/drivers/gpu/drm/nouveau/nouveau_connector.c b/drivers/gpu/drm/nouveau/nouveau_connector.c index 6ed9cb053dfa..7b557c354307 100644 --- a/drivers/gpu/drm/nouveau/nouveau_connector.c +++ b/drivers/gpu/drm/nouveau/nouveau_connector.c @@ -151,7 +151,7 @@ nouveau_conn_atomic_set_property(struct drm_connector *connector, /* ... except prior to G80, where the code * doesn't support such things. */ - if (disp->disp.oclass < NV50_DISP) + if (disp->disp.object.oclass < NV50_DISP) return -EINVAL; break; default: @@ -260,7 +260,7 @@ nouveau_conn_reset(struct drm_connector *connector) asyc->procamp.color_vibrance = 150; asyc->procamp.vibrant_hue = 90; - if (nouveau_display(connector->dev)->disp.oclass < NV50_DISP) { + if (nouveau_display(connector->dev)->disp.object.oclass < NV50_DISP) { switch (connector->connector_type) { case DRM_MODE_CONNECTOR_LVDS: /* See note in nouveau_conn_atomic_set_property(). */ @@ -314,7 +314,7 @@ nouveau_conn_attach_properties(struct drm_connector *connector) case DRM_MODE_CONNECTOR_TV: break; case DRM_MODE_CONNECTOR_VGA: - if (disp->disp.oclass < NV50_DISP) + if (disp->disp.object.oclass < NV50_DISP) break; /* Can only scale on DFPs. */ /* Fall-through. */ default: @@ -1005,7 +1005,7 @@ get_tmds_link_bandwidth(struct drm_connector *connector, bool hdmi) return 112000; } -static int +static enum drm_mode_status nouveau_connector_mode_valid(struct drm_connector *connector, struct drm_display_mode *mode) { @@ -1321,7 +1321,7 @@ nouveau_connector_create(struct drm_device *dev, int index) } /* HDMI 3D support */ - if ((disp->disp.oclass >= G82_DISP) + if ((disp->disp.object.oclass >= G82_DISP) && ((type == DRM_MODE_CONNECTOR_DisplayPort) || (type == DRM_MODE_CONNECTOR_eDP) || (type == DRM_MODE_CONNECTOR_HDMIA))) @@ -1343,7 +1343,7 @@ nouveau_connector_create(struct drm_device *dev, int index) case DCB_CONNECTOR_LVDS_SPWG: case DCB_CONNECTOR_eDP: /* see note in nouveau_connector_set_property() */ - if (disp->disp.oclass < NV50_DISP) { + if (disp->disp.object.oclass < NV50_DISP) { nv_connector->scaling_mode = DRM_MODE_SCALE_FULLSCREEN; break; } @@ -1366,8 +1366,8 @@ nouveau_connector_create(struct drm_device *dev, int index) break; } - ret = nvif_notify_init(&disp->disp, nouveau_connector_hotplug, true, - NV04_DISP_NTFY_CONN, + ret = nvif_notify_init(&disp->disp.object, nouveau_connector_hotplug, + true, NV04_DISP_NTFY_CONN, &(struct nvif_notify_conn_req_v0) { .mask = NVIF_NOTIFY_CONN_V0_ANY, .conn = index, diff --git a/drivers/gpu/drm/nouveau/nouveau_display.c b/drivers/gpu/drm/nouveau/nouveau_display.c index 7d0bec8dd03d..774b429142bc 100644 --- a/drivers/gpu/drm/nouveau/nouveau_display.c +++ b/drivers/gpu/drm/nouveau/nouveau_display.c @@ -116,7 +116,7 @@ nouveau_display_scanoutpos_head(struct drm_crtc *crtc, int *vpos, int *hpos, bool ret = false; do { - ret = nvif_mthd(&disp->disp, 0, &args, sizeof(args)); + ret = nvif_mthd(&disp->disp.object, 0, &args, sizeof(args)); if (ret != 0) return false; @@ -175,7 +175,7 @@ nouveau_display_vblank_init(struct drm_device *dev) list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) { struct nouveau_crtc *nv_crtc = nouveau_crtc(crtc); - ret = nvif_notify_init(&disp->disp, + ret = nvif_notify_init(&disp->disp.object, nouveau_display_vblank_handler, false, NV04_DISP_NTFY_VBLANK, &(struct nvif_notify_head_req_v0) { @@ -454,10 +454,10 @@ nouveau_display_create_properties(struct drm_device *dev) struct nouveau_display *disp = nouveau_display(dev); int gen; - if (disp->disp.oclass < NV50_DISP) + if (disp->disp.object.oclass < NV50_DISP) gen = 0; else - if (disp->disp.oclass < GF110_DISP) + if (disp->disp.object.oclass < GF110_DISP) gen = 1; else gen = 2; @@ -533,31 +533,10 @@ nouveau_display_create(struct drm_device *dev) drm_kms_helper_poll_disable(dev); if (nouveau_modeset != 2 && drm->vbios.dcb.entries) { - static const u16 oclass[] = { - GP102_DISP, - GP100_DISP, - GM200_DISP, - GM107_DISP, - GK110_DISP, - GK104_DISP, - GF110_DISP, - GT214_DISP, - GT206_DISP, - GT200_DISP, - G82_DISP, - NV50_DISP, - NV04_DISP, - }; - int i; - - for (i = 0, ret = -ENODEV; ret && i < ARRAY_SIZE(oclass); i++) { - ret = nvif_object_init(&drm->client.device.object, 0, - oclass[i], NULL, 0, &disp->disp); - } - + ret = nvif_disp_ctor(&drm->client.device, 0, &disp->disp); if (ret == 0) { nouveau_display_create_properties(dev); - if (disp->disp.oclass < NV50_DISP) + if (disp->disp.object.oclass < NV50_DISP) ret = nv04_display_create(dev); else ret = nv50_display_create(dev); @@ -611,7 +590,7 @@ nouveau_display_destroy(struct drm_device *dev) if (disp->dtor) disp->dtor(dev); - nvif_object_fini(&disp->disp); + nvif_disp_dtor(&disp->disp); nouveau_drm(dev)->display = NULL; kfree(disp); diff --git a/drivers/gpu/drm/nouveau/nouveau_display.h b/drivers/gpu/drm/nouveau/nouveau_display.h index 270ba56f2756..54aa7c3fa42d 100644 --- a/drivers/gpu/drm/nouveau/nouveau_display.h +++ b/drivers/gpu/drm/nouveau/nouveau_display.h @@ -2,6 +2,7 @@ #ifndef __NOUVEAU_DISPLAY_H__ #define __NOUVEAU_DISPLAY_H__ #include "nouveau_drv.h" +#include <nvif/disp.h> struct nouveau_framebuffer { struct drm_framebuffer base; @@ -38,7 +39,7 @@ struct nouveau_display { int (*init)(struct drm_device *); void (*fini)(struct drm_device *); - struct nvif_object disp; + struct nvif_disp disp; struct drm_property *dithering_mode; struct drm_property *dithering_depth; diff --git a/drivers/gpu/drm/nouveau/nouveau_dma.c b/drivers/gpu/drm/nouveau/nouveau_dma.c index 10e84f6ca2b7..945afd34138e 100644 --- a/drivers/gpu/drm/nouveau/nouveau_dma.c +++ b/drivers/gpu/drm/nouveau/nouveau_dma.c @@ -28,6 +28,8 @@ #include "nouveau_dma.h" #include "nouveau_vmm.h" +#include <nvif/user.h> + void OUT_RINGp(struct nouveau_channel *chan, const void *data, unsigned nr_dwords) { @@ -80,18 +82,11 @@ READ_GET(struct nouveau_channel *chan, uint64_t *prev_get, int *timeout) } void -nv50_dma_push(struct nouveau_channel *chan, struct nouveau_bo *bo, - int delta, int length) +nv50_dma_push(struct nouveau_channel *chan, u64 offset, int length) { - struct nouveau_cli *cli = (void *)chan->user.client; + struct nvif_user *user = &chan->drm->client.device.user; struct nouveau_bo *pb = chan->push.buffer; - struct nouveau_vma *vma; int ip = (chan->dma.ib_put * 2) + chan->dma.ib_base; - u64 offset; - - vma = nouveau_vma_find(bo, &cli->vmm); - BUG_ON(!vma); - offset = vma->addr + delta; BUG_ON(chan->dma.ib_free < 1); @@ -105,6 +100,8 @@ nv50_dma_push(struct nouveau_channel *chan, struct nouveau_bo *bo, nouveau_bo_rd32(pb, 0); nvif_wr32(&chan->user, 0x8c, chan->dma.ib_put); + if (user->func && user->func->doorbell) + user->func->doorbell(user, chan->chid); chan->dma.ib_free--; } diff --git a/drivers/gpu/drm/nouveau/nouveau_dma.h b/drivers/gpu/drm/nouveau/nouveau_dma.h index 74e10b14a7da..fc5e3f41282d 100644 --- a/drivers/gpu/drm/nouveau/nouveau_dma.h +++ b/drivers/gpu/drm/nouveau/nouveau_dma.h @@ -31,8 +31,7 @@ #include "nouveau_chan.h" int nouveau_dma_wait(struct nouveau_channel *, int slots, int size); -void nv50_dma_push(struct nouveau_channel *, struct nouveau_bo *, - int delta, int length); +void nv50_dma_push(struct nouveau_channel *, u64 addr, int length); /* * There's a hw race condition where you can't jump to your PUT offset, @@ -55,7 +54,6 @@ enum { NvSub2D = 3, /* DO NOT CHANGE - hardcoded for kepler gr fifo */ NvSubCopy = 4, /* DO NOT CHANGE - hardcoded for kepler gr fifo */ - FermiSw = 5, /* DO NOT CHANGE (well.. 6/7 will work...) */ }; /* Object handles - for stuff that's doesn't use handle == oclass. */ @@ -151,7 +149,7 @@ FIRE_RING(struct nouveau_channel *chan) chan->accel_done = true; if (chan->dma.ib_max) { - nv50_dma_push(chan, chan->push.buffer, chan->dma.put << 2, + nv50_dma_push(chan, chan->push.addr + (chan->dma.put << 2), (chan->dma.cur - chan->dma.put) << 2); } else { WRITE_PUT(chan->dma.cur); diff --git a/drivers/gpu/drm/nouveau/nouveau_drm.c b/drivers/gpu/drm/nouveau/nouveau_drm.c index bbbf353682e1..775443c9af94 100644 --- a/drivers/gpu/drm/nouveau/nouveau_drm.c +++ b/drivers/gpu/drm/nouveau/nouveau_drm.c @@ -38,6 +38,8 @@ #include <core/tegra.h> #include <nvif/driver.h> +#include <nvif/fifo.h> +#include <nvif/user.h> #include <nvif/class.h> #include <nvif/cl0002.h> @@ -112,24 +114,22 @@ nouveau_name(struct drm_device *dev) } static inline bool -nouveau_cli_work_ready(struct dma_fence *fence, bool wait) +nouveau_cli_work_ready(struct dma_fence *fence) { - if (!dma_fence_is_signaled(fence)) { - if (!wait) - return false; - WARN_ON(dma_fence_wait_timeout(fence, false, 2 * HZ) <= 0); - } + if (!dma_fence_is_signaled(fence)) + return false; dma_fence_put(fence); return true; } static void -nouveau_cli_work_flush(struct nouveau_cli *cli, bool wait) +nouveau_cli_work(struct work_struct *w) { + struct nouveau_cli *cli = container_of(w, typeof(*cli), work); struct nouveau_cli_work *work, *wtmp; mutex_lock(&cli->lock); list_for_each_entry_safe(work, wtmp, &cli->worker, head) { - if (!work->fence || nouveau_cli_work_ready(work->fence, wait)) { + if (!work->fence || nouveau_cli_work_ready(work->fence)) { list_del(&work->head); work->func(work); } @@ -158,16 +158,16 @@ nouveau_cli_work_queue(struct nouveau_cli *cli, struct dma_fence *fence, } static void -nouveau_cli_work(struct work_struct *w) -{ - struct nouveau_cli *cli = container_of(w, typeof(*cli), work); - nouveau_cli_work_flush(cli, false); -} - -static void nouveau_cli_fini(struct nouveau_cli *cli) { - nouveau_cli_work_flush(cli, true); + /* All our channels are dead now, which means all the fences they + * own are signalled, and all callback functions have been called. + * + * So, after flushing the workqueue, there should be nothing left. + */ + flush_work(&cli->work); + WARN_ON(!list_empty(&cli->worker)); + usif_client_fini(cli); nouveau_vmm_fini(&cli->vmm); nvif_mmu_fini(&cli->mmu); @@ -307,6 +307,16 @@ nouveau_accel_init(struct nouveau_drm *drm) if (nouveau_noaccel) return; + ret = nouveau_channels_init(drm); + if (ret) + return; + + if (drm->client.device.info.family >= NV_DEVICE_INFO_V0_VOLTA) { + ret = nvif_user_init(device); + if (ret) + return; + } + /* initialise synchronisation routines */ /*XXX: this is crap, but the fence/channel stuff is a little * backwards in some places. this will be fixed. @@ -338,6 +348,7 @@ nouveau_accel_init(struct nouveau_drm *drm) case KEPLER_CHANNEL_GPFIFO_B: case MAXWELL_CHANNEL_GPFIFO_A: case PASCAL_CHANNEL_GPFIFO_A: + case VOLTA_CHANNEL_GPFIFO_A: ret = nvc0_fence_create(drm); break; default: @@ -354,13 +365,12 @@ nouveau_accel_init(struct nouveau_drm *drm) if (device->info.family >= NV_DEVICE_INFO_V0_KEPLER) { ret = nouveau_channel_new(drm, &drm->client.device, - NVA06F_V0_ENGINE_CE0 | - NVA06F_V0_ENGINE_CE1, - 0, &drm->cechan); + nvif_fifo_runlist_ce(device), 0, + &drm->cechan); if (ret) NV_ERROR(drm, "failed to create ce channel, %d\n", ret); - arg0 = NVA06F_V0_ENGINE_GR; + arg0 = nvif_fifo_runlist(device, NV_DEVICE_INFO_ENGINE_GR); arg1 = 1; } else if (device->info.chipset >= 0xa3 && @@ -386,38 +396,36 @@ nouveau_accel_init(struct nouveau_drm *drm) return; } - ret = nvif_object_init(&drm->channel->user, NVDRM_NVSW, - nouveau_abi16_swclass(drm), NULL, 0, &drm->nvsw); - if (ret == 0) { - ret = RING_SPACE(drm->channel, 2); + if (device->info.family < NV_DEVICE_INFO_V0_TESLA) { + ret = nvif_object_init(&drm->channel->user, NVDRM_NVSW, + nouveau_abi16_swclass(drm), NULL, 0, + &drm->nvsw); if (ret == 0) { - if (device->info.family < NV_DEVICE_INFO_V0_FERMI) { + ret = RING_SPACE(drm->channel, 2); + if (ret == 0) { BEGIN_NV04(drm->channel, NvSubSw, 0, 1); - OUT_RING (drm->channel, NVDRM_NVSW); - } else - if (device->info.family < NV_DEVICE_INFO_V0_KEPLER) { - BEGIN_NVC0(drm->channel, FermiSw, 0, 1); - OUT_RING (drm->channel, 0x001f0000); + OUT_RING (drm->channel, drm->nvsw.handle); + } + + ret = nvif_notify_init(&drm->nvsw, + nouveau_flip_complete, + false, NV04_NVSW_NTFY_UEVENT, + NULL, 0, 0, &drm->flip); + if (ret == 0) + ret = nvif_notify_get(&drm->flip); + if (ret) { + nouveau_accel_fini(drm); + return; } } - ret = nvif_notify_init(&drm->nvsw, nouveau_flip_complete, - false, NV04_NVSW_NTFY_UEVENT, - NULL, 0, 0, &drm->flip); - if (ret == 0) - ret = nvif_notify_get(&drm->flip); if (ret) { + NV_ERROR(drm, "failed to allocate sw class, %d\n", ret); nouveau_accel_fini(drm); return; } } - if (ret) { - NV_ERROR(drm, "failed to allocate software object, %d\n", ret); - nouveau_accel_fini(drm); - return; - } - if (device->info.family < NV_DEVICE_INFO_V0_FERMI) { ret = nvkm_gpuobj_new(nvxx_device(&drm->client.device), 32, 0, false, NULL, &drm->notify); diff --git a/drivers/gpu/drm/nouveau/nouveau_drv.h b/drivers/gpu/drm/nouveau/nouveau_drv.h index 881b44b89a01..6e1acaec3400 100644 --- a/drivers/gpu/drm/nouveau/nouveau_drv.h +++ b/drivers/gpu/drm/nouveau/nouveau_drv.h @@ -170,6 +170,12 @@ struct nouveau_drm { /* synchronisation */ void *fence; + /* Global channel management. */ + struct { + int nr; + u64 context_base; + } chan; + /* context for accelerated drm-internal operations */ struct nouveau_channel *cechan; struct nouveau_channel *channel; diff --git a/drivers/gpu/drm/nouveau/nouveau_encoder.h b/drivers/gpu/drm/nouveau/nouveau_encoder.h index e28d966946a1..3517f920bf89 100644 --- a/drivers/gpu/drm/nouveau/nouveau_encoder.h +++ b/drivers/gpu/drm/nouveau/nouveau_encoder.h @@ -32,6 +32,7 @@ #include <drm/drm_encoder_slave.h> #include <drm/drm_dp_mst_helper.h> #include "dispnv04/disp.h" +struct nv50_head_atom; #define NV_DPMS_CLEARED 0x80 @@ -68,7 +69,7 @@ struct nouveau_encoder { void (*enc_save)(struct drm_encoder *encoder); void (*enc_restore)(struct drm_encoder *encoder); void (*update)(struct nouveau_encoder *, u8 head, - struct drm_display_mode *, u8 proto, u8 depth); + struct nv50_head_atom *, u8 proto, u8 depth); }; struct nouveau_encoder * diff --git a/drivers/gpu/drm/nouveau/nouveau_fence.c b/drivers/gpu/drm/nouveau/nouveau_fence.c index 503fa94dc06d..412d49bc6e56 100644 --- a/drivers/gpu/drm/nouveau/nouveau_fence.c +++ b/drivers/gpu/drm/nouveau/nouveau_fence.c @@ -74,15 +74,14 @@ nouveau_fence_signal(struct nouveau_fence *fence) } static struct nouveau_fence * -nouveau_local_fence(struct dma_fence *fence, struct nouveau_drm *drm) { - struct nouveau_fence_priv *priv = (void*)drm->fence; - +nouveau_local_fence(struct dma_fence *fence, struct nouveau_drm *drm) +{ if (fence->ops != &nouveau_fence_ops_legacy && fence->ops != &nouveau_fence_ops_uevent) return NULL; - if (fence->context < priv->context_base || - fence->context >= priv->context_base + priv->contexts) + if (fence->context < drm->chan.context_base || + fence->context >= drm->chan.context_base + drm->chan.nr) return NULL; return from_fence(fence); @@ -176,7 +175,7 @@ nouveau_fence_context_new(struct nouveau_channel *chan, struct nouveau_fence_cha INIT_LIST_HEAD(&fctx->flip); INIT_LIST_HEAD(&fctx->pending); spin_lock_init(&fctx->lock); - fctx->context = priv->context_base + chan->chid; + fctx->context = chan->drm->chan.context_base + chan->chid; if (chan == chan->drm->cechan) strcpy(fctx->name, "copy engine channel"); diff --git a/drivers/gpu/drm/nouveau/nouveau_fence.h b/drivers/gpu/drm/nouveau/nouveau_fence.h index 5bd8d30d1657..b999e6058046 100644 --- a/drivers/gpu/drm/nouveau/nouveau_fence.h +++ b/drivers/gpu/drm/nouveau/nouveau_fence.h @@ -55,8 +55,6 @@ struct nouveau_fence_priv { int (*context_new)(struct nouveau_channel *); void (*context_del)(struct nouveau_channel *); - u32 contexts; - u64 context_base; bool uevent; }; diff --git a/drivers/gpu/drm/nouveau/nouveau_gem.c b/drivers/gpu/drm/nouveau/nouveau_gem.c index e72a7e37eb0a..300daee74209 100644 --- a/drivers/gpu/drm/nouveau/nouveau_gem.c +++ b/drivers/gpu/drm/nouveau/nouveau_gem.c @@ -99,6 +99,7 @@ struct nouveau_gem_object_unmap { static void nouveau_gem_object_delete(struct nouveau_vma *vma) { + nouveau_fence_unref(&vma->fence); nouveau_vma_del(&vma); } @@ -114,25 +115,12 @@ nouveau_gem_object_delete_work(struct nouveau_cli_work *w) static void nouveau_gem_object_unmap(struct nouveau_bo *nvbo, struct nouveau_vma *vma) { - const bool mapped = nvbo->bo.mem.mem_type != TTM_PL_SYSTEM; - struct reservation_object *resv = nvbo->bo.resv; - struct reservation_object_list *fobj; + struct dma_fence *fence = vma->fence ? &vma->fence->base : NULL; struct nouveau_gem_object_unmap *work; - struct dma_fence *fence = NULL; - - fobj = reservation_object_get_list(resv); list_del_init(&vma->head); - if (fobj && fobj->shared_count > 1) - ttm_bo_wait(&nvbo->bo, false, false); - else if (fobj && fobj->shared_count == 1) - fence = rcu_dereference_protected(fobj->shared[0], - reservation_object_held(resv)); - else - fence = reservation_object_get_excl(nvbo->bo.resv); - - if (!fence || !mapped) { + if (!fence) { nouveau_gem_object_delete(vma); return; } @@ -344,9 +332,20 @@ validate_fini_no_ticket(struct validate_op *op, struct nouveau_fence *fence, nvbo = list_entry(op->list.next, struct nouveau_bo, entry); b = &pbbo[nvbo->pbbo_index]; - if (likely(fence)) + if (likely(fence)) { + struct nouveau_drm *drm = nouveau_bdev(nvbo->bo.bdev); + struct nouveau_vma *vma; + nouveau_bo_fence(nvbo, fence, !!b->write_domains); + if (drm->client.vmm.vmm.object.oclass >= NVIF_CLASS_VMM_NV50) { + vma = (void *)(unsigned long)b->user_priv; + nouveau_fence_unref(&vma->fence); + dma_fence_get(&fence->base); + vma->fence = fence; + } + } + if (unlikely(nvbo->validate_mapped)) { ttm_bo_kunmap(&nvbo->kmap); nvbo->validate_mapped = false; @@ -432,7 +431,20 @@ retry: } } - b->user_priv = (uint64_t)(unsigned long)nvbo; + if (cli->vmm.vmm.object.oclass >= NVIF_CLASS_VMM_NV50) { + struct nouveau_vmm *vmm = &cli->vmm; + struct nouveau_vma *vma = nouveau_vma_find(nvbo, vmm); + if (!vma) { + NV_PRINTK(err, cli, "vma not found!\n"); + ret = -EINVAL; + break; + } + + b->user_priv = (uint64_t)(unsigned long)vma; + } else { + b->user_priv = (uint64_t)(unsigned long)nvbo; + } + nvbo->reserved_by = file_priv; nvbo->pbbo_index = i; if ((b->valid_domains & NOUVEAU_GEM_DOMAIN_VRAM) && @@ -763,10 +775,10 @@ nouveau_gem_ioctl_pushbuf(struct drm_device *dev, void *data, } for (i = 0; i < req->nr_push; i++) { - struct nouveau_bo *nvbo = (void *)(unsigned long) + struct nouveau_vma *vma = (void *)(unsigned long) bo[push[i].bo_index].user_priv; - nv50_dma_push(chan, nvbo, push[i].offset, + nv50_dma_push(chan, vma->addr + push[i].offset, push[i].length); } } else diff --git a/drivers/gpu/drm/nouveau/nouveau_hwmon.c b/drivers/gpu/drm/nouveau/nouveau_hwmon.c index 7c965648df80..44178b4c3599 100644 --- a/drivers/gpu/drm/nouveau/nouveau_hwmon.c +++ b/drivers/gpu/drm/nouveau/nouveau_hwmon.c @@ -327,7 +327,7 @@ nouveau_temp_is_visible(const void *data, u32 attr, int channel) struct nouveau_drm *drm = nouveau_drm((struct drm_device *)data); struct nvkm_therm *therm = nvxx_therm(&drm->client.device); - if (therm && therm->attr_get && nvkm_therm_temp_get(therm) < 0) + if (!therm || !therm->attr_get || nvkm_therm_temp_get(therm) < 0) return 0; switch (attr) { @@ -351,8 +351,8 @@ nouveau_pwm_is_visible(const void *data, u32 attr, int channel) struct nouveau_drm *drm = nouveau_drm((struct drm_device *)data); struct nvkm_therm *therm = nvxx_therm(&drm->client.device); - if (therm && therm->attr_get && therm->fan_get && - therm->fan_get(therm) < 0) + if (!therm || !therm->attr_get || !therm->fan_get || + therm->fan_get(therm) < 0) return 0; switch (attr) { @@ -707,13 +707,20 @@ nouveau_hwmon_init(struct drm_device *dev) { #if defined(CONFIG_HWMON) || (defined(MODULE) && defined(CONFIG_HWMON_MODULE)) struct nouveau_drm *drm = nouveau_drm(dev); + struct nvkm_iccsense *iccsense = nvxx_iccsense(&drm->client.device); struct nvkm_therm *therm = nvxx_therm(&drm->client.device); + struct nvkm_volt *volt = nvxx_volt(&drm->client.device); const struct attribute_group *special_groups[N_ATTR_GROUPS]; struct nouveau_hwmon *hwmon; struct device *hwmon_dev; int ret = 0; int i = 0; + if (!iccsense && !therm && !volt) { + NV_DEBUG(drm, "Skipping hwmon registration\n"); + return 0; + } + hwmon = drm->hwmon = kzalloc(sizeof(*hwmon), GFP_KERNEL); if (!hwmon) return -ENOMEM; @@ -749,6 +756,9 @@ nouveau_hwmon_fini(struct drm_device *dev) #if defined(CONFIG_HWMON) || (defined(MODULE) && defined(CONFIG_HWMON_MODULE)) struct nouveau_hwmon *hwmon = nouveau_hwmon(dev); + if (!hwmon) + return; + if (hwmon->hwmon) hwmon_device_unregister(hwmon->hwmon); diff --git a/drivers/gpu/drm/nouveau/nouveau_ttm.c b/drivers/gpu/drm/nouveau/nouveau_ttm.c index dff51a0ee028..8c093ca4222e 100644 --- a/drivers/gpu/drm/nouveau/nouveau_ttm.c +++ b/drivers/gpu/drm/nouveau/nouveau_ttm.c @@ -63,7 +63,7 @@ nouveau_vram_manager_new(struct ttm_mem_type_manager *man, struct ttm_mem_reg *reg) { struct nouveau_bo *nvbo = nouveau_bo(bo); - struct nouveau_drm *drm = nvbo->cli->drm; + struct nouveau_drm *drm = nouveau_bdev(bo->bdev); struct nouveau_mem *mem; int ret; @@ -103,7 +103,7 @@ nouveau_gart_manager_new(struct ttm_mem_type_manager *man, struct ttm_mem_reg *reg) { struct nouveau_bo *nvbo = nouveau_bo(bo); - struct nouveau_drm *drm = nvbo->cli->drm; + struct nouveau_drm *drm = nouveau_bdev(bo->bdev); struct nouveau_mem *mem; int ret; @@ -131,7 +131,7 @@ nv04_gart_manager_new(struct ttm_mem_type_manager *man, struct ttm_mem_reg *reg) { struct nouveau_bo *nvbo = nouveau_bo(bo); - struct nouveau_drm *drm = nvbo->cli->drm; + struct nouveau_drm *drm = nouveau_bdev(bo->bdev); struct nouveau_mem *mem; int ret; diff --git a/drivers/gpu/drm/nouveau/nouveau_vmm.c b/drivers/gpu/drm/nouveau/nouveau_vmm.c index f5371d96b003..2032c3e4f6e5 100644 --- a/drivers/gpu/drm/nouveau/nouveau_vmm.c +++ b/drivers/gpu/drm/nouveau/nouveau_vmm.c @@ -92,6 +92,7 @@ nouveau_vma_new(struct nouveau_bo *nvbo, struct nouveau_vmm *vmm, vma->refs = 1; vma->addr = ~0ULL; vma->mem = NULL; + vma->fence = NULL; list_add_tail(&vma->head, &nvbo->vma_list); if (nvbo->bo.mem.mem_type != TTM_PL_SYSTEM && diff --git a/drivers/gpu/drm/nouveau/nouveau_vmm.h b/drivers/gpu/drm/nouveau/nouveau_vmm.h index 5c31f43678d3..7e3b118cf7c4 100644 --- a/drivers/gpu/drm/nouveau/nouveau_vmm.h +++ b/drivers/gpu/drm/nouveau/nouveau_vmm.h @@ -11,6 +11,8 @@ struct nouveau_vma { u64 addr; struct nouveau_mem *mem; + + struct nouveau_fence *fence; }; struct nouveau_vma *nouveau_vma_find(struct nouveau_bo *, struct nouveau_vmm *); diff --git a/drivers/gpu/drm/nouveau/nv04_fence.c b/drivers/gpu/drm/nouveau/nv04_fence.c index fa8f2375c398..c41e82be4893 100644 --- a/drivers/gpu/drm/nouveau/nv04_fence.c +++ b/drivers/gpu/drm/nouveau/nv04_fence.c @@ -109,7 +109,5 @@ nv04_fence_create(struct nouveau_drm *drm) priv->base.dtor = nv04_fence_destroy; priv->base.context_new = nv04_fence_context_new; priv->base.context_del = nv04_fence_context_del; - priv->base.contexts = 15; - priv->base.context_base = dma_fence_context_alloc(priv->base.contexts); return 0; } diff --git a/drivers/gpu/drm/nouveau/nv10_fence.c b/drivers/gpu/drm/nouveau/nv10_fence.c index 2998bde29211..4476b712dc84 100644 --- a/drivers/gpu/drm/nouveau/nv10_fence.c +++ b/drivers/gpu/drm/nouveau/nv10_fence.c @@ -103,8 +103,6 @@ nv10_fence_create(struct nouveau_drm *drm) priv->base.dtor = nv10_fence_destroy; priv->base.context_new = nv10_fence_context_new; priv->base.context_del = nv10_fence_context_del; - priv->base.contexts = 31; - priv->base.context_base = dma_fence_context_alloc(priv->base.contexts); spin_lock_init(&priv->lock); return 0; } diff --git a/drivers/gpu/drm/nouveau/nv17_fence.c b/drivers/gpu/drm/nouveau/nv17_fence.c index 6477b7069e14..5d613d43b84d 100644 --- a/drivers/gpu/drm/nouveau/nv17_fence.c +++ b/drivers/gpu/drm/nouveau/nv17_fence.c @@ -125,8 +125,6 @@ nv17_fence_create(struct nouveau_drm *drm) priv->base.resume = nv17_fence_resume; priv->base.context_new = nv17_fence_context_new; priv->base.context_del = nv10_fence_context_del; - priv->base.contexts = 31; - priv->base.context_base = dma_fence_context_alloc(priv->base.contexts); spin_lock_init(&priv->lock); ret = nouveau_bo_new(&drm->client, 4096, 0x1000, TTM_PL_FLAG_VRAM, diff --git a/drivers/gpu/drm/nouveau/nv50_display.c b/drivers/gpu/drm/nouveau/nv50_display.c deleted file mode 100644 index 8bd739cfd00d..000000000000 --- a/drivers/gpu/drm/nouveau/nv50_display.c +++ /dev/null @@ -1,4559 +0,0 @@ -/* - * Copyright 2011 Red Hat Inc. - * - * Permission is hereby granted, free of charge, to any person obtaining a - * copy of this software and associated documentation files (the "Software"), - * to deal in the Software without restriction, including without limitation - * the rights to use, copy, modify, merge, publish, distribute, sublicense, - * and/or sell copies of the Software, and to permit persons to whom the - * Software is furnished to do so, subject to the following conditions: - * - * The above copyright notice and this permission notice shall be included in - * all copies or substantial portions of the Software. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL - * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR - * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, - * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR - * OTHER DEALINGS IN THE SOFTWARE. - * - * Authors: Ben Skeggs - */ - -#include <linux/dma-mapping.h> -#include <linux/hdmi.h> - -#include <drm/drmP.h> -#include <drm/drm_atomic.h> -#include <drm/drm_atomic_helper.h> -#include <drm/drm_crtc_helper.h> -#include <drm/drm_dp_helper.h> -#include <drm/drm_fb_helper.h> -#include <drm/drm_plane_helper.h> -#include <drm/drm_edid.h> - -#include <nvif/class.h> -#include <nvif/cl0002.h> -#include <nvif/cl5070.h> -#include <nvif/cl507a.h> -#include <nvif/cl507b.h> -#include <nvif/cl507c.h> -#include <nvif/cl507d.h> -#include <nvif/cl507e.h> -#include <nvif/event.h> - -#include "nouveau_drv.h" -#include "nouveau_dma.h" -#include "nouveau_gem.h" -#include "nouveau_connector.h" -#include "nouveau_encoder.h" -#include "nouveau_crtc.h" -#include "nouveau_fence.h" -#include "nouveau_fbcon.h" -#include "nv50_display.h" - -#define EVO_DMA_NR 9 - -#define EVO_MASTER (0x00) -#define EVO_FLIP(c) (0x01 + (c)) -#define EVO_OVLY(c) (0x05 + (c)) -#define EVO_OIMM(c) (0x09 + (c)) -#define EVO_CURS(c) (0x0d + (c)) - -/* offsets in shared sync bo of various structures */ -#define EVO_SYNC(c, o) ((c) * 0x0100 + (o)) -#define EVO_MAST_NTFY EVO_SYNC( 0, 0x00) -#define EVO_FLIP_SEM0(c) EVO_SYNC((c) + 1, 0x00) -#define EVO_FLIP_SEM1(c) EVO_SYNC((c) + 1, 0x10) -#define EVO_FLIP_NTFY0(c) EVO_SYNC((c) + 1, 0x20) -#define EVO_FLIP_NTFY1(c) EVO_SYNC((c) + 1, 0x30) - -/****************************************************************************** - * Atomic state - *****************************************************************************/ -#define nv50_atom(p) container_of((p), struct nv50_atom, state) - -struct nv50_atom { - struct drm_atomic_state state; - - struct list_head outp; - bool lock_core; - bool flush_disable; -}; - -struct nv50_outp_atom { - struct list_head head; - - struct drm_encoder *encoder; - bool flush_disable; - - union { - struct { - bool ctrl:1; - }; - u8 mask; - } clr; - - union { - struct { - bool ctrl:1; - }; - u8 mask; - } set; -}; - -#define nv50_head_atom(p) container_of((p), struct nv50_head_atom, state) - -struct nv50_head_atom { - struct drm_crtc_state state; - - struct { - u16 iW; - u16 iH; - u16 oW; - u16 oH; - } view; - - struct nv50_head_mode { - bool interlace; - u32 clock; - struct { - u16 active; - u16 synce; - u16 blanke; - u16 blanks; - } h; - struct { - u32 active; - u16 synce; - u16 blanke; - u16 blanks; - u16 blank2s; - u16 blank2e; - u16 blankus; - } v; - } mode; - - struct { - bool visible; - u32 handle; - u64 offset:40; - u8 mode:4; - } lut; - - struct { - bool visible; - u32 handle; - u64 offset:40; - u8 format; - u8 kind:7; - u8 layout:1; - u8 block:4; - u32 pitch:20; - u16 x; - u16 y; - u16 w; - u16 h; - } core; - - struct { - bool visible; - u32 handle; - u64 offset:40; - u8 layout:1; - u8 format:1; - } curs; - - struct { - u8 depth; - u8 cpp; - u16 x; - u16 y; - u16 w; - u16 h; - } base; - - struct { - u8 cpp; - } ovly; - - struct { - bool enable:1; - u8 bits:2; - u8 mode:4; - } dither; - - struct { - struct { - u16 cos:12; - u16 sin:12; - } sat; - } procamp; - - union { - struct { - bool ilut:1; - bool core:1; - bool curs:1; - }; - u8 mask; - } clr; - - union { - struct { - bool ilut:1; - bool core:1; - bool curs:1; - bool view:1; - bool mode:1; - bool base:1; - bool ovly:1; - bool dither:1; - bool procamp:1; - }; - u16 mask; - } set; -}; - -static inline struct nv50_head_atom * -nv50_head_atom_get(struct drm_atomic_state *state, struct drm_crtc *crtc) -{ - struct drm_crtc_state *statec = drm_atomic_get_crtc_state(state, crtc); - if (IS_ERR(statec)) - return (void *)statec; - return nv50_head_atom(statec); -} - -#define nv50_wndw_atom(p) container_of((p), struct nv50_wndw_atom, state) - -struct nv50_wndw_atom { - struct drm_plane_state state; - u8 interval; - - struct { - u32 handle; - u16 offset:12; - bool awaken:1; - } ntfy; - - struct { - u32 handle; - u16 offset:12; - u32 acquire; - u32 release; - } sema; - - struct { - u8 enable:2; - } lut; - - struct { - u8 mode:2; - u8 interval:4; - - u8 format; - u8 kind:7; - u8 layout:1; - u8 block:4; - u32 pitch:20; - u16 w; - u16 h; - - u32 handle; - u64 offset; - } image; - - struct { - u16 x; - u16 y; - } point; - - union { - struct { - bool ntfy:1; - bool sema:1; - bool image:1; - }; - u8 mask; - } clr; - - union { - struct { - bool ntfy:1; - bool sema:1; - bool image:1; - bool lut:1; - bool point:1; - }; - u8 mask; - } set; -}; - -/****************************************************************************** - * EVO channel - *****************************************************************************/ - -struct nv50_chan { - struct nvif_object user; - struct nvif_device *device; -}; - -static int -nv50_chan_create(struct nvif_device *device, struct nvif_object *disp, - const s32 *oclass, u8 head, void *data, u32 size, - struct nv50_chan *chan) -{ - struct nvif_sclass *sclass; - int ret, i, n; - - chan->device = device; - - ret = n = nvif_object_sclass_get(disp, &sclass); - if (ret < 0) - return ret; - - while (oclass[0]) { - for (i = 0; i < n; i++) { - if (sclass[i].oclass == oclass[0]) { - ret = nvif_object_init(disp, 0, oclass[0], - data, size, &chan->user); - if (ret == 0) - nvif_object_map(&chan->user, NULL, 0); - nvif_object_sclass_put(&sclass); - return ret; - } - } - oclass++; - } - - nvif_object_sclass_put(&sclass); - return -ENOSYS; -} - -static void -nv50_chan_destroy(struct nv50_chan *chan) -{ - nvif_object_fini(&chan->user); -} - -/****************************************************************************** - * PIO EVO channel - *****************************************************************************/ - -struct nv50_pioc { - struct nv50_chan base; -}; - -static void -nv50_pioc_destroy(struct nv50_pioc *pioc) -{ - nv50_chan_destroy(&pioc->base); -} - -static int -nv50_pioc_create(struct nvif_device *device, struct nvif_object *disp, - const s32 *oclass, u8 head, void *data, u32 size, - struct nv50_pioc *pioc) -{ - return nv50_chan_create(device, disp, oclass, head, data, size, - &pioc->base); -} - -/****************************************************************************** - * Overlay Immediate - *****************************************************************************/ - -struct nv50_oimm { - struct nv50_pioc base; -}; - -static int -nv50_oimm_create(struct nvif_device *device, struct nvif_object *disp, - int head, struct nv50_oimm *oimm) -{ - struct nv50_disp_cursor_v0 args = { - .head = head, - }; - static const s32 oclass[] = { - GK104_DISP_OVERLAY, - GF110_DISP_OVERLAY, - GT214_DISP_OVERLAY, - G82_DISP_OVERLAY, - NV50_DISP_OVERLAY, - 0 - }; - - return nv50_pioc_create(device, disp, oclass, head, &args, sizeof(args), - &oimm->base); -} - -/****************************************************************************** - * DMA EVO channel - *****************************************************************************/ - -struct nv50_dmac_ctxdma { - struct list_head head; - struct nvif_object object; -}; - -struct nv50_dmac { - struct nv50_chan base; - dma_addr_t handle; - u32 *ptr; - - struct nvif_object sync; - struct nvif_object vram; - struct list_head ctxdma; - - /* Protects against concurrent pushbuf access to this channel, lock is - * grabbed by evo_wait (if the pushbuf reservation is successful) and - * dropped again by evo_kick. */ - struct mutex lock; -}; - -static void -nv50_dmac_ctxdma_del(struct nv50_dmac_ctxdma *ctxdma) -{ - nvif_object_fini(&ctxdma->object); - list_del(&ctxdma->head); - kfree(ctxdma); -} - -static struct nv50_dmac_ctxdma * -nv50_dmac_ctxdma_new(struct nv50_dmac *dmac, struct nouveau_framebuffer *fb) -{ - struct nouveau_drm *drm = nouveau_drm(fb->base.dev); - struct nv50_dmac_ctxdma *ctxdma; - const u8 kind = fb->nvbo->kind; - const u32 handle = 0xfb000000 | kind; - struct { - struct nv_dma_v0 base; - union { - struct nv50_dma_v0 nv50; - struct gf100_dma_v0 gf100; - struct gf119_dma_v0 gf119; - }; - } args = {}; - u32 argc = sizeof(args.base); - int ret; - - list_for_each_entry(ctxdma, &dmac->ctxdma, head) { - if (ctxdma->object.handle == handle) - return ctxdma; - } - - if (!(ctxdma = kzalloc(sizeof(*ctxdma), GFP_KERNEL))) - return ERR_PTR(-ENOMEM); - list_add(&ctxdma->head, &dmac->ctxdma); - - args.base.target = NV_DMA_V0_TARGET_VRAM; - args.base.access = NV_DMA_V0_ACCESS_RDWR; - args.base.start = 0; - args.base.limit = drm->client.device.info.ram_user - 1; - - if (drm->client.device.info.chipset < 0x80) { - args.nv50.part = NV50_DMA_V0_PART_256; - argc += sizeof(args.nv50); - } else - if (drm->client.device.info.chipset < 0xc0) { - args.nv50.part = NV50_DMA_V0_PART_256; - args.nv50.kind = kind; - argc += sizeof(args.nv50); - } else - if (drm->client.device.info.chipset < 0xd0) { - args.gf100.kind = kind; - argc += sizeof(args.gf100); - } else { - args.gf119.page = GF119_DMA_V0_PAGE_LP; - args.gf119.kind = kind; - argc += sizeof(args.gf119); - } - - ret = nvif_object_init(&dmac->base.user, handle, NV_DMA_IN_MEMORY, - &args, argc, &ctxdma->object); - if (ret) { - nv50_dmac_ctxdma_del(ctxdma); - return ERR_PTR(ret); - } - - return ctxdma; -} - -static void -nv50_dmac_destroy(struct nv50_dmac *dmac, struct nvif_object *disp) -{ - struct nvif_device *device = dmac->base.device; - struct nv50_dmac_ctxdma *ctxdma, *ctxtmp; - - list_for_each_entry_safe(ctxdma, ctxtmp, &dmac->ctxdma, head) { - nv50_dmac_ctxdma_del(ctxdma); - } - - nvif_object_fini(&dmac->vram); - nvif_object_fini(&dmac->sync); - - nv50_chan_destroy(&dmac->base); - - if (dmac->ptr) { - struct device *dev = nvxx_device(device)->dev; - dma_free_coherent(dev, PAGE_SIZE, dmac->ptr, dmac->handle); - } -} - -static int -nv50_dmac_create(struct nvif_device *device, struct nvif_object *disp, - const s32 *oclass, u8 head, void *data, u32 size, u64 syncbuf, - struct nv50_dmac *dmac) -{ - struct nv50_disp_core_channel_dma_v0 *args = data; - struct nvif_object pushbuf; - int ret; - - mutex_init(&dmac->lock); - INIT_LIST_HEAD(&dmac->ctxdma); - - dmac->ptr = dma_alloc_coherent(nvxx_device(device)->dev, PAGE_SIZE, - &dmac->handle, GFP_KERNEL); - if (!dmac->ptr) - return -ENOMEM; - - ret = nvif_object_init(&device->object, 0, NV_DMA_FROM_MEMORY, - &(struct nv_dma_v0) { - .target = NV_DMA_V0_TARGET_PCI_US, - .access = NV_DMA_V0_ACCESS_RD, - .start = dmac->handle + 0x0000, - .limit = dmac->handle + 0x0fff, - }, sizeof(struct nv_dma_v0), &pushbuf); - if (ret) - return ret; - - args->pushbuf = nvif_handle(&pushbuf); - - ret = nv50_chan_create(device, disp, oclass, head, data, size, - &dmac->base); - nvif_object_fini(&pushbuf); - if (ret) - return ret; - - ret = nvif_object_init(&dmac->base.user, 0xf0000000, NV_DMA_IN_MEMORY, - &(struct nv_dma_v0) { - .target = NV_DMA_V0_TARGET_VRAM, - .access = NV_DMA_V0_ACCESS_RDWR, - .start = syncbuf + 0x0000, - .limit = syncbuf + 0x0fff, - }, sizeof(struct nv_dma_v0), - &dmac->sync); - if (ret) - return ret; - - ret = nvif_object_init(&dmac->base.user, 0xf0000001, NV_DMA_IN_MEMORY, - &(struct nv_dma_v0) { - .target = NV_DMA_V0_TARGET_VRAM, - .access = NV_DMA_V0_ACCESS_RDWR, - .start = 0, - .limit = device->info.ram_user - 1, - }, sizeof(struct nv_dma_v0), - &dmac->vram); - if (ret) - return ret; - - return ret; -} - -/****************************************************************************** - * Core - *****************************************************************************/ - -struct nv50_mast { - struct nv50_dmac base; -}; - -static int -nv50_core_create(struct nvif_device *device, struct nvif_object *disp, - u64 syncbuf, struct nv50_mast *core) -{ - struct nv50_disp_core_channel_dma_v0 args = { - .pushbuf = 0xb0007d00, - }; - static const s32 oclass[] = { - GP102_DISP_CORE_CHANNEL_DMA, - GP100_DISP_CORE_CHANNEL_DMA, - GM200_DISP_CORE_CHANNEL_DMA, - GM107_DISP_CORE_CHANNEL_DMA, - GK110_DISP_CORE_CHANNEL_DMA, - GK104_DISP_CORE_CHANNEL_DMA, - GF110_DISP_CORE_CHANNEL_DMA, - GT214_DISP_CORE_CHANNEL_DMA, - GT206_DISP_CORE_CHANNEL_DMA, - GT200_DISP_CORE_CHANNEL_DMA, - G82_DISP_CORE_CHANNEL_DMA, - NV50_DISP_CORE_CHANNEL_DMA, - 0 - }; - - return nv50_dmac_create(device, disp, oclass, 0, &args, sizeof(args), - syncbuf, &core->base); -} - -/****************************************************************************** - * Base - *****************************************************************************/ - -struct nv50_sync { - struct nv50_dmac base; - u32 addr; - u32 data; -}; - -static int -nv50_base_create(struct nvif_device *device, struct nvif_object *disp, - int head, u64 syncbuf, struct nv50_sync *base) -{ - struct nv50_disp_base_channel_dma_v0 args = { - .pushbuf = 0xb0007c00 | head, - .head = head, - }; - static const s32 oclass[] = { - GK110_DISP_BASE_CHANNEL_DMA, - GK104_DISP_BASE_CHANNEL_DMA, - GF110_DISP_BASE_CHANNEL_DMA, - GT214_DISP_BASE_CHANNEL_DMA, - GT200_DISP_BASE_CHANNEL_DMA, - G82_DISP_BASE_CHANNEL_DMA, - NV50_DISP_BASE_CHANNEL_DMA, - 0 - }; - - return nv50_dmac_create(device, disp, oclass, head, &args, sizeof(args), - syncbuf, &base->base); -} - -/****************************************************************************** - * Overlay - *****************************************************************************/ - -struct nv50_ovly { - struct nv50_dmac base; -}; - -static int -nv50_ovly_create(struct nvif_device *device, struct nvif_object *disp, - int head, u64 syncbuf, struct nv50_ovly *ovly) -{ - struct nv50_disp_overlay_channel_dma_v0 args = { - .pushbuf = 0xb0007e00 | head, - .head = head, - }; - static const s32 oclass[] = { - GK104_DISP_OVERLAY_CONTROL_DMA, - GF110_DISP_OVERLAY_CONTROL_DMA, - GT214_DISP_OVERLAY_CHANNEL_DMA, - GT200_DISP_OVERLAY_CHANNEL_DMA, - G82_DISP_OVERLAY_CHANNEL_DMA, - NV50_DISP_OVERLAY_CHANNEL_DMA, - 0 - }; - - return nv50_dmac_create(device, disp, oclass, head, &args, sizeof(args), - syncbuf, &ovly->base); -} - -struct nv50_head { - struct nouveau_crtc base; - struct { - struct nouveau_bo *nvbo[2]; - int next; - } lut; - struct nv50_ovly ovly; - struct nv50_oimm oimm; -}; - -#define nv50_head(c) ((struct nv50_head *)nouveau_crtc(c)) -#define nv50_ovly(c) (&nv50_head(c)->ovly) -#define nv50_oimm(c) (&nv50_head(c)->oimm) -#define nv50_chan(c) (&(c)->base.base) -#define nv50_vers(c) nv50_chan(c)->user.oclass - -struct nv50_disp { - struct nvif_object *disp; - struct nv50_mast mast; - - struct nouveau_bo *sync; - - struct mutex mutex; -}; - -static struct nv50_disp * -nv50_disp(struct drm_device *dev) -{ - return nouveau_display(dev)->priv; -} - -#define nv50_mast(d) (&nv50_disp(d)->mast) - -/****************************************************************************** - * EVO channel helpers - *****************************************************************************/ -static u32 * -evo_wait(void *evoc, int nr) -{ - struct nv50_dmac *dmac = evoc; - struct nvif_device *device = dmac->base.device; - u32 put = nvif_rd32(&dmac->base.user, 0x0000) / 4; - - mutex_lock(&dmac->lock); - if (put + nr >= (PAGE_SIZE / 4) - 8) { - dmac->ptr[put] = 0x20000000; - - nvif_wr32(&dmac->base.user, 0x0000, 0x00000000); - if (nvif_msec(device, 2000, - if (!nvif_rd32(&dmac->base.user, 0x0004)) - break; - ) < 0) { - mutex_unlock(&dmac->lock); - pr_err("nouveau: evo channel stalled\n"); - return NULL; - } - - put = 0; - } - - return dmac->ptr + put; -} - -static void -evo_kick(u32 *push, void *evoc) -{ - struct nv50_dmac *dmac = evoc; - nvif_wr32(&dmac->base.user, 0x0000, (push - dmac->ptr) << 2); - mutex_unlock(&dmac->lock); -} - -#define evo_mthd(p, m, s) do { \ - const u32 _m = (m), _s = (s); \ - if (drm_debug & DRM_UT_KMS) \ - pr_err("%04x %d %s\n", _m, _s, __func__); \ - *((p)++) = ((_s << 18) | _m); \ -} while(0) - -#define evo_data(p, d) do { \ - const u32 _d = (d); \ - if (drm_debug & DRM_UT_KMS) \ - pr_err("\t%08x\n", _d); \ - *((p)++) = _d; \ -} while(0) - -/****************************************************************************** - * Plane - *****************************************************************************/ -#define nv50_wndw(p) container_of((p), struct nv50_wndw, plane) - -struct nv50_wndw { - const struct nv50_wndw_func *func; - struct nv50_dmac *dmac; - - struct drm_plane plane; - - struct nvif_notify notify; - u16 ntfy; - u16 sema; - u32 data; -}; - -struct nv50_wndw_func { - void *(*dtor)(struct nv50_wndw *); - int (*acquire)(struct nv50_wndw *, struct nv50_wndw_atom *asyw, - struct nv50_head_atom *asyh); - void (*release)(struct nv50_wndw *, struct nv50_wndw_atom *asyw, - struct nv50_head_atom *asyh); - void (*prepare)(struct nv50_wndw *, struct nv50_head_atom *asyh, - struct nv50_wndw_atom *asyw); - - void (*sema_set)(struct nv50_wndw *, struct nv50_wndw_atom *); - void (*sema_clr)(struct nv50_wndw *); - void (*ntfy_set)(struct nv50_wndw *, struct nv50_wndw_atom *); - void (*ntfy_clr)(struct nv50_wndw *); - int (*ntfy_wait_begun)(struct nv50_wndw *, struct nv50_wndw_atom *); - void (*image_set)(struct nv50_wndw *, struct nv50_wndw_atom *); - void (*image_clr)(struct nv50_wndw *); - void (*lut)(struct nv50_wndw *, struct nv50_wndw_atom *); - void (*point)(struct nv50_wndw *, struct nv50_wndw_atom *); - - u32 (*update)(struct nv50_wndw *, u32 interlock); -}; - -static int -nv50_wndw_wait_armed(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw) -{ - if (asyw->set.ntfy) - return wndw->func->ntfy_wait_begun(wndw, asyw); - return 0; -} - -static u32 -nv50_wndw_flush_clr(struct nv50_wndw *wndw, u32 interlock, bool flush, - struct nv50_wndw_atom *asyw) -{ - if (asyw->clr.sema && (!asyw->set.sema || flush)) - wndw->func->sema_clr(wndw); - if (asyw->clr.ntfy && (!asyw->set.ntfy || flush)) - wndw->func->ntfy_clr(wndw); - if (asyw->clr.image && (!asyw->set.image || flush)) - wndw->func->image_clr(wndw); - - return flush ? wndw->func->update(wndw, interlock) : 0; -} - -static u32 -nv50_wndw_flush_set(struct nv50_wndw *wndw, u32 interlock, - struct nv50_wndw_atom *asyw) -{ - if (interlock) { - asyw->image.mode = 0; - asyw->image.interval = 1; - } - - if (asyw->set.sema ) wndw->func->sema_set (wndw, asyw); - if (asyw->set.ntfy ) wndw->func->ntfy_set (wndw, asyw); - if (asyw->set.image) wndw->func->image_set(wndw, asyw); - if (asyw->set.lut ) wndw->func->lut (wndw, asyw); - if (asyw->set.point) wndw->func->point (wndw, asyw); - - return wndw->func->update(wndw, interlock); -} - -static void -nv50_wndw_atomic_check_release(struct nv50_wndw *wndw, - struct nv50_wndw_atom *asyw, - struct nv50_head_atom *asyh) -{ - struct nouveau_drm *drm = nouveau_drm(wndw->plane.dev); - NV_ATOMIC(drm, "%s release\n", wndw->plane.name); - wndw->func->release(wndw, asyw, asyh); - asyw->ntfy.handle = 0; - asyw->sema.handle = 0; -} - -static int -nv50_wndw_atomic_check_acquire(struct nv50_wndw *wndw, - struct nv50_wndw_atom *asyw, - struct nv50_head_atom *asyh) -{ - struct nouveau_framebuffer *fb = nouveau_framebuffer(asyw->state.fb); - struct nouveau_drm *drm = nouveau_drm(wndw->plane.dev); - int ret; - - NV_ATOMIC(drm, "%s acquire\n", wndw->plane.name); - - asyw->image.w = fb->base.width; - asyw->image.h = fb->base.height; - asyw->image.kind = fb->nvbo->kind; - - if (asyh->state.pageflip_flags & DRM_MODE_PAGE_FLIP_ASYNC) - asyw->interval = 0; - else - asyw->interval = 1; - - if (asyw->image.kind) { - asyw->image.layout = 0; - if (drm->client.device.info.chipset >= 0xc0) - asyw->image.block = fb->nvbo->mode >> 4; - else - asyw->image.block = fb->nvbo->mode; - asyw->image.pitch = (fb->base.pitches[0] / 4) << 4; - } else { - asyw->image.layout = 1; - asyw->image.block = 0; - asyw->image.pitch = fb->base.pitches[0]; - } - - ret = wndw->func->acquire(wndw, asyw, asyh); - if (ret) - return ret; - - if (asyw->set.image) { - if (!(asyw->image.mode = asyw->interval ? 0 : 1)) - asyw->image.interval = asyw->interval; - else - asyw->image.interval = 0; - } - - return 0; -} - -static int -nv50_wndw_atomic_check(struct drm_plane *plane, struct drm_plane_state *state) -{ - struct nouveau_drm *drm = nouveau_drm(plane->dev); - struct nv50_wndw *wndw = nv50_wndw(plane); - struct nv50_wndw_atom *armw = nv50_wndw_atom(wndw->plane.state); - struct nv50_wndw_atom *asyw = nv50_wndw_atom(state); - struct nv50_head_atom *harm = NULL, *asyh = NULL; - bool varm = false, asyv = false, asym = false; - int ret; - - NV_ATOMIC(drm, "%s atomic_check\n", plane->name); - if (asyw->state.crtc) { - asyh = nv50_head_atom_get(asyw->state.state, asyw->state.crtc); - if (IS_ERR(asyh)) - return PTR_ERR(asyh); - asym = drm_atomic_crtc_needs_modeset(&asyh->state); - asyv = asyh->state.active; - } - - if (armw->state.crtc) { - harm = nv50_head_atom_get(asyw->state.state, armw->state.crtc); - if (IS_ERR(harm)) - return PTR_ERR(harm); - varm = harm->state.crtc->state->active; - } - - if (asyv) { - asyw->point.x = asyw->state.crtc_x; - asyw->point.y = asyw->state.crtc_y; - if (memcmp(&armw->point, &asyw->point, sizeof(asyw->point))) - asyw->set.point = true; - - ret = nv50_wndw_atomic_check_acquire(wndw, asyw, asyh); - if (ret) - return ret; - } else - if (varm) { - nv50_wndw_atomic_check_release(wndw, asyw, harm); - } else { - return 0; - } - - if (!asyv || asym) { - asyw->clr.ntfy = armw->ntfy.handle != 0; - asyw->clr.sema = armw->sema.handle != 0; - if (wndw->func->image_clr) - asyw->clr.image = armw->image.handle != 0; - asyw->set.lut = wndw->func->lut && asyv; - } - - return 0; -} - -static void -nv50_wndw_cleanup_fb(struct drm_plane *plane, struct drm_plane_state *old_state) -{ - struct nouveau_framebuffer *fb = nouveau_framebuffer(old_state->fb); - struct nouveau_drm *drm = nouveau_drm(plane->dev); - - NV_ATOMIC(drm, "%s cleanup: %p\n", plane->name, old_state->fb); - if (!old_state->fb) - return; - - nouveau_bo_unpin(fb->nvbo); -} - -static int -nv50_wndw_prepare_fb(struct drm_plane *plane, struct drm_plane_state *state) -{ - struct nouveau_framebuffer *fb = nouveau_framebuffer(state->fb); - struct nouveau_drm *drm = nouveau_drm(plane->dev); - struct nv50_wndw *wndw = nv50_wndw(plane); - struct nv50_wndw_atom *asyw = nv50_wndw_atom(state); - struct nv50_head_atom *asyh; - struct nv50_dmac_ctxdma *ctxdma; - int ret; - - NV_ATOMIC(drm, "%s prepare: %p\n", plane->name, state->fb); - if (!asyw->state.fb) - return 0; - - ret = nouveau_bo_pin(fb->nvbo, TTM_PL_FLAG_VRAM, true); - if (ret) - return ret; - - ctxdma = nv50_dmac_ctxdma_new(wndw->dmac, fb); - if (IS_ERR(ctxdma)) { - nouveau_bo_unpin(fb->nvbo); - return PTR_ERR(ctxdma); - } - - asyw->state.fence = reservation_object_get_excl_rcu(fb->nvbo->bo.resv); - asyw->image.handle = ctxdma->object.handle; - asyw->image.offset = fb->nvbo->bo.offset; - - if (wndw->func->prepare) { - asyh = nv50_head_atom_get(asyw->state.state, asyw->state.crtc); - if (IS_ERR(asyh)) - return PTR_ERR(asyh); - - wndw->func->prepare(wndw, asyh, asyw); - } - - return 0; -} - -static const struct drm_plane_helper_funcs -nv50_wndw_helper = { - .prepare_fb = nv50_wndw_prepare_fb, - .cleanup_fb = nv50_wndw_cleanup_fb, - .atomic_check = nv50_wndw_atomic_check, -}; - -static void -nv50_wndw_atomic_destroy_state(struct drm_plane *plane, - struct drm_plane_state *state) -{ - struct nv50_wndw_atom *asyw = nv50_wndw_atom(state); - __drm_atomic_helper_plane_destroy_state(&asyw->state); - kfree(asyw); -} - -static struct drm_plane_state * -nv50_wndw_atomic_duplicate_state(struct drm_plane *plane) -{ - struct nv50_wndw_atom *armw = nv50_wndw_atom(plane->state); - struct nv50_wndw_atom *asyw; - if (!(asyw = kmalloc(sizeof(*asyw), GFP_KERNEL))) - return NULL; - __drm_atomic_helper_plane_duplicate_state(plane, &asyw->state); - asyw->interval = 1; - asyw->sema = armw->sema; - asyw->ntfy = armw->ntfy; - asyw->image = armw->image; - asyw->point = armw->point; - asyw->lut = armw->lut; - asyw->clr.mask = 0; - asyw->set.mask = 0; - return &asyw->state; -} - -static void -nv50_wndw_reset(struct drm_plane *plane) -{ - struct nv50_wndw_atom *asyw; - - if (WARN_ON(!(asyw = kzalloc(sizeof(*asyw), GFP_KERNEL)))) - return; - - if (plane->state) - plane->funcs->atomic_destroy_state(plane, plane->state); - plane->state = &asyw->state; - plane->state->plane = plane; - plane->state->rotation = DRM_MODE_ROTATE_0; -} - -static void -nv50_wndw_destroy(struct drm_plane *plane) -{ - struct nv50_wndw *wndw = nv50_wndw(plane); - void *data; - nvif_notify_fini(&wndw->notify); - data = wndw->func->dtor(wndw); - drm_plane_cleanup(&wndw->plane); - kfree(data); -} - -static const struct drm_plane_funcs -nv50_wndw = { - .update_plane = drm_atomic_helper_update_plane, - .disable_plane = drm_atomic_helper_disable_plane, - .destroy = nv50_wndw_destroy, - .reset = nv50_wndw_reset, - .atomic_duplicate_state = nv50_wndw_atomic_duplicate_state, - .atomic_destroy_state = nv50_wndw_atomic_destroy_state, -}; - -static void -nv50_wndw_fini(struct nv50_wndw *wndw) -{ - nvif_notify_put(&wndw->notify); -} - -static void -nv50_wndw_init(struct nv50_wndw *wndw) -{ - nvif_notify_get(&wndw->notify); -} - -static int -nv50_wndw_ctor(const struct nv50_wndw_func *func, struct drm_device *dev, - enum drm_plane_type type, const char *name, int index, - struct nv50_dmac *dmac, const u32 *format, int nformat, - struct nv50_wndw *wndw) -{ - int ret; - - wndw->func = func; - wndw->dmac = dmac; - - ret = drm_universal_plane_init(dev, &wndw->plane, 0, &nv50_wndw, - format, nformat, NULL, - type, "%s-%d", name, index); - if (ret) - return ret; - - drm_plane_helper_add(&wndw->plane, &nv50_wndw_helper); - return 0; -} - -/****************************************************************************** - * Cursor plane - *****************************************************************************/ -#define nv50_curs(p) container_of((p), struct nv50_curs, wndw) - -struct nv50_curs { - struct nv50_wndw wndw; - struct nvif_object chan; -}; - -static u32 -nv50_curs_update(struct nv50_wndw *wndw, u32 interlock) -{ - struct nv50_curs *curs = nv50_curs(wndw); - nvif_wr32(&curs->chan, 0x0080, 0x00000000); - return 0; -} - -static void -nv50_curs_point(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw) -{ - struct nv50_curs *curs = nv50_curs(wndw); - nvif_wr32(&curs->chan, 0x0084, (asyw->point.y << 16) | asyw->point.x); -} - -static void -nv50_curs_prepare(struct nv50_wndw *wndw, struct nv50_head_atom *asyh, - struct nv50_wndw_atom *asyw) -{ - u32 handle = nv50_disp(wndw->plane.dev)->mast.base.vram.handle; - u32 offset = asyw->image.offset; - if (asyh->curs.handle != handle || asyh->curs.offset != offset) { - asyh->curs.handle = handle; - asyh->curs.offset = offset; - asyh->set.curs = asyh->curs.visible; - } -} - -static void -nv50_curs_release(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw, - struct nv50_head_atom *asyh) -{ - asyh->curs.visible = false; -} - -static int -nv50_curs_acquire(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw, - struct nv50_head_atom *asyh) -{ - int ret; - - ret = drm_atomic_helper_check_plane_state(&asyw->state, &asyh->state, - DRM_PLANE_HELPER_NO_SCALING, - DRM_PLANE_HELPER_NO_SCALING, - true, true); - asyh->curs.visible = asyw->state.visible; - if (ret || !asyh->curs.visible) - return ret; - - switch (asyw->state.fb->width) { - case 32: asyh->curs.layout = 0; break; - case 64: asyh->curs.layout = 1; break; - default: - return -EINVAL; - } - - if (asyw->state.fb->width != asyw->state.fb->height) - return -EINVAL; - - switch (asyw->state.fb->format->format) { - case DRM_FORMAT_ARGB8888: asyh->curs.format = 1; break; - default: - WARN_ON(1); - return -EINVAL; - } - - return 0; -} - -static void * -nv50_curs_dtor(struct nv50_wndw *wndw) -{ - struct nv50_curs *curs = nv50_curs(wndw); - nvif_object_fini(&curs->chan); - return curs; -} - -static const u32 -nv50_curs_format[] = { - DRM_FORMAT_ARGB8888, -}; - -static const struct nv50_wndw_func -nv50_curs = { - .dtor = nv50_curs_dtor, - .acquire = nv50_curs_acquire, - .release = nv50_curs_release, - .prepare = nv50_curs_prepare, - .point = nv50_curs_point, - .update = nv50_curs_update, -}; - -static int -nv50_curs_new(struct nouveau_drm *drm, struct nv50_head *head, - struct nv50_curs **pcurs) -{ - static const struct nvif_mclass curses[] = { - { GK104_DISP_CURSOR, 0 }, - { GF110_DISP_CURSOR, 0 }, - { GT214_DISP_CURSOR, 0 }, - { G82_DISP_CURSOR, 0 }, - { NV50_DISP_CURSOR, 0 }, - {} - }; - struct nv50_disp_cursor_v0 args = { - .head = head->base.index, - }; - struct nv50_disp *disp = nv50_disp(drm->dev); - struct nv50_curs *curs; - int cid, ret; - - cid = nvif_mclass(disp->disp, curses); - if (cid < 0) { - NV_ERROR(drm, "No supported cursor immediate class\n"); - return cid; - } - - if (!(curs = *pcurs = kzalloc(sizeof(*curs), GFP_KERNEL))) - return -ENOMEM; - - ret = nv50_wndw_ctor(&nv50_curs, drm->dev, DRM_PLANE_TYPE_CURSOR, - "curs", head->base.index, &disp->mast.base, - nv50_curs_format, ARRAY_SIZE(nv50_curs_format), - &curs->wndw); - if (ret) { - kfree(curs); - return ret; - } - - ret = nvif_object_init(disp->disp, 0, curses[cid].oclass, &args, - sizeof(args), &curs->chan); - if (ret) { - NV_ERROR(drm, "curs%04x allocation failed: %d\n", - curses[cid].oclass, ret); - return ret; - } - - return 0; -} - -/****************************************************************************** - * Primary plane - *****************************************************************************/ -#define nv50_base(p) container_of((p), struct nv50_base, wndw) - -struct nv50_base { - struct nv50_wndw wndw; - struct nv50_sync chan; - int id; -}; - -static int -nv50_base_notify(struct nvif_notify *notify) -{ - return NVIF_NOTIFY_KEEP; -} - -static void -nv50_base_lut(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw) -{ - struct nv50_base *base = nv50_base(wndw); - u32 *push; - if ((push = evo_wait(&base->chan, 2))) { - evo_mthd(push, 0x00e0, 1); - evo_data(push, asyw->lut.enable << 30); - evo_kick(push, &base->chan); - } -} - -static void -nv50_base_image_clr(struct nv50_wndw *wndw) -{ - struct nv50_base *base = nv50_base(wndw); - u32 *push; - if ((push = evo_wait(&base->chan, 4))) { - evo_mthd(push, 0x0084, 1); - evo_data(push, 0x00000000); - evo_mthd(push, 0x00c0, 1); - evo_data(push, 0x00000000); - evo_kick(push, &base->chan); - } -} - -static void -nv50_base_image_set(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw) -{ - struct nv50_base *base = nv50_base(wndw); - const s32 oclass = base->chan.base.base.user.oclass; - u32 *push; - if ((push = evo_wait(&base->chan, 10))) { - evo_mthd(push, 0x0084, 1); - evo_data(push, (asyw->image.mode << 8) | - (asyw->image.interval << 4)); - evo_mthd(push, 0x00c0, 1); - evo_data(push, asyw->image.handle); - if (oclass < G82_DISP_BASE_CHANNEL_DMA) { - evo_mthd(push, 0x0800, 5); - evo_data(push, asyw->image.offset >> 8); - evo_data(push, 0x00000000); - evo_data(push, (asyw->image.h << 16) | asyw->image.w); - evo_data(push, (asyw->image.layout << 20) | - asyw->image.pitch | - asyw->image.block); - evo_data(push, (asyw->image.kind << 16) | - (asyw->image.format << 8)); - } else - if (oclass < GF110_DISP_BASE_CHANNEL_DMA) { - evo_mthd(push, 0x0800, 5); - evo_data(push, asyw->image.offset >> 8); - evo_data(push, 0x00000000); - evo_data(push, (asyw->image.h << 16) | asyw->image.w); - evo_data(push, (asyw->image.layout << 20) | - asyw->image.pitch | - asyw->image.block); - evo_data(push, asyw->image.format << 8); - } else { - evo_mthd(push, 0x0400, 5); - evo_data(push, asyw->image.offset >> 8); - evo_data(push, 0x00000000); - evo_data(push, (asyw->image.h << 16) | asyw->image.w); - evo_data(push, (asyw->image.layout << 24) | - asyw->image.pitch | - asyw->image.block); - evo_data(push, asyw->image.format << 8); - } - evo_kick(push, &base->chan); - } -} - -static void -nv50_base_ntfy_clr(struct nv50_wndw *wndw) -{ - struct nv50_base *base = nv50_base(wndw); - u32 *push; - if ((push = evo_wait(&base->chan, 2))) { - evo_mthd(push, 0x00a4, 1); - evo_data(push, 0x00000000); - evo_kick(push, &base->chan); - } -} - -static void -nv50_base_ntfy_set(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw) -{ - struct nv50_base *base = nv50_base(wndw); - u32 *push; - if ((push = evo_wait(&base->chan, 3))) { - evo_mthd(push, 0x00a0, 2); - evo_data(push, (asyw->ntfy.awaken << 30) | asyw->ntfy.offset); - evo_data(push, asyw->ntfy.handle); - evo_kick(push, &base->chan); - } -} - -static void -nv50_base_sema_clr(struct nv50_wndw *wndw) -{ - struct nv50_base *base = nv50_base(wndw); - u32 *push; - if ((push = evo_wait(&base->chan, 2))) { - evo_mthd(push, 0x0094, 1); - evo_data(push, 0x00000000); - evo_kick(push, &base->chan); - } -} - -static void -nv50_base_sema_set(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw) -{ - struct nv50_base *base = nv50_base(wndw); - u32 *push; - if ((push = evo_wait(&base->chan, 5))) { - evo_mthd(push, 0x0088, 4); - evo_data(push, asyw->sema.offset); - evo_data(push, asyw->sema.acquire); - evo_data(push, asyw->sema.release); - evo_data(push, asyw->sema.handle); - evo_kick(push, &base->chan); - } -} - -static u32 -nv50_base_update(struct nv50_wndw *wndw, u32 interlock) -{ - struct nv50_base *base = nv50_base(wndw); - u32 *push; - - if (!(push = evo_wait(&base->chan, 2))) - return 0; - evo_mthd(push, 0x0080, 1); - evo_data(push, interlock); - evo_kick(push, &base->chan); - - if (base->chan.base.base.user.oclass < GF110_DISP_BASE_CHANNEL_DMA) - return interlock ? 2 << (base->id * 8) : 0; - return interlock ? 2 << (base->id * 4) : 0; -} - -static int -nv50_base_ntfy_wait_begun(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw) -{ - struct nouveau_drm *drm = nouveau_drm(wndw->plane.dev); - struct nv50_disp *disp = nv50_disp(wndw->plane.dev); - if (nvif_msec(&drm->client.device, 2000ULL, - u32 data = nouveau_bo_rd32(disp->sync, asyw->ntfy.offset / 4); - if ((data & 0xc0000000) == 0x40000000) - break; - usleep_range(1, 2); - ) < 0) - return -ETIMEDOUT; - return 0; -} - -static void -nv50_base_release(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw, - struct nv50_head_atom *asyh) -{ - asyh->base.cpp = 0; -} - -static int -nv50_base_acquire(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw, - struct nv50_head_atom *asyh) -{ - const struct drm_framebuffer *fb = asyw->state.fb; - int ret; - - if (!fb->format->depth) - return -EINVAL; - - ret = drm_atomic_helper_check_plane_state(&asyw->state, &asyh->state, - DRM_PLANE_HELPER_NO_SCALING, - DRM_PLANE_HELPER_NO_SCALING, - false, true); - if (ret) - return ret; - - asyh->base.depth = fb->format->depth; - asyh->base.cpp = fb->format->cpp[0]; - asyh->base.x = asyw->state.src.x1 >> 16; - asyh->base.y = asyw->state.src.y1 >> 16; - asyh->base.w = asyw->state.fb->width; - asyh->base.h = asyw->state.fb->height; - - switch (fb->format->format) { - case DRM_FORMAT_C8 : asyw->image.format = 0x1e; break; - case DRM_FORMAT_RGB565 : asyw->image.format = 0xe8; break; - case DRM_FORMAT_XRGB1555 : - case DRM_FORMAT_ARGB1555 : asyw->image.format = 0xe9; break; - case DRM_FORMAT_XRGB8888 : - case DRM_FORMAT_ARGB8888 : asyw->image.format = 0xcf; break; - case DRM_FORMAT_XBGR2101010: - case DRM_FORMAT_ABGR2101010: asyw->image.format = 0xd1; break; - case DRM_FORMAT_XBGR8888 : - case DRM_FORMAT_ABGR8888 : asyw->image.format = 0xd5; break; - default: - WARN_ON(1); - return -EINVAL; - } - - asyw->lut.enable = 1; - asyw->set.image = true; - return 0; -} - -static void * -nv50_base_dtor(struct nv50_wndw *wndw) -{ - struct nv50_disp *disp = nv50_disp(wndw->plane.dev); - struct nv50_base *base = nv50_base(wndw); - nv50_dmac_destroy(&base->chan.base, disp->disp); - return base; -} - -static const u32 -nv50_base_format[] = { - DRM_FORMAT_C8, - DRM_FORMAT_RGB565, - DRM_FORMAT_XRGB1555, - DRM_FORMAT_ARGB1555, - DRM_FORMAT_XRGB8888, - DRM_FORMAT_ARGB8888, - DRM_FORMAT_XBGR2101010, - DRM_FORMAT_ABGR2101010, - DRM_FORMAT_XBGR8888, - DRM_FORMAT_ABGR8888, -}; - -static const struct nv50_wndw_func -nv50_base = { - .dtor = nv50_base_dtor, - .acquire = nv50_base_acquire, - .release = nv50_base_release, - .sema_set = nv50_base_sema_set, - .sema_clr = nv50_base_sema_clr, - .ntfy_set = nv50_base_ntfy_set, - .ntfy_clr = nv50_base_ntfy_clr, - .ntfy_wait_begun = nv50_base_ntfy_wait_begun, - .image_set = nv50_base_image_set, - .image_clr = nv50_base_image_clr, - .lut = nv50_base_lut, - .update = nv50_base_update, -}; - -static int -nv50_base_new(struct nouveau_drm *drm, struct nv50_head *head, - struct nv50_base **pbase) -{ - struct nv50_disp *disp = nv50_disp(drm->dev); - struct nv50_base *base; - int ret; - - if (!(base = *pbase = kzalloc(sizeof(*base), GFP_KERNEL))) - return -ENOMEM; - base->id = head->base.index; - base->wndw.ntfy = EVO_FLIP_NTFY0(base->id); - base->wndw.sema = EVO_FLIP_SEM0(base->id); - base->wndw.data = 0x00000000; - - ret = nv50_wndw_ctor(&nv50_base, drm->dev, DRM_PLANE_TYPE_PRIMARY, - "base", base->id, &base->chan.base, - nv50_base_format, ARRAY_SIZE(nv50_base_format), - &base->wndw); - if (ret) { - kfree(base); - return ret; - } - - ret = nv50_base_create(&drm->client.device, disp->disp, base->id, - disp->sync->bo.offset, &base->chan); - if (ret) - return ret; - - return nvif_notify_init(&base->chan.base.base.user, nv50_base_notify, - false, - NV50_DISP_BASE_CHANNEL_DMA_V0_NTFY_UEVENT, - &(struct nvif_notify_uevent_req) {}, - sizeof(struct nvif_notify_uevent_req), - sizeof(struct nvif_notify_uevent_rep), - &base->wndw.notify); -} - -/****************************************************************************** - * Head - *****************************************************************************/ -static void -nv50_head_procamp(struct nv50_head *head, struct nv50_head_atom *asyh) -{ - struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->mast.base; - u32 *push; - if ((push = evo_wait(core, 2))) { - if (core->base.user.oclass < GF110_DISP_CORE_CHANNEL_DMA) - evo_mthd(push, 0x08a8 + (head->base.index * 0x400), 1); - else - evo_mthd(push, 0x0498 + (head->base.index * 0x300), 1); - evo_data(push, (asyh->procamp.sat.sin << 20) | - (asyh->procamp.sat.cos << 8)); - evo_kick(push, core); - } -} - -static void -nv50_head_dither(struct nv50_head *head, struct nv50_head_atom *asyh) -{ - struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->mast.base; - u32 *push; - if ((push = evo_wait(core, 2))) { - if (core->base.user.oclass < GF110_DISP_CORE_CHANNEL_DMA) - evo_mthd(push, 0x08a0 + (head->base.index * 0x0400), 1); - else - if (core->base.user.oclass < GK104_DISP_CORE_CHANNEL_DMA) - evo_mthd(push, 0x0490 + (head->base.index * 0x0300), 1); - else - evo_mthd(push, 0x04a0 + (head->base.index * 0x0300), 1); - evo_data(push, (asyh->dither.mode << 3) | - (asyh->dither.bits << 1) | - asyh->dither.enable); - evo_kick(push, core); - } -} - -static void -nv50_head_ovly(struct nv50_head *head, struct nv50_head_atom *asyh) -{ - struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->mast.base; - u32 bounds = 0; - u32 *push; - - if (asyh->base.cpp) { - switch (asyh->base.cpp) { - case 8: bounds |= 0x00000500; break; - case 4: bounds |= 0x00000300; break; - case 2: bounds |= 0x00000100; break; - default: - WARN_ON(1); - break; - } - bounds |= 0x00000001; - } - - if ((push = evo_wait(core, 2))) { - if (core->base.user.oclass < GF110_DISP_CORE_CHANNEL_DMA) - evo_mthd(push, 0x0904 + head->base.index * 0x400, 1); - else - evo_mthd(push, 0x04d4 + head->base.index * 0x300, 1); - evo_data(push, bounds); - evo_kick(push, core); - } -} - -static void -nv50_head_base(struct nv50_head *head, struct nv50_head_atom *asyh) -{ - struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->mast.base; - u32 bounds = 0; - u32 *push; - - if (asyh->base.cpp) { - switch (asyh->base.cpp) { - case 8: bounds |= 0x00000500; break; - case 4: bounds |= 0x00000300; break; - case 2: bounds |= 0x00000100; break; - case 1: bounds |= 0x00000000; break; - default: - WARN_ON(1); - break; - } - bounds |= 0x00000001; - } - - if ((push = evo_wait(core, 2))) { - if (core->base.user.oclass < GF110_DISP_CORE_CHANNEL_DMA) - evo_mthd(push, 0x0900 + head->base.index * 0x400, 1); - else - evo_mthd(push, 0x04d0 + head->base.index * 0x300, 1); - evo_data(push, bounds); - evo_kick(push, core); - } -} - -static void -nv50_head_curs_clr(struct nv50_head *head) -{ - struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->mast.base; - u32 *push; - if ((push = evo_wait(core, 4))) { - if (core->base.user.oclass < G82_DISP_CORE_CHANNEL_DMA) { - evo_mthd(push, 0x0880 + head->base.index * 0x400, 1); - evo_data(push, 0x05000000); - } else - if (core->base.user.oclass < GF110_DISP_CORE_CHANNEL_DMA) { - evo_mthd(push, 0x0880 + head->base.index * 0x400, 1); - evo_data(push, 0x05000000); - evo_mthd(push, 0x089c + head->base.index * 0x400, 1); - evo_data(push, 0x00000000); - } else { - evo_mthd(push, 0x0480 + head->base.index * 0x300, 1); - evo_data(push, 0x05000000); - evo_mthd(push, 0x048c + head->base.index * 0x300, 1); - evo_data(push, 0x00000000); - } - evo_kick(push, core); - } -} - -static void -nv50_head_curs_set(struct nv50_head *head, struct nv50_head_atom *asyh) -{ - struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->mast.base; - u32 *push; - if ((push = evo_wait(core, 5))) { - if (core->base.user.oclass < G82_DISP_BASE_CHANNEL_DMA) { - evo_mthd(push, 0x0880 + head->base.index * 0x400, 2); - evo_data(push, 0x80000000 | (asyh->curs.layout << 26) | - (asyh->curs.format << 24)); - evo_data(push, asyh->curs.offset >> 8); - } else - if (core->base.user.oclass < GF110_DISP_BASE_CHANNEL_DMA) { - evo_mthd(push, 0x0880 + head->base.index * 0x400, 2); - evo_data(push, 0x80000000 | (asyh->curs.layout << 26) | - (asyh->curs.format << 24)); - evo_data(push, asyh->curs.offset >> 8); - evo_mthd(push, 0x089c + head->base.index * 0x400, 1); - evo_data(push, asyh->curs.handle); - } else { - evo_mthd(push, 0x0480 + head->base.index * 0x300, 2); - evo_data(push, 0x80000000 | (asyh->curs.layout << 26) | - (asyh->curs.format << 24)); - evo_data(push, asyh->curs.offset >> 8); - evo_mthd(push, 0x048c + head->base.index * 0x300, 1); - evo_data(push, asyh->curs.handle); - } - evo_kick(push, core); - } -} - -static void -nv50_head_core_clr(struct nv50_head *head) -{ - struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->mast.base; - u32 *push; - if ((push = evo_wait(core, 2))) { - if (core->base.user.oclass < GF110_DISP_CORE_CHANNEL_DMA) - evo_mthd(push, 0x0874 + head->base.index * 0x400, 1); - else - evo_mthd(push, 0x0474 + head->base.index * 0x300, 1); - evo_data(push, 0x00000000); - evo_kick(push, core); - } -} - -static void -nv50_head_core_set(struct nv50_head *head, struct nv50_head_atom *asyh) -{ - struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->mast.base; - u32 *push; - if ((push = evo_wait(core, 9))) { - if (core->base.user.oclass < G82_DISP_CORE_CHANNEL_DMA) { - evo_mthd(push, 0x0860 + head->base.index * 0x400, 1); - evo_data(push, asyh->core.offset >> 8); - evo_mthd(push, 0x0868 + head->base.index * 0x400, 4); - evo_data(push, (asyh->core.h << 16) | asyh->core.w); - evo_data(push, asyh->core.layout << 20 | - (asyh->core.pitch >> 8) << 8 | - asyh->core.block); - evo_data(push, asyh->core.kind << 16 | - asyh->core.format << 8); - evo_data(push, asyh->core.handle); - evo_mthd(push, 0x08c0 + head->base.index * 0x400, 1); - evo_data(push, (asyh->core.y << 16) | asyh->core.x); - /* EVO will complain with INVALID_STATE if we have an - * active cursor and (re)specify HeadSetContextDmaIso - * without also updating HeadSetOffsetCursor. - */ - asyh->set.curs = asyh->curs.visible; - } else - if (core->base.user.oclass < GF110_DISP_CORE_CHANNEL_DMA) { - evo_mthd(push, 0x0860 + head->base.index * 0x400, 1); - evo_data(push, asyh->core.offset >> 8); - evo_mthd(push, 0x0868 + head->base.index * 0x400, 4); - evo_data(push, (asyh->core.h << 16) | asyh->core.w); - evo_data(push, asyh->core.layout << 20 | - (asyh->core.pitch >> 8) << 8 | - asyh->core.block); - evo_data(push, asyh->core.format << 8); - evo_data(push, asyh->core.handle); - evo_mthd(push, 0x08c0 + head->base.index * 0x400, 1); - evo_data(push, (asyh->core.y << 16) | asyh->core.x); - } else { - evo_mthd(push, 0x0460 + head->base.index * 0x300, 1); - evo_data(push, asyh->core.offset >> 8); - evo_mthd(push, 0x0468 + head->base.index * 0x300, 4); - evo_data(push, (asyh->core.h << 16) | asyh->core.w); - evo_data(push, asyh->core.layout << 24 | - (asyh->core.pitch >> 8) << 8 | - asyh->core.block); - evo_data(push, asyh->core.format << 8); - evo_data(push, asyh->core.handle); - evo_mthd(push, 0x04b0 + head->base.index * 0x300, 1); - evo_data(push, (asyh->core.y << 16) | asyh->core.x); - } - evo_kick(push, core); - } -} - -static void -nv50_head_lut_clr(struct nv50_head *head) -{ - struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->mast.base; - u32 *push; - if ((push = evo_wait(core, 4))) { - if (core->base.user.oclass < G82_DISP_CORE_CHANNEL_DMA) { - evo_mthd(push, 0x0840 + (head->base.index * 0x400), 1); - evo_data(push, 0x40000000); - } else - if (core->base.user.oclass < GF110_DISP_CORE_CHANNEL_DMA) { - evo_mthd(push, 0x0840 + (head->base.index * 0x400), 1); - evo_data(push, 0x40000000); - evo_mthd(push, 0x085c + (head->base.index * 0x400), 1); - evo_data(push, 0x00000000); - } else { - evo_mthd(push, 0x0440 + (head->base.index * 0x300), 1); - evo_data(push, 0x03000000); - evo_mthd(push, 0x045c + (head->base.index * 0x300), 1); - evo_data(push, 0x00000000); - } - evo_kick(push, core); - } -} - -static void -nv50_head_lut_load(struct drm_property_blob *blob, int mode, - struct nouveau_bo *nvbo) -{ - struct drm_color_lut *in = (struct drm_color_lut *)blob->data; - void __iomem *lut = (u8 *)nvbo_kmap_obj_iovirtual(nvbo); - const int size = blob->length / sizeof(*in); - int bits, shift, i; - u16 zero, r, g, b; - - /* This can't happen.. But it shuts the compiler up. */ - if (WARN_ON(size != 256)) - return; - - switch (mode) { - case 0: /* LORES. */ - case 1: /* HIRES. */ - bits = 11; - shift = 3; - zero = 0x0000; - break; - case 7: /* INTERPOLATE_257_UNITY_RANGE. */ - bits = 14; - shift = 0; - zero = 0x6000; - break; - default: - WARN_ON(1); - return; - } - - for (i = 0; i < size; i++) { - r = (drm_color_lut_extract(in[i]. red, bits) + zero) << shift; - g = (drm_color_lut_extract(in[i].green, bits) + zero) << shift; - b = (drm_color_lut_extract(in[i]. blue, bits) + zero) << shift; - writew(r, lut + (i * 0x08) + 0); - writew(g, lut + (i * 0x08) + 2); - writew(b, lut + (i * 0x08) + 4); - } - - /* INTERPOLATE modes require a "next" entry to interpolate with, - * so we replicate the last entry to deal with this for now. - */ - writew(r, lut + (i * 0x08) + 0); - writew(g, lut + (i * 0x08) + 2); - writew(b, lut + (i * 0x08) + 4); -} - -static void -nv50_head_lut_set(struct nv50_head *head, struct nv50_head_atom *asyh) -{ - struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->mast.base; - u32 *push; - if ((push = evo_wait(core, 7))) { - if (core->base.user.oclass < G82_DISP_CORE_CHANNEL_DMA) { - evo_mthd(push, 0x0840 + (head->base.index * 0x400), 2); - evo_data(push, 0x80000000 | asyh->lut.mode << 30); - evo_data(push, asyh->lut.offset >> 8); - } else - if (core->base.user.oclass < GF110_DISP_CORE_CHANNEL_DMA) { - evo_mthd(push, 0x0840 + (head->base.index * 0x400), 2); - evo_data(push, 0x80000000 | asyh->lut.mode << 30); - evo_data(push, asyh->lut.offset >> 8); - evo_mthd(push, 0x085c + (head->base.index * 0x400), 1); - evo_data(push, asyh->lut.handle); - } else { - evo_mthd(push, 0x0440 + (head->base.index * 0x300), 4); - evo_data(push, 0x80000000 | asyh->lut.mode << 24); - evo_data(push, asyh->lut.offset >> 8); - evo_data(push, 0x00000000); - evo_data(push, 0x00000000); - evo_mthd(push, 0x045c + (head->base.index * 0x300), 1); - evo_data(push, asyh->lut.handle); - } - evo_kick(push, core); - } -} - -static void -nv50_head_mode(struct nv50_head *head, struct nv50_head_atom *asyh) -{ - struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->mast.base; - struct nv50_head_mode *m = &asyh->mode; - u32 *push; - if ((push = evo_wait(core, 14))) { - if (core->base.user.oclass < GF110_DISP_CORE_CHANNEL_DMA) { - evo_mthd(push, 0x0804 + (head->base.index * 0x400), 2); - evo_data(push, 0x00800000 | m->clock); - evo_data(push, m->interlace ? 0x00000002 : 0x00000000); - evo_mthd(push, 0x0810 + (head->base.index * 0x400), 7); - evo_data(push, 0x00000000); - evo_data(push, (m->v.active << 16) | m->h.active ); - evo_data(push, (m->v.synce << 16) | m->h.synce ); - evo_data(push, (m->v.blanke << 16) | m->h.blanke ); - evo_data(push, (m->v.blanks << 16) | m->h.blanks ); - evo_data(push, (m->v.blank2e << 16) | m->v.blank2s); - evo_data(push, asyh->mode.v.blankus); - evo_mthd(push, 0x082c + (head->base.index * 0x400), 1); - evo_data(push, 0x00000000); - } else { - evo_mthd(push, 0x0410 + (head->base.index * 0x300), 6); - evo_data(push, 0x00000000); - evo_data(push, (m->v.active << 16) | m->h.active ); - evo_data(push, (m->v.synce << 16) | m->h.synce ); - evo_data(push, (m->v.blanke << 16) | m->h.blanke ); - evo_data(push, (m->v.blanks << 16) | m->h.blanks ); - evo_data(push, (m->v.blank2e << 16) | m->v.blank2s); - evo_mthd(push, 0x042c + (head->base.index * 0x300), 2); - evo_data(push, 0x00000000); /* ??? */ - evo_data(push, 0xffffff00); - evo_mthd(push, 0x0450 + (head->base.index * 0x300), 3); - evo_data(push, m->clock * 1000); - evo_data(push, 0x00200000); /* ??? */ - evo_data(push, m->clock * 1000); - } - evo_kick(push, core); - } -} - -static void -nv50_head_view(struct nv50_head *head, struct nv50_head_atom *asyh) -{ - struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->mast.base; - u32 *push; - if ((push = evo_wait(core, 10))) { - if (core->base.user.oclass < GF110_DISP_CORE_CHANNEL_DMA) { - evo_mthd(push, 0x08a4 + (head->base.index * 0x400), 1); - evo_data(push, 0x00000000); - evo_mthd(push, 0x08c8 + (head->base.index * 0x400), 1); - evo_data(push, (asyh->view.iH << 16) | asyh->view.iW); - evo_mthd(push, 0x08d8 + (head->base.index * 0x400), 2); - evo_data(push, (asyh->view.oH << 16) | asyh->view.oW); - evo_data(push, (asyh->view.oH << 16) | asyh->view.oW); - } else { - evo_mthd(push, 0x0494 + (head->base.index * 0x300), 1); - evo_data(push, 0x00000000); - evo_mthd(push, 0x04b8 + (head->base.index * 0x300), 1); - evo_data(push, (asyh->view.iH << 16) | asyh->view.iW); - evo_mthd(push, 0x04c0 + (head->base.index * 0x300), 3); - evo_data(push, (asyh->view.oH << 16) | asyh->view.oW); - evo_data(push, (asyh->view.oH << 16) | asyh->view.oW); - evo_data(push, (asyh->view.oH << 16) | asyh->view.oW); - } - evo_kick(push, core); - } -} - -static void -nv50_head_flush_clr(struct nv50_head *head, struct nv50_head_atom *asyh, bool y) -{ - if (asyh->clr.ilut && (!asyh->set.ilut || y)) - nv50_head_lut_clr(head); - if (asyh->clr.core && (!asyh->set.core || y)) - nv50_head_core_clr(head); - if (asyh->clr.curs && (!asyh->set.curs || y)) - nv50_head_curs_clr(head); -} - -static void -nv50_head_flush_set(struct nv50_head *head, struct nv50_head_atom *asyh) -{ - if (asyh->set.view ) nv50_head_view (head, asyh); - if (asyh->set.mode ) nv50_head_mode (head, asyh); - if (asyh->set.ilut ) { - struct nouveau_bo *nvbo = head->lut.nvbo[head->lut.next]; - struct drm_property_blob *blob = asyh->state.gamma_lut; - if (blob) - nv50_head_lut_load(blob, asyh->lut.mode, nvbo); - asyh->lut.offset = nvbo->bo.offset; - head->lut.next ^= 1; - nv50_head_lut_set(head, asyh); - } - if (asyh->set.core ) nv50_head_core_set(head, asyh); - if (asyh->set.curs ) nv50_head_curs_set(head, asyh); - if (asyh->set.base ) nv50_head_base (head, asyh); - if (asyh->set.ovly ) nv50_head_ovly (head, asyh); - if (asyh->set.dither ) nv50_head_dither (head, asyh); - if (asyh->set.procamp) nv50_head_procamp (head, asyh); -} - -static void -nv50_head_atomic_check_procamp(struct nv50_head_atom *armh, - struct nv50_head_atom *asyh, - struct nouveau_conn_atom *asyc) -{ - const int vib = asyc->procamp.color_vibrance - 100; - const int hue = asyc->procamp.vibrant_hue - 90; - const int adj = (vib > 0) ? 50 : 0; - asyh->procamp.sat.cos = ((vib * 2047 + adj) / 100) & 0xfff; - asyh->procamp.sat.sin = ((hue * 2047) / 100) & 0xfff; - asyh->set.procamp = true; -} - -static void -nv50_head_atomic_check_dither(struct nv50_head_atom *armh, - struct nv50_head_atom *asyh, - struct nouveau_conn_atom *asyc) -{ - struct drm_connector *connector = asyc->state.connector; - u32 mode = 0x00; - - if (asyc->dither.mode == DITHERING_MODE_AUTO) { - if (asyh->base.depth > connector->display_info.bpc * 3) - mode = DITHERING_MODE_DYNAMIC2X2; - } else { - mode = asyc->dither.mode; - } - - if (asyc->dither.depth == DITHERING_DEPTH_AUTO) { - if (connector->display_info.bpc >= 8) - mode |= DITHERING_DEPTH_8BPC; - } else { - mode |= asyc->dither.depth; - } - - asyh->dither.enable = mode; - asyh->dither.bits = mode >> 1; - asyh->dither.mode = mode >> 3; - asyh->set.dither = true; -} - -static void -nv50_head_atomic_check_view(struct nv50_head_atom *armh, - struct nv50_head_atom *asyh, - struct nouveau_conn_atom *asyc) -{ - struct drm_connector *connector = asyc->state.connector; - struct drm_display_mode *omode = &asyh->state.adjusted_mode; - struct drm_display_mode *umode = &asyh->state.mode; - int mode = asyc->scaler.mode; - struct edid *edid; - int umode_vdisplay, omode_hdisplay, omode_vdisplay; - - if (connector->edid_blob_ptr) - edid = (struct edid *)connector->edid_blob_ptr->data; - else - edid = NULL; - - if (!asyc->scaler.full) { - if (mode == DRM_MODE_SCALE_NONE) - omode = umode; - } else { - /* Non-EDID LVDS/eDP mode. */ - mode = DRM_MODE_SCALE_FULLSCREEN; - } - - /* For the user-specified mode, we must ignore doublescan and - * the like, but honor frame packing. - */ - umode_vdisplay = umode->vdisplay; - if ((umode->flags & DRM_MODE_FLAG_3D_MASK) == DRM_MODE_FLAG_3D_FRAME_PACKING) - umode_vdisplay += umode->vtotal; - asyh->view.iW = umode->hdisplay; - asyh->view.iH = umode_vdisplay; - /* For the output mode, we can just use the stock helper. */ - drm_mode_get_hv_timing(omode, &omode_hdisplay, &omode_vdisplay); - asyh->view.oW = omode_hdisplay; - asyh->view.oH = omode_vdisplay; - - /* Add overscan compensation if necessary, will keep the aspect - * ratio the same as the backend mode unless overridden by the - * user setting both hborder and vborder properties. - */ - if ((asyc->scaler.underscan.mode == UNDERSCAN_ON || - (asyc->scaler.underscan.mode == UNDERSCAN_AUTO && - drm_detect_hdmi_monitor(edid)))) { - u32 bX = asyc->scaler.underscan.hborder; - u32 bY = asyc->scaler.underscan.vborder; - u32 r = (asyh->view.oH << 19) / asyh->view.oW; - - if (bX) { - asyh->view.oW -= (bX * 2); - if (bY) asyh->view.oH -= (bY * 2); - else asyh->view.oH = ((asyh->view.oW * r) + (r / 2)) >> 19; - } else { - asyh->view.oW -= (asyh->view.oW >> 4) + 32; - if (bY) asyh->view.oH -= (bY * 2); - else asyh->view.oH = ((asyh->view.oW * r) + (r / 2)) >> 19; - } - } - - /* Handle CENTER/ASPECT scaling, taking into account the areas - * removed already for overscan compensation. - */ - switch (mode) { - case DRM_MODE_SCALE_CENTER: - asyh->view.oW = min((u16)umode->hdisplay, asyh->view.oW); - asyh->view.oH = min((u16)umode_vdisplay, asyh->view.oH); - /* fall-through */ - case DRM_MODE_SCALE_ASPECT: - if (asyh->view.oH < asyh->view.oW) { - u32 r = (asyh->view.iW << 19) / asyh->view.iH; - asyh->view.oW = ((asyh->view.oH * r) + (r / 2)) >> 19; - } else { - u32 r = (asyh->view.iH << 19) / asyh->view.iW; - asyh->view.oH = ((asyh->view.oW * r) + (r / 2)) >> 19; - } - break; - default: - break; - } - - asyh->set.view = true; -} - -static void -nv50_head_atomic_check_lut(struct nv50_head *head, - struct nv50_head_atom *armh, - struct nv50_head_atom *asyh) -{ - struct nv50_disp *disp = nv50_disp(head->base.base.dev); - - /* An I8 surface without an input LUT makes no sense, and - * EVO will throw an error if you try. - * - * Legacy clients actually cause this due to the order in - * which they call ioctls, so we will enable the LUT with - * whatever contents the buffer already contains to avoid - * triggering the error check. - */ - if (!asyh->state.gamma_lut && asyh->base.cpp != 1) { - asyh->lut.handle = 0; - asyh->clr.ilut = armh->lut.visible; - return; - } - - if (disp->disp->oclass < GF110_DISP) { - asyh->lut.mode = (asyh->base.cpp == 1) ? 0 : 1; - asyh->set.ilut = true; - } else { - asyh->lut.mode = 7; - asyh->set.ilut = asyh->state.color_mgmt_changed; - } - asyh->lut.handle = disp->mast.base.vram.handle; -} - -static void -nv50_head_atomic_check_mode(struct nv50_head *head, struct nv50_head_atom *asyh) -{ - struct drm_display_mode *mode = &asyh->state.adjusted_mode; - struct nv50_head_mode *m = &asyh->mode; - u32 blankus; - - drm_mode_set_crtcinfo(mode, CRTC_INTERLACE_HALVE_V | CRTC_STEREO_DOUBLE); - - /* - * DRM modes are defined in terms of a repeating interval - * starting with the active display area. The hardware modes - * are defined in terms of a repeating interval starting one - * unit (pixel or line) into the sync pulse. So, add bias. - */ - - m->h.active = mode->crtc_htotal; - m->h.synce = mode->crtc_hsync_end - mode->crtc_hsync_start - 1; - m->h.blanke = mode->crtc_hblank_end - mode->crtc_hsync_start - 1; - m->h.blanks = m->h.blanke + mode->crtc_hdisplay; - - m->v.active = mode->crtc_vtotal; - m->v.synce = mode->crtc_vsync_end - mode->crtc_vsync_start - 1; - m->v.blanke = mode->crtc_vblank_end - mode->crtc_vsync_start - 1; - m->v.blanks = m->v.blanke + mode->crtc_vdisplay; - - /*XXX: Safe underestimate, even "0" works */ - blankus = (m->v.active - mode->crtc_vdisplay - 2) * m->h.active; - blankus *= 1000; - blankus /= mode->crtc_clock; - m->v.blankus = blankus; - - if (mode->flags & DRM_MODE_FLAG_INTERLACE) { - m->v.blank2e = m->v.active + m->v.blanke; - m->v.blank2s = m->v.blank2e + mode->crtc_vdisplay; - m->v.active = (m->v.active * 2) + 1; - m->interlace = true; - } else { - m->v.blank2e = 0; - m->v.blank2s = 1; - m->interlace = false; - } - m->clock = mode->crtc_clock; - - asyh->set.mode = true; -} - -static int -nv50_head_atomic_check(struct drm_crtc *crtc, struct drm_crtc_state *state) -{ - struct nouveau_drm *drm = nouveau_drm(crtc->dev); - struct nv50_disp *disp = nv50_disp(crtc->dev); - struct nv50_head *head = nv50_head(crtc); - struct nv50_head_atom *armh = nv50_head_atom(crtc->state); - struct nv50_head_atom *asyh = nv50_head_atom(state); - struct nouveau_conn_atom *asyc = NULL; - struct drm_connector_state *conns; - struct drm_connector *conn; - int i; - - NV_ATOMIC(drm, "%s atomic_check %d\n", crtc->name, asyh->state.active); - if (asyh->state.active) { - for_each_new_connector_in_state(asyh->state.state, conn, conns, i) { - if (conns->crtc == crtc) { - asyc = nouveau_conn_atom(conns); - break; - } - } - - if (armh->state.active) { - if (asyc) { - if (asyh->state.mode_changed) - asyc->set.scaler = true; - if (armh->base.depth != asyh->base.depth) - asyc->set.dither = true; - } - } else { - if (asyc) - asyc->set.mask = ~0; - asyh->set.mask = ~0; - } - - if (asyh->state.mode_changed) - nv50_head_atomic_check_mode(head, asyh); - - if (asyh->state.color_mgmt_changed || - asyh->base.cpp != armh->base.cpp) - nv50_head_atomic_check_lut(head, armh, asyh); - asyh->lut.visible = asyh->lut.handle != 0; - - if (asyc) { - if (asyc->set.scaler) - nv50_head_atomic_check_view(armh, asyh, asyc); - if (asyc->set.dither) - nv50_head_atomic_check_dither(armh, asyh, asyc); - if (asyc->set.procamp) - nv50_head_atomic_check_procamp(armh, asyh, asyc); - } - - if ((asyh->core.visible = (asyh->base.cpp != 0))) { - asyh->core.x = asyh->base.x; - asyh->core.y = asyh->base.y; - asyh->core.w = asyh->base.w; - asyh->core.h = asyh->base.h; - } else - if ((asyh->core.visible = asyh->curs.visible) || - (asyh->core.visible = asyh->lut.visible)) { - /*XXX: We need to either find some way of having the - * primary base layer appear black, while still - * being able to display the other layers, or we - * need to allocate a dummy black surface here. - */ - asyh->core.x = 0; - asyh->core.y = 0; - asyh->core.w = asyh->state.mode.hdisplay; - asyh->core.h = asyh->state.mode.vdisplay; - } - asyh->core.handle = disp->mast.base.vram.handle; - asyh->core.offset = 0; - asyh->core.format = 0xcf; - asyh->core.kind = 0; - asyh->core.layout = 1; - asyh->core.block = 0; - asyh->core.pitch = ALIGN(asyh->core.w, 64) * 4; - asyh->set.base = armh->base.cpp != asyh->base.cpp; - asyh->set.ovly = armh->ovly.cpp != asyh->ovly.cpp; - } else { - asyh->lut.visible = false; - asyh->core.visible = false; - asyh->curs.visible = false; - asyh->base.cpp = 0; - asyh->ovly.cpp = 0; - } - - if (!drm_atomic_crtc_needs_modeset(&asyh->state)) { - if (asyh->core.visible) { - if (memcmp(&armh->core, &asyh->core, sizeof(asyh->core))) - asyh->set.core = true; - } else - if (armh->core.visible) { - asyh->clr.core = true; - } - - if (asyh->curs.visible) { - if (memcmp(&armh->curs, &asyh->curs, sizeof(asyh->curs))) - asyh->set.curs = true; - } else - if (armh->curs.visible) { - asyh->clr.curs = true; - } - } else { - asyh->clr.ilut = armh->lut.visible; - asyh->clr.core = armh->core.visible; - asyh->clr.curs = armh->curs.visible; - asyh->set.ilut = asyh->lut.visible; - asyh->set.core = asyh->core.visible; - asyh->set.curs = asyh->curs.visible; - } - - if (asyh->clr.mask || asyh->set.mask) - nv50_atom(asyh->state.state)->lock_core = true; - return 0; -} - -static const struct drm_crtc_helper_funcs -nv50_head_help = { - .atomic_check = nv50_head_atomic_check, -}; - -static void -nv50_head_atomic_destroy_state(struct drm_crtc *crtc, - struct drm_crtc_state *state) -{ - struct nv50_head_atom *asyh = nv50_head_atom(state); - __drm_atomic_helper_crtc_destroy_state(&asyh->state); - kfree(asyh); -} - -static struct drm_crtc_state * -nv50_head_atomic_duplicate_state(struct drm_crtc *crtc) -{ - struct nv50_head_atom *armh = nv50_head_atom(crtc->state); - struct nv50_head_atom *asyh; - if (!(asyh = kmalloc(sizeof(*asyh), GFP_KERNEL))) - return NULL; - __drm_atomic_helper_crtc_duplicate_state(crtc, &asyh->state); - asyh->view = armh->view; - asyh->mode = armh->mode; - asyh->lut = armh->lut; - asyh->core = armh->core; - asyh->curs = armh->curs; - asyh->base = armh->base; - asyh->ovly = armh->ovly; - asyh->dither = armh->dither; - asyh->procamp = armh->procamp; - asyh->clr.mask = 0; - asyh->set.mask = 0; - return &asyh->state; -} - -static void -__drm_atomic_helper_crtc_reset(struct drm_crtc *crtc, - struct drm_crtc_state *state) -{ - if (crtc->state) - crtc->funcs->atomic_destroy_state(crtc, crtc->state); - crtc->state = state; - crtc->state->crtc = crtc; -} - -static void -nv50_head_reset(struct drm_crtc *crtc) -{ - struct nv50_head_atom *asyh; - - if (WARN_ON(!(asyh = kzalloc(sizeof(*asyh), GFP_KERNEL)))) - return; - - __drm_atomic_helper_crtc_reset(crtc, &asyh->state); -} - -static void -nv50_head_destroy(struct drm_crtc *crtc) -{ - struct nv50_disp *disp = nv50_disp(crtc->dev); - struct nv50_head *head = nv50_head(crtc); - int i; - - nv50_dmac_destroy(&head->ovly.base, disp->disp); - nv50_pioc_destroy(&head->oimm.base); - - for (i = 0; i < ARRAY_SIZE(head->lut.nvbo); i++) - nouveau_bo_unmap_unpin_unref(&head->lut.nvbo[i]); - - drm_crtc_cleanup(crtc); - kfree(crtc); -} - -static const struct drm_crtc_funcs -nv50_head_func = { - .reset = nv50_head_reset, - .gamma_set = drm_atomic_helper_legacy_gamma_set, - .destroy = nv50_head_destroy, - .set_config = drm_atomic_helper_set_config, - .page_flip = drm_atomic_helper_page_flip, - .atomic_duplicate_state = nv50_head_atomic_duplicate_state, - .atomic_destroy_state = nv50_head_atomic_destroy_state, -}; - -static int -nv50_head_create(struct drm_device *dev, int index) -{ - struct nouveau_drm *drm = nouveau_drm(dev); - struct nvif_device *device = &drm->client.device; - struct nv50_disp *disp = nv50_disp(dev); - struct nv50_head *head; - struct nv50_base *base; - struct nv50_curs *curs; - struct drm_crtc *crtc; - int ret, i; - - head = kzalloc(sizeof(*head), GFP_KERNEL); - if (!head) - return -ENOMEM; - - head->base.index = index; - ret = nv50_base_new(drm, head, &base); - if (ret == 0) - ret = nv50_curs_new(drm, head, &curs); - if (ret) { - kfree(head); - return ret; - } - - crtc = &head->base.base; - drm_crtc_init_with_planes(dev, crtc, &base->wndw.plane, - &curs->wndw.plane, &nv50_head_func, - "head-%d", head->base.index); - drm_crtc_helper_add(crtc, &nv50_head_help); - drm_mode_crtc_set_gamma_size(crtc, 256); - - for (i = 0; i < ARRAY_SIZE(head->lut.nvbo); i++) { - ret = nouveau_bo_new_pin_map(&drm->client, 1025 * 8, 0x100, - TTM_PL_FLAG_VRAM, - &head->lut.nvbo[i]); - if (ret) - goto out; - } - - /* allocate overlay resources */ - ret = nv50_oimm_create(device, disp->disp, index, &head->oimm); - if (ret) - goto out; - - ret = nv50_ovly_create(device, disp->disp, index, disp->sync->bo.offset, - &head->ovly); - if (ret) - goto out; - -out: - if (ret) - nv50_head_destroy(crtc); - return ret; -} - -/****************************************************************************** - * Output path helpers - *****************************************************************************/ -static void -nv50_outp_release(struct nouveau_encoder *nv_encoder) -{ - struct nv50_disp *disp = nv50_disp(nv_encoder->base.base.dev); - struct { - struct nv50_disp_mthd_v1 base; - } args = { - .base.version = 1, - .base.method = NV50_DISP_MTHD_V1_RELEASE, - .base.hasht = nv_encoder->dcb->hasht, - .base.hashm = nv_encoder->dcb->hashm, - }; - - nvif_mthd(disp->disp, 0, &args, sizeof(args)); - nv_encoder->or = -1; - nv_encoder->link = 0; -} - -static int -nv50_outp_acquire(struct nouveau_encoder *nv_encoder) -{ - struct nouveau_drm *drm = nouveau_drm(nv_encoder->base.base.dev); - struct nv50_disp *disp = nv50_disp(drm->dev); - struct { - struct nv50_disp_mthd_v1 base; - struct nv50_disp_acquire_v0 info; - } args = { - .base.version = 1, - .base.method = NV50_DISP_MTHD_V1_ACQUIRE, - .base.hasht = nv_encoder->dcb->hasht, - .base.hashm = nv_encoder->dcb->hashm, - }; - int ret; - - ret = nvif_mthd(disp->disp, 0, &args, sizeof(args)); - if (ret) { - NV_ERROR(drm, "error acquiring output path: %d\n", ret); - return ret; - } - - nv_encoder->or = args.info.or; - nv_encoder->link = args.info.link; - return 0; -} - -static int -nv50_outp_atomic_check_view(struct drm_encoder *encoder, - struct drm_crtc_state *crtc_state, - struct drm_connector_state *conn_state, - struct drm_display_mode *native_mode) -{ - struct drm_display_mode *adjusted_mode = &crtc_state->adjusted_mode; - struct drm_display_mode *mode = &crtc_state->mode; - struct drm_connector *connector = conn_state->connector; - struct nouveau_conn_atom *asyc = nouveau_conn_atom(conn_state); - struct nouveau_drm *drm = nouveau_drm(encoder->dev); - - NV_ATOMIC(drm, "%s atomic_check\n", encoder->name); - asyc->scaler.full = false; - if (!native_mode) - return 0; - - if (asyc->scaler.mode == DRM_MODE_SCALE_NONE) { - switch (connector->connector_type) { - case DRM_MODE_CONNECTOR_LVDS: - case DRM_MODE_CONNECTOR_eDP: - /* Force use of scaler for non-EDID modes. */ - if (adjusted_mode->type & DRM_MODE_TYPE_DRIVER) - break; - mode = native_mode; - asyc->scaler.full = true; - break; - default: - break; - } - } else { - mode = native_mode; - } - - if (!drm_mode_equal(adjusted_mode, mode)) { - drm_mode_copy(adjusted_mode, mode); - crtc_state->mode_changed = true; - } - - return 0; -} - -static int -nv50_outp_atomic_check(struct drm_encoder *encoder, - struct drm_crtc_state *crtc_state, - struct drm_connector_state *conn_state) -{ - struct nouveau_connector *nv_connector = - nouveau_connector(conn_state->connector); - return nv50_outp_atomic_check_view(encoder, crtc_state, conn_state, - nv_connector->native_mode); -} - -/****************************************************************************** - * DAC - *****************************************************************************/ -static void -nv50_dac_disable(struct drm_encoder *encoder) -{ - struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder); - struct nv50_mast *mast = nv50_mast(encoder->dev); - const int or = nv_encoder->or; - u32 *push; - - if (nv_encoder->crtc) { - push = evo_wait(mast, 4); - if (push) { - if (nv50_vers(mast) < GF110_DISP_CORE_CHANNEL_DMA) { - evo_mthd(push, 0x0400 + (or * 0x080), 1); - evo_data(push, 0x00000000); - } else { - evo_mthd(push, 0x0180 + (or * 0x020), 1); - evo_data(push, 0x00000000); - } - evo_kick(push, mast); - } - } - - nv_encoder->crtc = NULL; - nv50_outp_release(nv_encoder); -} - -static void -nv50_dac_enable(struct drm_encoder *encoder) -{ - struct nv50_mast *mast = nv50_mast(encoder->dev); - struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder); - struct nouveau_crtc *nv_crtc = nouveau_crtc(encoder->crtc); - struct drm_display_mode *mode = &nv_crtc->base.state->adjusted_mode; - u32 *push; - - nv50_outp_acquire(nv_encoder); - - push = evo_wait(mast, 8); - if (push) { - if (nv50_vers(mast) < GF110_DISP_CORE_CHANNEL_DMA) { - u32 syncs = 0x00000000; - - if (mode->flags & DRM_MODE_FLAG_NHSYNC) - syncs |= 0x00000001; - if (mode->flags & DRM_MODE_FLAG_NVSYNC) - syncs |= 0x00000002; - - evo_mthd(push, 0x0400 + (nv_encoder->or * 0x080), 2); - evo_data(push, 1 << nv_crtc->index); - evo_data(push, syncs); - } else { - u32 magic = 0x31ec6000 | (nv_crtc->index << 25); - u32 syncs = 0x00000001; - - if (mode->flags & DRM_MODE_FLAG_NHSYNC) - syncs |= 0x00000008; - if (mode->flags & DRM_MODE_FLAG_NVSYNC) - syncs |= 0x00000010; - - if (mode->flags & DRM_MODE_FLAG_INTERLACE) - magic |= 0x00000001; - - evo_mthd(push, 0x0404 + (nv_crtc->index * 0x300), 2); - evo_data(push, syncs); - evo_data(push, magic); - evo_mthd(push, 0x0180 + (nv_encoder->or * 0x020), 1); - evo_data(push, 1 << nv_crtc->index); - } - - evo_kick(push, mast); - } - - nv_encoder->crtc = encoder->crtc; -} - -static enum drm_connector_status -nv50_dac_detect(struct drm_encoder *encoder, struct drm_connector *connector) -{ - struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder); - struct nv50_disp *disp = nv50_disp(encoder->dev); - struct { - struct nv50_disp_mthd_v1 base; - struct nv50_disp_dac_load_v0 load; - } args = { - .base.version = 1, - .base.method = NV50_DISP_MTHD_V1_DAC_LOAD, - .base.hasht = nv_encoder->dcb->hasht, - .base.hashm = nv_encoder->dcb->hashm, - }; - int ret; - - args.load.data = nouveau_drm(encoder->dev)->vbios.dactestval; - if (args.load.data == 0) - args.load.data = 340; - - ret = nvif_mthd(disp->disp, 0, &args, sizeof(args)); - if (ret || !args.load.load) - return connector_status_disconnected; - - return connector_status_connected; -} - -static const struct drm_encoder_helper_funcs -nv50_dac_help = { - .atomic_check = nv50_outp_atomic_check, - .enable = nv50_dac_enable, - .disable = nv50_dac_disable, - .detect = nv50_dac_detect -}; - -static void -nv50_dac_destroy(struct drm_encoder *encoder) -{ - drm_encoder_cleanup(encoder); - kfree(encoder); -} - -static const struct drm_encoder_funcs -nv50_dac_func = { - .destroy = nv50_dac_destroy, -}; - -static int -nv50_dac_create(struct drm_connector *connector, struct dcb_output *dcbe) -{ - struct nouveau_drm *drm = nouveau_drm(connector->dev); - struct nvkm_i2c *i2c = nvxx_i2c(&drm->client.device); - struct nvkm_i2c_bus *bus; - struct nouveau_encoder *nv_encoder; - struct drm_encoder *encoder; - int type = DRM_MODE_ENCODER_DAC; - - nv_encoder = kzalloc(sizeof(*nv_encoder), GFP_KERNEL); - if (!nv_encoder) - return -ENOMEM; - nv_encoder->dcb = dcbe; - - bus = nvkm_i2c_bus_find(i2c, dcbe->i2c_index); - if (bus) - nv_encoder->i2c = &bus->i2c; - - encoder = to_drm_encoder(nv_encoder); - encoder->possible_crtcs = dcbe->heads; - encoder->possible_clones = 0; - drm_encoder_init(connector->dev, encoder, &nv50_dac_func, type, - "dac-%04x-%04x", dcbe->hasht, dcbe->hashm); - drm_encoder_helper_add(encoder, &nv50_dac_help); - - drm_mode_connector_attach_encoder(connector, encoder); - return 0; -} - -/****************************************************************************** - * Audio - *****************************************************************************/ -static void -nv50_audio_disable(struct drm_encoder *encoder, struct nouveau_crtc *nv_crtc) -{ - struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder); - struct nv50_disp *disp = nv50_disp(encoder->dev); - struct { - struct nv50_disp_mthd_v1 base; - struct nv50_disp_sor_hda_eld_v0 eld; - } args = { - .base.version = 1, - .base.method = NV50_DISP_MTHD_V1_SOR_HDA_ELD, - .base.hasht = nv_encoder->dcb->hasht, - .base.hashm = (0xf0ff & nv_encoder->dcb->hashm) | - (0x0100 << nv_crtc->index), - }; - - nvif_mthd(disp->disp, 0, &args, sizeof(args)); -} - -static void -nv50_audio_enable(struct drm_encoder *encoder, struct drm_display_mode *mode) -{ - struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder); - struct nouveau_crtc *nv_crtc = nouveau_crtc(encoder->crtc); - struct nouveau_connector *nv_connector; - struct nv50_disp *disp = nv50_disp(encoder->dev); - struct __packed { - struct { - struct nv50_disp_mthd_v1 mthd; - struct nv50_disp_sor_hda_eld_v0 eld; - } base; - u8 data[sizeof(nv_connector->base.eld)]; - } args = { - .base.mthd.version = 1, - .base.mthd.method = NV50_DISP_MTHD_V1_SOR_HDA_ELD, - .base.mthd.hasht = nv_encoder->dcb->hasht, - .base.mthd.hashm = (0xf0ff & nv_encoder->dcb->hashm) | - (0x0100 << nv_crtc->index), - }; - - nv_connector = nouveau_encoder_connector_get(nv_encoder); - if (!drm_detect_monitor_audio(nv_connector->edid)) - return; - - memcpy(args.data, nv_connector->base.eld, sizeof(args.data)); - - nvif_mthd(disp->disp, 0, &args, - sizeof(args.base) + drm_eld_size(args.data)); -} - -/****************************************************************************** - * HDMI - *****************************************************************************/ -static void -nv50_hdmi_disable(struct drm_encoder *encoder, struct nouveau_crtc *nv_crtc) -{ - struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder); - struct nv50_disp *disp = nv50_disp(encoder->dev); - struct { - struct nv50_disp_mthd_v1 base; - struct nv50_disp_sor_hdmi_pwr_v0 pwr; - } args = { - .base.version = 1, - .base.method = NV50_DISP_MTHD_V1_SOR_HDMI_PWR, - .base.hasht = nv_encoder->dcb->hasht, - .base.hashm = (0xf0ff & nv_encoder->dcb->hashm) | - (0x0100 << nv_crtc->index), - }; - - nvif_mthd(disp->disp, 0, &args, sizeof(args)); -} - -static void -nv50_hdmi_enable(struct drm_encoder *encoder, struct drm_display_mode *mode) -{ - struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder); - struct nouveau_crtc *nv_crtc = nouveau_crtc(encoder->crtc); - struct nv50_disp *disp = nv50_disp(encoder->dev); - struct { - struct nv50_disp_mthd_v1 base; - struct nv50_disp_sor_hdmi_pwr_v0 pwr; - u8 infoframes[2 * 17]; /* two frames, up to 17 bytes each */ - } args = { - .base.version = 1, - .base.method = NV50_DISP_MTHD_V1_SOR_HDMI_PWR, - .base.hasht = nv_encoder->dcb->hasht, - .base.hashm = (0xf0ff & nv_encoder->dcb->hashm) | - (0x0100 << nv_crtc->index), - .pwr.state = 1, - .pwr.rekey = 56, /* binary driver, and tegra, constant */ - }; - struct nouveau_connector *nv_connector; - u32 max_ac_packet; - union hdmi_infoframe avi_frame; - union hdmi_infoframe vendor_frame; - int ret; - int size; - - nv_connector = nouveau_encoder_connector_get(nv_encoder); - if (!drm_detect_hdmi_monitor(nv_connector->edid)) - return; - - ret = drm_hdmi_avi_infoframe_from_display_mode(&avi_frame.avi, mode, - false); - if (!ret) { - /* We have an AVI InfoFrame, populate it to the display */ - args.pwr.avi_infoframe_length - = hdmi_infoframe_pack(&avi_frame, args.infoframes, 17); - } - - ret = drm_hdmi_vendor_infoframe_from_display_mode(&vendor_frame.vendor.hdmi, - &nv_connector->base, mode); - if (!ret) { - /* We have a Vendor InfoFrame, populate it to the display */ - args.pwr.vendor_infoframe_length - = hdmi_infoframe_pack(&vendor_frame, - args.infoframes - + args.pwr.avi_infoframe_length, - 17); - } - - max_ac_packet = mode->htotal - mode->hdisplay; - max_ac_packet -= args.pwr.rekey; - max_ac_packet -= 18; /* constant from tegra */ - args.pwr.max_ac_packet = max_ac_packet / 32; - - size = sizeof(args.base) - + sizeof(args.pwr) - + args.pwr.avi_infoframe_length - + args.pwr.vendor_infoframe_length; - nvif_mthd(disp->disp, 0, &args, size); - nv50_audio_enable(encoder, mode); -} - -/****************************************************************************** - * MST - *****************************************************************************/ -#define nv50_mstm(p) container_of((p), struct nv50_mstm, mgr) -#define nv50_mstc(p) container_of((p), struct nv50_mstc, connector) -#define nv50_msto(p) container_of((p), struct nv50_msto, encoder) - -struct nv50_mstm { - struct nouveau_encoder *outp; - - struct drm_dp_mst_topology_mgr mgr; - struct nv50_msto *msto[4]; - - bool modified; - bool disabled; - int links; -}; - -struct nv50_mstc { - struct nv50_mstm *mstm; - struct drm_dp_mst_port *port; - struct drm_connector connector; - - struct drm_display_mode *native; - struct edid *edid; - - int pbn; -}; - -struct nv50_msto { - struct drm_encoder encoder; - - struct nv50_head *head; - struct nv50_mstc *mstc; - bool disabled; -}; - -static struct drm_dp_payload * -nv50_msto_payload(struct nv50_msto *msto) -{ - struct nouveau_drm *drm = nouveau_drm(msto->encoder.dev); - struct nv50_mstc *mstc = msto->mstc; - struct nv50_mstm *mstm = mstc->mstm; - int vcpi = mstc->port->vcpi.vcpi, i; - - NV_ATOMIC(drm, "%s: vcpi %d\n", msto->encoder.name, vcpi); - for (i = 0; i < mstm->mgr.max_payloads; i++) { - struct drm_dp_payload *payload = &mstm->mgr.payloads[i]; - NV_ATOMIC(drm, "%s: %d: vcpi %d start 0x%02x slots 0x%02x\n", - mstm->outp->base.base.name, i, payload->vcpi, - payload->start_slot, payload->num_slots); - } - - for (i = 0; i < mstm->mgr.max_payloads; i++) { - struct drm_dp_payload *payload = &mstm->mgr.payloads[i]; - if (payload->vcpi == vcpi) - return payload; - } - - return NULL; -} - -static void -nv50_msto_cleanup(struct nv50_msto *msto) -{ - struct nouveau_drm *drm = nouveau_drm(msto->encoder.dev); - struct nv50_mstc *mstc = msto->mstc; - struct nv50_mstm *mstm = mstc->mstm; - - NV_ATOMIC(drm, "%s: msto cleanup\n", msto->encoder.name); - if (mstc->port && mstc->port->vcpi.vcpi > 0 && !nv50_msto_payload(msto)) - drm_dp_mst_deallocate_vcpi(&mstm->mgr, mstc->port); - if (msto->disabled) { - msto->mstc = NULL; - msto->head = NULL; - msto->disabled = false; - } -} - -static void -nv50_msto_prepare(struct nv50_msto *msto) -{ - struct nouveau_drm *drm = nouveau_drm(msto->encoder.dev); - struct nv50_mstc *mstc = msto->mstc; - struct nv50_mstm *mstm = mstc->mstm; - struct { - struct nv50_disp_mthd_v1 base; - struct nv50_disp_sor_dp_mst_vcpi_v0 vcpi; - } args = { - .base.version = 1, - .base.method = NV50_DISP_MTHD_V1_SOR_DP_MST_VCPI, - .base.hasht = mstm->outp->dcb->hasht, - .base.hashm = (0xf0ff & mstm->outp->dcb->hashm) | - (0x0100 << msto->head->base.index), - }; - - NV_ATOMIC(drm, "%s: msto prepare\n", msto->encoder.name); - if (mstc->port && mstc->port->vcpi.vcpi > 0) { - struct drm_dp_payload *payload = nv50_msto_payload(msto); - if (payload) { - args.vcpi.start_slot = payload->start_slot; - args.vcpi.num_slots = payload->num_slots; - args.vcpi.pbn = mstc->port->vcpi.pbn; - args.vcpi.aligned_pbn = mstc->port->vcpi.aligned_pbn; - } - } - - NV_ATOMIC(drm, "%s: %s: %02x %02x %04x %04x\n", - msto->encoder.name, msto->head->base.base.name, - args.vcpi.start_slot, args.vcpi.num_slots, - args.vcpi.pbn, args.vcpi.aligned_pbn); - nvif_mthd(&drm->display->disp, 0, &args, sizeof(args)); -} - -static int -nv50_msto_atomic_check(struct drm_encoder *encoder, - struct drm_crtc_state *crtc_state, - struct drm_connector_state *conn_state) -{ - struct nv50_mstc *mstc = nv50_mstc(conn_state->connector); - struct nv50_mstm *mstm = mstc->mstm; - int bpp = conn_state->connector->display_info.bpc * 3; - int slots; - - mstc->pbn = drm_dp_calc_pbn_mode(crtc_state->adjusted_mode.clock, bpp); - - slots = drm_dp_find_vcpi_slots(&mstm->mgr, mstc->pbn); - if (slots < 0) - return slots; - - return nv50_outp_atomic_check_view(encoder, crtc_state, conn_state, - mstc->native); -} - -static void -nv50_msto_enable(struct drm_encoder *encoder) -{ - struct nv50_head *head = nv50_head(encoder->crtc); - struct nv50_msto *msto = nv50_msto(encoder); - struct nv50_mstc *mstc = NULL; - struct nv50_mstm *mstm = NULL; - struct drm_connector *connector; - struct drm_connector_list_iter conn_iter; - u8 proto, depth; - int slots; - bool r; - - drm_connector_list_iter_begin(encoder->dev, &conn_iter); - drm_for_each_connector_iter(connector, &conn_iter) { - if (connector->state->best_encoder == &msto->encoder) { - mstc = nv50_mstc(connector); - mstm = mstc->mstm; - break; - } - } - drm_connector_list_iter_end(&conn_iter); - - if (WARN_ON(!mstc)) - return; - - slots = drm_dp_find_vcpi_slots(&mstm->mgr, mstc->pbn); - r = drm_dp_mst_allocate_vcpi(&mstm->mgr, mstc->port, mstc->pbn, slots); - WARN_ON(!r); - - if (!mstm->links++) - nv50_outp_acquire(mstm->outp); - - if (mstm->outp->link & 1) - proto = 0x8; - else - proto = 0x9; - - switch (mstc->connector.display_info.bpc) { - case 6: depth = 0x2; break; - case 8: depth = 0x5; break; - case 10: - default: depth = 0x6; break; - } - - mstm->outp->update(mstm->outp, head->base.index, - &head->base.base.state->adjusted_mode, proto, depth); - - msto->head = head; - msto->mstc = mstc; - mstm->modified = true; -} - -static void -nv50_msto_disable(struct drm_encoder *encoder) -{ - struct nv50_msto *msto = nv50_msto(encoder); - struct nv50_mstc *mstc = msto->mstc; - struct nv50_mstm *mstm = mstc->mstm; - - if (mstc->port) - drm_dp_mst_reset_vcpi_slots(&mstm->mgr, mstc->port); - - mstm->outp->update(mstm->outp, msto->head->base.index, NULL, 0, 0); - mstm->modified = true; - if (!--mstm->links) - mstm->disabled = true; - msto->disabled = true; -} - -static const struct drm_encoder_helper_funcs -nv50_msto_help = { - .disable = nv50_msto_disable, - .enable = nv50_msto_enable, - .atomic_check = nv50_msto_atomic_check, -}; - -static void -nv50_msto_destroy(struct drm_encoder *encoder) -{ - struct nv50_msto *msto = nv50_msto(encoder); - drm_encoder_cleanup(&msto->encoder); - kfree(msto); -} - -static const struct drm_encoder_funcs -nv50_msto = { - .destroy = nv50_msto_destroy, -}; - -static int -nv50_msto_new(struct drm_device *dev, u32 heads, const char *name, int id, - struct nv50_msto **pmsto) -{ - struct nv50_msto *msto; - int ret; - - if (!(msto = *pmsto = kzalloc(sizeof(*msto), GFP_KERNEL))) - return -ENOMEM; - - ret = drm_encoder_init(dev, &msto->encoder, &nv50_msto, - DRM_MODE_ENCODER_DPMST, "%s-mst-%d", name, id); - if (ret) { - kfree(*pmsto); - *pmsto = NULL; - return ret; - } - - drm_encoder_helper_add(&msto->encoder, &nv50_msto_help); - msto->encoder.possible_crtcs = heads; - return 0; -} - -static struct drm_encoder * -nv50_mstc_atomic_best_encoder(struct drm_connector *connector, - struct drm_connector_state *connector_state) -{ - struct nv50_head *head = nv50_head(connector_state->crtc); - struct nv50_mstc *mstc = nv50_mstc(connector); - if (mstc->port) { - struct nv50_mstm *mstm = mstc->mstm; - return &mstm->msto[head->base.index]->encoder; - } - return NULL; -} - -static struct drm_encoder * -nv50_mstc_best_encoder(struct drm_connector *connector) -{ - struct nv50_mstc *mstc = nv50_mstc(connector); - if (mstc->port) { - struct nv50_mstm *mstm = mstc->mstm; - return &mstm->msto[0]->encoder; - } - return NULL; -} - -static enum drm_mode_status -nv50_mstc_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode) -{ - return MODE_OK; -} - -static int -nv50_mstc_get_modes(struct drm_connector *connector) -{ - struct nv50_mstc *mstc = nv50_mstc(connector); - int ret = 0; - - mstc->edid = drm_dp_mst_get_edid(&mstc->connector, mstc->port->mgr, mstc->port); - drm_mode_connector_update_edid_property(&mstc->connector, mstc->edid); - if (mstc->edid) - ret = drm_add_edid_modes(&mstc->connector, mstc->edid); - - if (!mstc->connector.display_info.bpc) - mstc->connector.display_info.bpc = 8; - - if (mstc->native) - drm_mode_destroy(mstc->connector.dev, mstc->native); - mstc->native = nouveau_conn_native_mode(&mstc->connector); - return ret; -} - -static const struct drm_connector_helper_funcs -nv50_mstc_help = { - .get_modes = nv50_mstc_get_modes, - .mode_valid = nv50_mstc_mode_valid, - .best_encoder = nv50_mstc_best_encoder, - .atomic_best_encoder = nv50_mstc_atomic_best_encoder, -}; - -static enum drm_connector_status -nv50_mstc_detect(struct drm_connector *connector, bool force) -{ - struct nv50_mstc *mstc = nv50_mstc(connector); - if (!mstc->port) - return connector_status_disconnected; - return drm_dp_mst_detect_port(connector, mstc->port->mgr, mstc->port); -} - -static void -nv50_mstc_destroy(struct drm_connector *connector) -{ - struct nv50_mstc *mstc = nv50_mstc(connector); - drm_connector_cleanup(&mstc->connector); - kfree(mstc); -} - -static const struct drm_connector_funcs -nv50_mstc = { - .reset = nouveau_conn_reset, - .detect = nv50_mstc_detect, - .fill_modes = drm_helper_probe_single_connector_modes, - .destroy = nv50_mstc_destroy, - .atomic_duplicate_state = nouveau_conn_atomic_duplicate_state, - .atomic_destroy_state = nouveau_conn_atomic_destroy_state, - .atomic_set_property = nouveau_conn_atomic_set_property, - .atomic_get_property = nouveau_conn_atomic_get_property, -}; - -static int -nv50_mstc_new(struct nv50_mstm *mstm, struct drm_dp_mst_port *port, - const char *path, struct nv50_mstc **pmstc) -{ - struct drm_device *dev = mstm->outp->base.base.dev; - struct nv50_mstc *mstc; - int ret, i; - - if (!(mstc = *pmstc = kzalloc(sizeof(*mstc), GFP_KERNEL))) - return -ENOMEM; - mstc->mstm = mstm; - mstc->port = port; - - ret = drm_connector_init(dev, &mstc->connector, &nv50_mstc, - DRM_MODE_CONNECTOR_DisplayPort); - if (ret) { - kfree(*pmstc); - *pmstc = NULL; - return ret; - } - - drm_connector_helper_add(&mstc->connector, &nv50_mstc_help); - - mstc->connector.funcs->reset(&mstc->connector); - nouveau_conn_attach_properties(&mstc->connector); - - for (i = 0; i < ARRAY_SIZE(mstm->msto) && mstm->msto[i]; i++) - drm_mode_connector_attach_encoder(&mstc->connector, &mstm->msto[i]->encoder); - - drm_object_attach_property(&mstc->connector.base, dev->mode_config.path_property, 0); - drm_object_attach_property(&mstc->connector.base, dev->mode_config.tile_property, 0); - drm_mode_connector_set_path_property(&mstc->connector, path); - return 0; -} - -static void -nv50_mstm_cleanup(struct nv50_mstm *mstm) -{ - struct nouveau_drm *drm = nouveau_drm(mstm->outp->base.base.dev); - struct drm_encoder *encoder; - int ret; - - NV_ATOMIC(drm, "%s: mstm cleanup\n", mstm->outp->base.base.name); - ret = drm_dp_check_act_status(&mstm->mgr); - - ret = drm_dp_update_payload_part2(&mstm->mgr); - - drm_for_each_encoder(encoder, mstm->outp->base.base.dev) { - if (encoder->encoder_type == DRM_MODE_ENCODER_DPMST) { - struct nv50_msto *msto = nv50_msto(encoder); - struct nv50_mstc *mstc = msto->mstc; - if (mstc && mstc->mstm == mstm) - nv50_msto_cleanup(msto); - } - } - - mstm->modified = false; -} - -static void -nv50_mstm_prepare(struct nv50_mstm *mstm) -{ - struct nouveau_drm *drm = nouveau_drm(mstm->outp->base.base.dev); - struct drm_encoder *encoder; - int ret; - - NV_ATOMIC(drm, "%s: mstm prepare\n", mstm->outp->base.base.name); - ret = drm_dp_update_payload_part1(&mstm->mgr); - - drm_for_each_encoder(encoder, mstm->outp->base.base.dev) { - if (encoder->encoder_type == DRM_MODE_ENCODER_DPMST) { - struct nv50_msto *msto = nv50_msto(encoder); - struct nv50_mstc *mstc = msto->mstc; - if (mstc && mstc->mstm == mstm) - nv50_msto_prepare(msto); - } - } - - if (mstm->disabled) { - if (!mstm->links) - nv50_outp_release(mstm->outp); - mstm->disabled = false; - } -} - -static void -nv50_mstm_hotplug(struct drm_dp_mst_topology_mgr *mgr) -{ - struct nv50_mstm *mstm = nv50_mstm(mgr); - drm_kms_helper_hotplug_event(mstm->outp->base.base.dev); -} - -static void -nv50_mstm_destroy_connector(struct drm_dp_mst_topology_mgr *mgr, - struct drm_connector *connector) -{ - struct nouveau_drm *drm = nouveau_drm(connector->dev); - struct nv50_mstc *mstc = nv50_mstc(connector); - - drm_connector_unregister(&mstc->connector); - - drm_modeset_lock_all(drm->dev); - drm_fb_helper_remove_one_connector(&drm->fbcon->helper, &mstc->connector); - mstc->port = NULL; - drm_modeset_unlock_all(drm->dev); - - drm_connector_unreference(&mstc->connector); -} - -static void -nv50_mstm_register_connector(struct drm_connector *connector) -{ - struct nouveau_drm *drm = nouveau_drm(connector->dev); - - drm_modeset_lock_all(drm->dev); - drm_fb_helper_add_one_connector(&drm->fbcon->helper, connector); - drm_modeset_unlock_all(drm->dev); - - drm_connector_register(connector); -} - -static struct drm_connector * -nv50_mstm_add_connector(struct drm_dp_mst_topology_mgr *mgr, - struct drm_dp_mst_port *port, const char *path) -{ - struct nv50_mstm *mstm = nv50_mstm(mgr); - struct nv50_mstc *mstc; - int ret; - - ret = nv50_mstc_new(mstm, port, path, &mstc); - if (ret) { - if (mstc) - mstc->connector.funcs->destroy(&mstc->connector); - return NULL; - } - - return &mstc->connector; -} - -static const struct drm_dp_mst_topology_cbs -nv50_mstm = { - .add_connector = nv50_mstm_add_connector, - .register_connector = nv50_mstm_register_connector, - .destroy_connector = nv50_mstm_destroy_connector, - .hotplug = nv50_mstm_hotplug, -}; - -void -nv50_mstm_service(struct nv50_mstm *mstm) -{ - struct drm_dp_aux *aux = mstm ? mstm->mgr.aux : NULL; - bool handled = true; - int ret; - u8 esi[8] = {}; - - if (!aux) - return; - - while (handled) { - ret = drm_dp_dpcd_read(aux, DP_SINK_COUNT_ESI, esi, 8); - if (ret != 8) { - drm_dp_mst_topology_mgr_set_mst(&mstm->mgr, false); - return; - } - - drm_dp_mst_hpd_irq(&mstm->mgr, esi, &handled); - if (!handled) - break; - - drm_dp_dpcd_write(aux, DP_SINK_COUNT_ESI + 1, &esi[1], 3); - } -} - -void -nv50_mstm_remove(struct nv50_mstm *mstm) -{ - if (mstm) - drm_dp_mst_topology_mgr_set_mst(&mstm->mgr, false); -} - -static int -nv50_mstm_enable(struct nv50_mstm *mstm, u8 dpcd, int state) -{ - struct nouveau_encoder *outp = mstm->outp; - struct { - struct nv50_disp_mthd_v1 base; - struct nv50_disp_sor_dp_mst_link_v0 mst; - } args = { - .base.version = 1, - .base.method = NV50_DISP_MTHD_V1_SOR_DP_MST_LINK, - .base.hasht = outp->dcb->hasht, - .base.hashm = outp->dcb->hashm, - .mst.state = state, - }; - struct nouveau_drm *drm = nouveau_drm(outp->base.base.dev); - struct nvif_object *disp = &drm->display->disp; - int ret; - - if (dpcd >= 0x12) { - ret = drm_dp_dpcd_readb(mstm->mgr.aux, DP_MSTM_CTRL, &dpcd); - if (ret < 0) - return ret; - - dpcd &= ~DP_MST_EN; - if (state) - dpcd |= DP_MST_EN; - - ret = drm_dp_dpcd_writeb(mstm->mgr.aux, DP_MSTM_CTRL, dpcd); - if (ret < 0) - return ret; - } - - return nvif_mthd(disp, 0, &args, sizeof(args)); -} - -int -nv50_mstm_detect(struct nv50_mstm *mstm, u8 dpcd[8], int allow) -{ - int ret, state = 0; - - if (!mstm) - return 0; - - if (dpcd[0] >= 0x12) { - ret = drm_dp_dpcd_readb(mstm->mgr.aux, DP_MSTM_CAP, &dpcd[1]); - if (ret < 0) - return ret; - - if (!(dpcd[1] & DP_MST_CAP)) - dpcd[0] = 0x11; - else - state = allow; - } - - ret = nv50_mstm_enable(mstm, dpcd[0], state); - if (ret) - return ret; - - ret = drm_dp_mst_topology_mgr_set_mst(&mstm->mgr, state); - if (ret) - return nv50_mstm_enable(mstm, dpcd[0], 0); - - return mstm->mgr.mst_state; -} - -static void -nv50_mstm_fini(struct nv50_mstm *mstm) -{ - if (mstm && mstm->mgr.mst_state) - drm_dp_mst_topology_mgr_suspend(&mstm->mgr); -} - -static void -nv50_mstm_init(struct nv50_mstm *mstm) -{ - if (mstm && mstm->mgr.mst_state) - drm_dp_mst_topology_mgr_resume(&mstm->mgr); -} - -static void -nv50_mstm_del(struct nv50_mstm **pmstm) -{ - struct nv50_mstm *mstm = *pmstm; - if (mstm) { - kfree(*pmstm); - *pmstm = NULL; - } -} - -static int -nv50_mstm_new(struct nouveau_encoder *outp, struct drm_dp_aux *aux, int aux_max, - int conn_base_id, struct nv50_mstm **pmstm) -{ - const int max_payloads = hweight8(outp->dcb->heads); - struct drm_device *dev = outp->base.base.dev; - struct nv50_mstm *mstm; - int ret, i; - u8 dpcd; - - /* This is a workaround for some monitors not functioning - * correctly in MST mode on initial module load. I think - * some bad interaction with the VBIOS may be responsible. - * - * A good ol' off and on again seems to work here ;) - */ - ret = drm_dp_dpcd_readb(aux, DP_DPCD_REV, &dpcd); - if (ret >= 0 && dpcd >= 0x12) - drm_dp_dpcd_writeb(aux, DP_MSTM_CTRL, 0); - - if (!(mstm = *pmstm = kzalloc(sizeof(*mstm), GFP_KERNEL))) - return -ENOMEM; - mstm->outp = outp; - mstm->mgr.cbs = &nv50_mstm; - - ret = drm_dp_mst_topology_mgr_init(&mstm->mgr, dev, aux, aux_max, - max_payloads, conn_base_id); - if (ret) - return ret; - - for (i = 0; i < max_payloads; i++) { - ret = nv50_msto_new(dev, outp->dcb->heads, outp->base.base.name, - i, &mstm->msto[i]); - if (ret) - return ret; - } - - return 0; -} - -/****************************************************************************** - * SOR - *****************************************************************************/ -static void -nv50_sor_update(struct nouveau_encoder *nv_encoder, u8 head, - struct drm_display_mode *mode, u8 proto, u8 depth) -{ - struct nv50_dmac *core = &nv50_mast(nv_encoder->base.base.dev)->base; - u32 *push; - - if (!mode) { - nv_encoder->ctrl &= ~BIT(head); - if (!(nv_encoder->ctrl & 0x0000000f)) - nv_encoder->ctrl = 0; - } else { - nv_encoder->ctrl |= proto << 8; - nv_encoder->ctrl |= BIT(head); - } - - if ((push = evo_wait(core, 6))) { - if (core->base.user.oclass < GF110_DISP_CORE_CHANNEL_DMA) { - if (mode) { - if (mode->flags & DRM_MODE_FLAG_NHSYNC) - nv_encoder->ctrl |= 0x00001000; - if (mode->flags & DRM_MODE_FLAG_NVSYNC) - nv_encoder->ctrl |= 0x00002000; - nv_encoder->ctrl |= depth << 16; - } - evo_mthd(push, 0x0600 + (nv_encoder->or * 0x40), 1); - } else { - if (mode) { - u32 magic = 0x31ec6000 | (head << 25); - u32 syncs = 0x00000001; - if (mode->flags & DRM_MODE_FLAG_NHSYNC) - syncs |= 0x00000008; - if (mode->flags & DRM_MODE_FLAG_NVSYNC) - syncs |= 0x00000010; - if (mode->flags & DRM_MODE_FLAG_INTERLACE) - magic |= 0x00000001; - - evo_mthd(push, 0x0404 + (head * 0x300), 2); - evo_data(push, syncs | (depth << 6)); - evo_data(push, magic); - } - evo_mthd(push, 0x0200 + (nv_encoder->or * 0x20), 1); - } - evo_data(push, nv_encoder->ctrl); - evo_kick(push, core); - } -} - -static void -nv50_sor_disable(struct drm_encoder *encoder) -{ - struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder); - struct nouveau_crtc *nv_crtc = nouveau_crtc(nv_encoder->crtc); - - nv_encoder->crtc = NULL; - - if (nv_crtc) { - struct nvkm_i2c_aux *aux = nv_encoder->aux; - u8 pwr; - - if (aux) { - int ret = nvkm_rdaux(aux, DP_SET_POWER, &pwr, 1); - if (ret == 0) { - pwr &= ~DP_SET_POWER_MASK; - pwr |= DP_SET_POWER_D3; - nvkm_wraux(aux, DP_SET_POWER, &pwr, 1); - } - } - - nv_encoder->update(nv_encoder, nv_crtc->index, NULL, 0, 0); - nv50_audio_disable(encoder, nv_crtc); - nv50_hdmi_disable(&nv_encoder->base.base, nv_crtc); - nv50_outp_release(nv_encoder); - } -} - -static void -nv50_sor_enable(struct drm_encoder *encoder) -{ - struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder); - struct nouveau_crtc *nv_crtc = nouveau_crtc(encoder->crtc); - struct drm_display_mode *mode = &nv_crtc->base.state->adjusted_mode; - struct { - struct nv50_disp_mthd_v1 base; - struct nv50_disp_sor_lvds_script_v0 lvds; - } lvds = { - .base.version = 1, - .base.method = NV50_DISP_MTHD_V1_SOR_LVDS_SCRIPT, - .base.hasht = nv_encoder->dcb->hasht, - .base.hashm = nv_encoder->dcb->hashm, - }; - struct nv50_disp *disp = nv50_disp(encoder->dev); - struct drm_device *dev = encoder->dev; - struct nouveau_drm *drm = nouveau_drm(dev); - struct nouveau_connector *nv_connector; - struct nvbios *bios = &drm->vbios; - u8 proto = 0xf; - u8 depth = 0x0; - - nv_connector = nouveau_encoder_connector_get(nv_encoder); - nv_encoder->crtc = encoder->crtc; - nv50_outp_acquire(nv_encoder); - - switch (nv_encoder->dcb->type) { - case DCB_OUTPUT_TMDS: - if (nv_encoder->link & 1) { - proto = 0x1; - /* Only enable dual-link if: - * - Need to (i.e. rate > 165MHz) - * - DCB says we can - * - Not an HDMI monitor, since there's no dual-link - * on HDMI. - */ - if (mode->clock >= 165000 && - nv_encoder->dcb->duallink_possible && - !drm_detect_hdmi_monitor(nv_connector->edid)) - proto |= 0x4; - } else { - proto = 0x2; - } - - nv50_hdmi_enable(&nv_encoder->base.base, mode); - break; - case DCB_OUTPUT_LVDS: - proto = 0x0; - - if (bios->fp_no_ddc) { - if (bios->fp.dual_link) - lvds.lvds.script |= 0x0100; - if (bios->fp.if_is_24bit) - lvds.lvds.script |= 0x0200; - } else { - if (nv_connector->type == DCB_CONNECTOR_LVDS_SPWG) { - if (((u8 *)nv_connector->edid)[121] == 2) - lvds.lvds.script |= 0x0100; - } else - if (mode->clock >= bios->fp.duallink_transition_clk) { - lvds.lvds.script |= 0x0100; - } - - if (lvds.lvds.script & 0x0100) { - if (bios->fp.strapless_is_24bit & 2) - lvds.lvds.script |= 0x0200; - } else { - if (bios->fp.strapless_is_24bit & 1) - lvds.lvds.script |= 0x0200; - } - - if (nv_connector->base.display_info.bpc == 8) - lvds.lvds.script |= 0x0200; - } - - nvif_mthd(disp->disp, 0, &lvds, sizeof(lvds)); - break; - case DCB_OUTPUT_DP: - if (nv_connector->base.display_info.bpc == 6) - depth = 0x2; - else - if (nv_connector->base.display_info.bpc == 8) - depth = 0x5; - else - depth = 0x6; - - if (nv_encoder->link & 1) - proto = 0x8; - else - proto = 0x9; - - nv50_audio_enable(encoder, mode); - break; - default: - BUG(); - break; - } - - nv_encoder->update(nv_encoder, nv_crtc->index, mode, proto, depth); -} - -static const struct drm_encoder_helper_funcs -nv50_sor_help = { - .atomic_check = nv50_outp_atomic_check, - .enable = nv50_sor_enable, - .disable = nv50_sor_disable, -}; - -static void -nv50_sor_destroy(struct drm_encoder *encoder) -{ - struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder); - nv50_mstm_del(&nv_encoder->dp.mstm); - drm_encoder_cleanup(encoder); - kfree(encoder); -} - -static const struct drm_encoder_funcs -nv50_sor_func = { - .destroy = nv50_sor_destroy, -}; - -static int -nv50_sor_create(struct drm_connector *connector, struct dcb_output *dcbe) -{ - struct nouveau_connector *nv_connector = nouveau_connector(connector); - struct nouveau_drm *drm = nouveau_drm(connector->dev); - struct nvkm_i2c *i2c = nvxx_i2c(&drm->client.device); - struct nouveau_encoder *nv_encoder; - struct drm_encoder *encoder; - int type, ret; - - switch (dcbe->type) { - case DCB_OUTPUT_LVDS: type = DRM_MODE_ENCODER_LVDS; break; - case DCB_OUTPUT_TMDS: - case DCB_OUTPUT_DP: - default: - type = DRM_MODE_ENCODER_TMDS; - break; - } - - nv_encoder = kzalloc(sizeof(*nv_encoder), GFP_KERNEL); - if (!nv_encoder) - return -ENOMEM; - nv_encoder->dcb = dcbe; - nv_encoder->update = nv50_sor_update; - - encoder = to_drm_encoder(nv_encoder); - encoder->possible_crtcs = dcbe->heads; - encoder->possible_clones = 0; - drm_encoder_init(connector->dev, encoder, &nv50_sor_func, type, - "sor-%04x-%04x", dcbe->hasht, dcbe->hashm); - drm_encoder_helper_add(encoder, &nv50_sor_help); - - drm_mode_connector_attach_encoder(connector, encoder); - - if (dcbe->type == DCB_OUTPUT_DP) { - struct nv50_disp *disp = nv50_disp(encoder->dev); - struct nvkm_i2c_aux *aux = - nvkm_i2c_aux_find(i2c, dcbe->i2c_index); - if (aux) { - if (disp->disp->oclass < GF110_DISP) { - /* HW has no support for address-only - * transactions, so we're required to - * use custom I2C-over-AUX code. - */ - nv_encoder->i2c = &aux->i2c; - } else { - nv_encoder->i2c = &nv_connector->aux.ddc; - } - nv_encoder->aux = aux; - } - - /*TODO: Use DP Info Table to check for support. */ - if (disp->disp->oclass >= GF110_DISP) { - ret = nv50_mstm_new(nv_encoder, &nv_connector->aux, 16, - nv_connector->base.base.id, - &nv_encoder->dp.mstm); - if (ret) - return ret; - } - } else { - struct nvkm_i2c_bus *bus = - nvkm_i2c_bus_find(i2c, dcbe->i2c_index); - if (bus) - nv_encoder->i2c = &bus->i2c; - } - - return 0; -} - -/****************************************************************************** - * PIOR - *****************************************************************************/ -static int -nv50_pior_atomic_check(struct drm_encoder *encoder, - struct drm_crtc_state *crtc_state, - struct drm_connector_state *conn_state) -{ - int ret = nv50_outp_atomic_check(encoder, crtc_state, conn_state); - if (ret) - return ret; - crtc_state->adjusted_mode.clock *= 2; - return 0; -} - -static void -nv50_pior_disable(struct drm_encoder *encoder) -{ - struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder); - struct nv50_mast *mast = nv50_mast(encoder->dev); - const int or = nv_encoder->or; - u32 *push; - - if (nv_encoder->crtc) { - push = evo_wait(mast, 4); - if (push) { - if (nv50_vers(mast) < GF110_DISP_CORE_CHANNEL_DMA) { - evo_mthd(push, 0x0700 + (or * 0x040), 1); - evo_data(push, 0x00000000); - } - evo_kick(push, mast); - } - } - - nv_encoder->crtc = NULL; - nv50_outp_release(nv_encoder); -} - -static void -nv50_pior_enable(struct drm_encoder *encoder) -{ - struct nv50_mast *mast = nv50_mast(encoder->dev); - struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder); - struct nouveau_crtc *nv_crtc = nouveau_crtc(encoder->crtc); - struct nouveau_connector *nv_connector; - struct drm_display_mode *mode = &nv_crtc->base.state->adjusted_mode; - u8 owner = 1 << nv_crtc->index; - u8 proto, depth; - u32 *push; - - nv50_outp_acquire(nv_encoder); - - nv_connector = nouveau_encoder_connector_get(nv_encoder); - switch (nv_connector->base.display_info.bpc) { - case 10: depth = 0x6; break; - case 8: depth = 0x5; break; - case 6: depth = 0x2; break; - default: depth = 0x0; break; - } - - switch (nv_encoder->dcb->type) { - case DCB_OUTPUT_TMDS: - case DCB_OUTPUT_DP: - proto = 0x0; - break; - default: - BUG(); - break; - } - - push = evo_wait(mast, 8); - if (push) { - if (nv50_vers(mast) < GF110_DISP_CORE_CHANNEL_DMA) { - u32 ctrl = (depth << 16) | (proto << 8) | owner; - if (mode->flags & DRM_MODE_FLAG_NHSYNC) - ctrl |= 0x00001000; - if (mode->flags & DRM_MODE_FLAG_NVSYNC) - ctrl |= 0x00002000; - evo_mthd(push, 0x0700 + (nv_encoder->or * 0x040), 1); - evo_data(push, ctrl); - } - - evo_kick(push, mast); - } - - nv_encoder->crtc = encoder->crtc; -} - -static const struct drm_encoder_helper_funcs -nv50_pior_help = { - .atomic_check = nv50_pior_atomic_check, - .enable = nv50_pior_enable, - .disable = nv50_pior_disable, -}; - -static void -nv50_pior_destroy(struct drm_encoder *encoder) -{ - drm_encoder_cleanup(encoder); - kfree(encoder); -} - -static const struct drm_encoder_funcs -nv50_pior_func = { - .destroy = nv50_pior_destroy, -}; - -static int -nv50_pior_create(struct drm_connector *connector, struct dcb_output *dcbe) -{ - struct nouveau_connector *nv_connector = nouveau_connector(connector); - struct nouveau_drm *drm = nouveau_drm(connector->dev); - struct nvkm_i2c *i2c = nvxx_i2c(&drm->client.device); - struct nvkm_i2c_bus *bus = NULL; - struct nvkm_i2c_aux *aux = NULL; - struct i2c_adapter *ddc; - struct nouveau_encoder *nv_encoder; - struct drm_encoder *encoder; - int type; - - switch (dcbe->type) { - case DCB_OUTPUT_TMDS: - bus = nvkm_i2c_bus_find(i2c, NVKM_I2C_BUS_EXT(dcbe->extdev)); - ddc = bus ? &bus->i2c : NULL; - type = DRM_MODE_ENCODER_TMDS; - break; - case DCB_OUTPUT_DP: - aux = nvkm_i2c_aux_find(i2c, NVKM_I2C_AUX_EXT(dcbe->extdev)); - ddc = aux ? &nv_connector->aux.ddc : NULL; - type = DRM_MODE_ENCODER_TMDS; - break; - default: - return -ENODEV; - } - - nv_encoder = kzalloc(sizeof(*nv_encoder), GFP_KERNEL); - if (!nv_encoder) - return -ENOMEM; - nv_encoder->dcb = dcbe; - nv_encoder->i2c = ddc; - nv_encoder->aux = aux; - - encoder = to_drm_encoder(nv_encoder); - encoder->possible_crtcs = dcbe->heads; - encoder->possible_clones = 0; - drm_encoder_init(connector->dev, encoder, &nv50_pior_func, type, - "pior-%04x-%04x", dcbe->hasht, dcbe->hashm); - drm_encoder_helper_add(encoder, &nv50_pior_help); - - drm_mode_connector_attach_encoder(connector, encoder); - return 0; -} - -/****************************************************************************** - * Atomic - *****************************************************************************/ - -static void -nv50_disp_atomic_commit_core(struct nouveau_drm *drm, u32 interlock) -{ - struct nv50_disp *disp = nv50_disp(drm->dev); - struct nv50_dmac *core = &disp->mast.base; - struct nv50_mstm *mstm; - struct drm_encoder *encoder; - u32 *push; - - NV_ATOMIC(drm, "commit core %08x\n", interlock); - - drm_for_each_encoder(encoder, drm->dev) { - if (encoder->encoder_type != DRM_MODE_ENCODER_DPMST) { - mstm = nouveau_encoder(encoder)->dp.mstm; - if (mstm && mstm->modified) - nv50_mstm_prepare(mstm); - } - } - - if ((push = evo_wait(core, 5))) { - evo_mthd(push, 0x0084, 1); - evo_data(push, 0x80000000); - evo_mthd(push, 0x0080, 2); - evo_data(push, interlock); - evo_data(push, 0x00000000); - nouveau_bo_wr32(disp->sync, 0, 0x00000000); - evo_kick(push, core); - if (nvif_msec(&drm->client.device, 2000ULL, - if (nouveau_bo_rd32(disp->sync, 0)) - break; - usleep_range(1, 2); - ) < 0) - NV_ERROR(drm, "EVO timeout\n"); - } - - drm_for_each_encoder(encoder, drm->dev) { - if (encoder->encoder_type != DRM_MODE_ENCODER_DPMST) { - mstm = nouveau_encoder(encoder)->dp.mstm; - if (mstm && mstm->modified) - nv50_mstm_cleanup(mstm); - } - } -} - -static void -nv50_disp_atomic_commit_tail(struct drm_atomic_state *state) -{ - struct drm_device *dev = state->dev; - struct drm_crtc_state *new_crtc_state, *old_crtc_state; - struct drm_crtc *crtc; - struct drm_plane_state *new_plane_state; - struct drm_plane *plane; - struct nouveau_drm *drm = nouveau_drm(dev); - struct nv50_disp *disp = nv50_disp(dev); - struct nv50_atom *atom = nv50_atom(state); - struct nv50_outp_atom *outp, *outt; - u32 interlock_core = 0; - u32 interlock_chan = 0; - int i; - - NV_ATOMIC(drm, "commit %d %d\n", atom->lock_core, atom->flush_disable); - drm_atomic_helper_wait_for_fences(dev, state, false); - drm_atomic_helper_wait_for_dependencies(state); - drm_atomic_helper_update_legacy_modeset_state(dev, state); - - if (atom->lock_core) - mutex_lock(&disp->mutex); - - /* Disable head(s). */ - for_each_oldnew_crtc_in_state(state, crtc, old_crtc_state, new_crtc_state, i) { - struct nv50_head_atom *asyh = nv50_head_atom(new_crtc_state); - struct nv50_head *head = nv50_head(crtc); - - NV_ATOMIC(drm, "%s: clr %04x (set %04x)\n", crtc->name, - asyh->clr.mask, asyh->set.mask); - if (old_crtc_state->active && !new_crtc_state->active) - drm_crtc_vblank_off(crtc); - - if (asyh->clr.mask) { - nv50_head_flush_clr(head, asyh, atom->flush_disable); - interlock_core |= 1; - } - } - - /* Disable plane(s). */ - for_each_new_plane_in_state(state, plane, new_plane_state, i) { - struct nv50_wndw_atom *asyw = nv50_wndw_atom(new_plane_state); - struct nv50_wndw *wndw = nv50_wndw(plane); - - NV_ATOMIC(drm, "%s: clr %02x (set %02x)\n", plane->name, - asyw->clr.mask, asyw->set.mask); - if (!asyw->clr.mask) - continue; - - interlock_chan |= nv50_wndw_flush_clr(wndw, interlock_core, - atom->flush_disable, - asyw); - } - - /* Disable output path(s). */ - list_for_each_entry(outp, &atom->outp, head) { - const struct drm_encoder_helper_funcs *help; - struct drm_encoder *encoder; - - encoder = outp->encoder; - help = encoder->helper_private; - - NV_ATOMIC(drm, "%s: clr %02x (set %02x)\n", encoder->name, - outp->clr.mask, outp->set.mask); - - if (outp->clr.mask) { - help->disable(encoder); - interlock_core |= 1; - if (outp->flush_disable) { - nv50_disp_atomic_commit_core(drm, interlock_chan); - interlock_core = 0; - interlock_chan = 0; - } - } - } - - /* Flush disable. */ - if (interlock_core) { - if (atom->flush_disable) { - nv50_disp_atomic_commit_core(drm, interlock_chan); - interlock_core = 0; - interlock_chan = 0; - } - } - - /* Update output path(s). */ - list_for_each_entry_safe(outp, outt, &atom->outp, head) { - const struct drm_encoder_helper_funcs *help; - struct drm_encoder *encoder; - - encoder = outp->encoder; - help = encoder->helper_private; - - NV_ATOMIC(drm, "%s: set %02x (clr %02x)\n", encoder->name, - outp->set.mask, outp->clr.mask); - - if (outp->set.mask) { - help->enable(encoder); - interlock_core = 1; - } - - list_del(&outp->head); - kfree(outp); - } - - /* Update head(s). */ - for_each_oldnew_crtc_in_state(state, crtc, old_crtc_state, new_crtc_state, i) { - struct nv50_head_atom *asyh = nv50_head_atom(new_crtc_state); - struct nv50_head *head = nv50_head(crtc); - - NV_ATOMIC(drm, "%s: set %04x (clr %04x)\n", crtc->name, - asyh->set.mask, asyh->clr.mask); - - if (asyh->set.mask) { - nv50_head_flush_set(head, asyh); - interlock_core = 1; - } - - if (new_crtc_state->active) { - if (!old_crtc_state->active) - drm_crtc_vblank_on(crtc); - if (new_crtc_state->event) - drm_crtc_vblank_get(crtc); - } - } - - /* Update plane(s). */ - for_each_new_plane_in_state(state, plane, new_plane_state, i) { - struct nv50_wndw_atom *asyw = nv50_wndw_atom(new_plane_state); - struct nv50_wndw *wndw = nv50_wndw(plane); - - NV_ATOMIC(drm, "%s: set %02x (clr %02x)\n", plane->name, - asyw->set.mask, asyw->clr.mask); - if ( !asyw->set.mask && - (!asyw->clr.mask || atom->flush_disable)) - continue; - - interlock_chan |= nv50_wndw_flush_set(wndw, interlock_core, asyw); - } - - /* Flush update. */ - if (interlock_core) { - if (!interlock_chan && atom->state.legacy_cursor_update) { - u32 *push = evo_wait(&disp->mast, 2); - if (push) { - evo_mthd(push, 0x0080, 1); - evo_data(push, 0x00000000); - evo_kick(push, &disp->mast); - } - } else { - nv50_disp_atomic_commit_core(drm, interlock_chan); - } - } - - if (atom->lock_core) - mutex_unlock(&disp->mutex); - - /* Wait for HW to signal completion. */ - for_each_new_plane_in_state(state, plane, new_plane_state, i) { - struct nv50_wndw_atom *asyw = nv50_wndw_atom(new_plane_state); - struct nv50_wndw *wndw = nv50_wndw(plane); - int ret = nv50_wndw_wait_armed(wndw, asyw); - if (ret) - NV_ERROR(drm, "%s: timeout\n", plane->name); - } - - for_each_new_crtc_in_state(state, crtc, new_crtc_state, i) { - if (new_crtc_state->event) { - unsigned long flags; - /* Get correct count/ts if racing with vblank irq */ - if (new_crtc_state->active) - drm_crtc_accurate_vblank_count(crtc); - spin_lock_irqsave(&crtc->dev->event_lock, flags); - drm_crtc_send_vblank_event(crtc, new_crtc_state->event); - spin_unlock_irqrestore(&crtc->dev->event_lock, flags); - - new_crtc_state->event = NULL; - if (new_crtc_state->active) - drm_crtc_vblank_put(crtc); - } - } - - drm_atomic_helper_commit_hw_done(state); - drm_atomic_helper_cleanup_planes(dev, state); - drm_atomic_helper_commit_cleanup_done(state); - drm_atomic_state_put(state); -} - -static void -nv50_disp_atomic_commit_work(struct work_struct *work) -{ - struct drm_atomic_state *state = - container_of(work, typeof(*state), commit_work); - nv50_disp_atomic_commit_tail(state); -} - -static int -nv50_disp_atomic_commit(struct drm_device *dev, - struct drm_atomic_state *state, bool nonblock) -{ - struct nouveau_drm *drm = nouveau_drm(dev); - struct nv50_disp *disp = nv50_disp(dev); - struct drm_plane_state *new_plane_state; - struct drm_plane *plane; - struct drm_crtc *crtc; - bool active = false; - int ret, i; - - ret = pm_runtime_get_sync(dev->dev); - if (ret < 0 && ret != -EACCES) - return ret; - - ret = drm_atomic_helper_setup_commit(state, nonblock); - if (ret) - goto done; - - INIT_WORK(&state->commit_work, nv50_disp_atomic_commit_work); - - ret = drm_atomic_helper_prepare_planes(dev, state); - if (ret) - goto done; - - if (!nonblock) { - ret = drm_atomic_helper_wait_for_fences(dev, state, true); - if (ret) - goto err_cleanup; - } - - ret = drm_atomic_helper_swap_state(state, true); - if (ret) - goto err_cleanup; - - for_each_new_plane_in_state(state, plane, new_plane_state, i) { - struct nv50_wndw_atom *asyw = nv50_wndw_atom(new_plane_state); - struct nv50_wndw *wndw = nv50_wndw(plane); - - if (asyw->set.image) { - asyw->ntfy.handle = wndw->dmac->sync.handle; - asyw->ntfy.offset = wndw->ntfy; - asyw->ntfy.awaken = false; - asyw->set.ntfy = true; - nouveau_bo_wr32(disp->sync, wndw->ntfy / 4, 0x00000000); - wndw->ntfy ^= 0x10; - } - } - - drm_atomic_state_get(state); - - if (nonblock) - queue_work(system_unbound_wq, &state->commit_work); - else - nv50_disp_atomic_commit_tail(state); - - drm_for_each_crtc(crtc, dev) { - if (crtc->state->enable) { - if (!drm->have_disp_power_ref) { - drm->have_disp_power_ref = true; - return 0; - } - active = true; - break; - } - } - - if (!active && drm->have_disp_power_ref) { - pm_runtime_put_autosuspend(dev->dev); - drm->have_disp_power_ref = false; - } - -err_cleanup: - if (ret) - drm_atomic_helper_cleanup_planes(dev, state); -done: - pm_runtime_put_autosuspend(dev->dev); - return ret; -} - -static struct nv50_outp_atom * -nv50_disp_outp_atomic_add(struct nv50_atom *atom, struct drm_encoder *encoder) -{ - struct nv50_outp_atom *outp; - - list_for_each_entry(outp, &atom->outp, head) { - if (outp->encoder == encoder) - return outp; - } - - outp = kzalloc(sizeof(*outp), GFP_KERNEL); - if (!outp) - return ERR_PTR(-ENOMEM); - - list_add(&outp->head, &atom->outp); - outp->encoder = encoder; - return outp; -} - -static int -nv50_disp_outp_atomic_check_clr(struct nv50_atom *atom, - struct drm_connector_state *old_connector_state) -{ - struct drm_encoder *encoder = old_connector_state->best_encoder; - struct drm_crtc_state *old_crtc_state, *new_crtc_state; - struct drm_crtc *crtc; - struct nv50_outp_atom *outp; - - if (!(crtc = old_connector_state->crtc)) - return 0; - - old_crtc_state = drm_atomic_get_old_crtc_state(&atom->state, crtc); - new_crtc_state = drm_atomic_get_new_crtc_state(&atom->state, crtc); - if (old_crtc_state->active && drm_atomic_crtc_needs_modeset(new_crtc_state)) { - outp = nv50_disp_outp_atomic_add(atom, encoder); - if (IS_ERR(outp)) - return PTR_ERR(outp); - - if (outp->encoder->encoder_type == DRM_MODE_ENCODER_DPMST) { - outp->flush_disable = true; - atom->flush_disable = true; - } - outp->clr.ctrl = true; - atom->lock_core = true; - } - - return 0; -} - -static int -nv50_disp_outp_atomic_check_set(struct nv50_atom *atom, - struct drm_connector_state *connector_state) -{ - struct drm_encoder *encoder = connector_state->best_encoder; - struct drm_crtc_state *new_crtc_state; - struct drm_crtc *crtc; - struct nv50_outp_atom *outp; - - if (!(crtc = connector_state->crtc)) - return 0; - - new_crtc_state = drm_atomic_get_new_crtc_state(&atom->state, crtc); - if (new_crtc_state->active && drm_atomic_crtc_needs_modeset(new_crtc_state)) { - outp = nv50_disp_outp_atomic_add(atom, encoder); - if (IS_ERR(outp)) - return PTR_ERR(outp); - - outp->set.ctrl = true; - atom->lock_core = true; - } - - return 0; -} - -static int -nv50_disp_atomic_check(struct drm_device *dev, struct drm_atomic_state *state) -{ - struct nv50_atom *atom = nv50_atom(state); - struct drm_connector_state *old_connector_state, *new_connector_state; - struct drm_connector *connector; - int ret, i; - - ret = drm_atomic_helper_check(dev, state); - if (ret) - return ret; - - for_each_oldnew_connector_in_state(state, connector, old_connector_state, new_connector_state, i) { - ret = nv50_disp_outp_atomic_check_clr(atom, old_connector_state); - if (ret) - return ret; - - ret = nv50_disp_outp_atomic_check_set(atom, new_connector_state); - if (ret) - return ret; - } - - return 0; -} - -static void -nv50_disp_atomic_state_clear(struct drm_atomic_state *state) -{ - struct nv50_atom *atom = nv50_atom(state); - struct nv50_outp_atom *outp, *outt; - - list_for_each_entry_safe(outp, outt, &atom->outp, head) { - list_del(&outp->head); - kfree(outp); - } - - drm_atomic_state_default_clear(state); -} - -static void -nv50_disp_atomic_state_free(struct drm_atomic_state *state) -{ - struct nv50_atom *atom = nv50_atom(state); - drm_atomic_state_default_release(&atom->state); - kfree(atom); -} - -static struct drm_atomic_state * -nv50_disp_atomic_state_alloc(struct drm_device *dev) -{ - struct nv50_atom *atom; - if (!(atom = kzalloc(sizeof(*atom), GFP_KERNEL)) || - drm_atomic_state_init(dev, &atom->state) < 0) { - kfree(atom); - return NULL; - } - INIT_LIST_HEAD(&atom->outp); - return &atom->state; -} - -static const struct drm_mode_config_funcs -nv50_disp_func = { - .fb_create = nouveau_user_framebuffer_create, - .output_poll_changed = drm_fb_helper_output_poll_changed, - .atomic_check = nv50_disp_atomic_check, - .atomic_commit = nv50_disp_atomic_commit, - .atomic_state_alloc = nv50_disp_atomic_state_alloc, - .atomic_state_clear = nv50_disp_atomic_state_clear, - .atomic_state_free = nv50_disp_atomic_state_free, -}; - -/****************************************************************************** - * Init - *****************************************************************************/ - -void -nv50_display_fini(struct drm_device *dev) -{ - struct nouveau_encoder *nv_encoder; - struct drm_encoder *encoder; - struct drm_plane *plane; - - drm_for_each_plane(plane, dev) { - struct nv50_wndw *wndw = nv50_wndw(plane); - if (plane->funcs != &nv50_wndw) - continue; - nv50_wndw_fini(wndw); - } - - list_for_each_entry(encoder, &dev->mode_config.encoder_list, head) { - if (encoder->encoder_type != DRM_MODE_ENCODER_DPMST) { - nv_encoder = nouveau_encoder(encoder); - nv50_mstm_fini(nv_encoder->dp.mstm); - } - } -} - -int -nv50_display_init(struct drm_device *dev) -{ - struct drm_encoder *encoder; - struct drm_plane *plane; - u32 *push; - - push = evo_wait(nv50_mast(dev), 32); - if (!push) - return -EBUSY; - - evo_mthd(push, 0x0088, 1); - evo_data(push, nv50_mast(dev)->base.sync.handle); - evo_kick(push, nv50_mast(dev)); - - list_for_each_entry(encoder, &dev->mode_config.encoder_list, head) { - if (encoder->encoder_type != DRM_MODE_ENCODER_DPMST) { - struct nouveau_encoder *nv_encoder = - nouveau_encoder(encoder); - nv50_mstm_init(nv_encoder->dp.mstm); - } - } - - drm_for_each_plane(plane, dev) { - struct nv50_wndw *wndw = nv50_wndw(plane); - if (plane->funcs != &nv50_wndw) - continue; - nv50_wndw_init(wndw); - } - - return 0; -} - -void -nv50_display_destroy(struct drm_device *dev) -{ - struct nv50_disp *disp = nv50_disp(dev); - - nv50_dmac_destroy(&disp->mast.base, disp->disp); - - nouveau_bo_unmap(disp->sync); - if (disp->sync) - nouveau_bo_unpin(disp->sync); - nouveau_bo_ref(NULL, &disp->sync); - - nouveau_display(dev)->priv = NULL; - kfree(disp); -} - -MODULE_PARM_DESC(atomic, "Expose atomic ioctl (default: disabled)"); -static int nouveau_atomic = 0; -module_param_named(atomic, nouveau_atomic, int, 0400); - -int -nv50_display_create(struct drm_device *dev) -{ - struct nvif_device *device = &nouveau_drm(dev)->client.device; - struct nouveau_drm *drm = nouveau_drm(dev); - struct dcb_table *dcb = &drm->vbios.dcb; - struct drm_connector *connector, *tmp; - struct nv50_disp *disp; - struct dcb_output *dcbe; - int crtcs, ret, i; - - disp = kzalloc(sizeof(*disp), GFP_KERNEL); - if (!disp) - return -ENOMEM; - - mutex_init(&disp->mutex); - - nouveau_display(dev)->priv = disp; - nouveau_display(dev)->dtor = nv50_display_destroy; - nouveau_display(dev)->init = nv50_display_init; - nouveau_display(dev)->fini = nv50_display_fini; - disp->disp = &nouveau_display(dev)->disp; - dev->mode_config.funcs = &nv50_disp_func; - dev->driver->driver_features |= DRIVER_PREFER_XBGR_30BPP; - if (nouveau_atomic) - dev->driver->driver_features |= DRIVER_ATOMIC; - - /* small shared memory area we use for notifiers and semaphores */ - ret = nouveau_bo_new(&drm->client, 4096, 0x1000, TTM_PL_FLAG_VRAM, - 0, 0x0000, NULL, NULL, &disp->sync); - if (!ret) { - ret = nouveau_bo_pin(disp->sync, TTM_PL_FLAG_VRAM, true); - if (!ret) { - ret = nouveau_bo_map(disp->sync); - if (ret) - nouveau_bo_unpin(disp->sync); - } - if (ret) - nouveau_bo_ref(NULL, &disp->sync); - } - - if (ret) - goto out; - - /* allocate master evo channel */ - ret = nv50_core_create(device, disp->disp, disp->sync->bo.offset, - &disp->mast); - if (ret) - goto out; - - /* create crtc objects to represent the hw heads */ - if (disp->disp->oclass >= GF110_DISP) - crtcs = nvif_rd32(&device->object, 0x612004) & 0xf; - else - crtcs = 0x3; - - for (i = 0; i < fls(crtcs); i++) { - if (!(crtcs & (1 << i))) - continue; - ret = nv50_head_create(dev, i); - if (ret) - goto out; - } - - /* create encoder/connector objects based on VBIOS DCB table */ - for (i = 0, dcbe = &dcb->entry[0]; i < dcb->entries; i++, dcbe++) { - connector = nouveau_connector_create(dev, dcbe->connector); - if (IS_ERR(connector)) - continue; - - if (dcbe->location == DCB_LOC_ON_CHIP) { - switch (dcbe->type) { - case DCB_OUTPUT_TMDS: - case DCB_OUTPUT_LVDS: - case DCB_OUTPUT_DP: - ret = nv50_sor_create(connector, dcbe); - break; - case DCB_OUTPUT_ANALOG: - ret = nv50_dac_create(connector, dcbe); - break; - default: - ret = -ENODEV; - break; - } - } else { - ret = nv50_pior_create(connector, dcbe); - } - - if (ret) { - NV_WARN(drm, "failed to create encoder %d/%d/%d: %d\n", - dcbe->location, dcbe->type, - ffs(dcbe->or) - 1, ret); - ret = 0; - } - } - - /* cull any connectors we created that don't have an encoder */ - list_for_each_entry_safe(connector, tmp, &dev->mode_config.connector_list, head) { - if (connector->encoder_ids[0]) - continue; - - NV_WARN(drm, "%s has no encoders, removing\n", - connector->name); - connector->funcs->destroy(connector); - } - -out: - if (ret) - nv50_display_destroy(dev); - return ret; -} diff --git a/drivers/gpu/drm/nouveau/nv50_display.h b/drivers/gpu/drm/nouveau/nv50_display.h index 918187cee84b..fbd3b15583bc 100644 --- a/drivers/gpu/drm/nouveau/nv50_display.h +++ b/drivers/gpu/drm/nouveau/nv50_display.h @@ -28,7 +28,6 @@ #define __NV50_DISPLAY_H__ #include "nouveau_display.h" -#include "nouveau_crtc.h" #include "nouveau_reg.h" int nv50_display_create(struct drm_device *); diff --git a/drivers/gpu/drm/nouveau/nv50_fence.c b/drivers/gpu/drm/nouveau/nv50_fence.c index a369d978e267..a00ecc3de053 100644 --- a/drivers/gpu/drm/nouveau/nv50_fence.c +++ b/drivers/gpu/drm/nouveau/nv50_fence.c @@ -78,8 +78,6 @@ nv50_fence_create(struct nouveau_drm *drm) priv->base.resume = nv17_fence_resume; priv->base.context_new = nv50_fence_context_new; priv->base.context_del = nv10_fence_context_del; - priv->base.contexts = 127; - priv->base.context_base = dma_fence_context_alloc(priv->base.contexts); spin_lock_init(&priv->lock); ret = nouveau_bo_new(&drm->client, 4096, 0x1000, TTM_PL_FLAG_VRAM, diff --git a/drivers/gpu/drm/nouveau/nv84_fence.c b/drivers/gpu/drm/nouveau/nv84_fence.c index 5f0c0c27d5dc..090664899247 100644 --- a/drivers/gpu/drm/nouveau/nv84_fence.c +++ b/drivers/gpu/drm/nouveau/nv84_fence.c @@ -141,9 +141,9 @@ nv84_fence_suspend(struct nouveau_drm *drm) struct nv84_fence_priv *priv = drm->fence; int i; - priv->suspend = vmalloc(priv->base.contexts * sizeof(u32)); + priv->suspend = vmalloc(drm->chan.nr * sizeof(u32)); if (priv->suspend) { - for (i = 0; i < priv->base.contexts; i++) + for (i = 0; i < drm->chan.nr; i++) priv->suspend[i] = nouveau_bo_rd32(priv->bo, i*4); } @@ -157,7 +157,7 @@ nv84_fence_resume(struct nouveau_drm *drm) int i; if (priv->suspend) { - for (i = 0; i < priv->base.contexts; i++) + for (i = 0; i < drm->chan.nr; i++) nouveau_bo_wr32(priv->bo, i*4, priv->suspend[i]); vfree(priv->suspend); priv->suspend = NULL; @@ -179,7 +179,6 @@ nv84_fence_destroy(struct nouveau_drm *drm) int nv84_fence_create(struct nouveau_drm *drm) { - struct nvkm_fifo *fifo = nvxx_fifo(&drm->client.device); struct nv84_fence_priv *priv; u32 domain; int ret; @@ -194,8 +193,6 @@ nv84_fence_create(struct nouveau_drm *drm) priv->base.context_new = nv84_fence_context_new; priv->base.context_del = nv84_fence_context_del; - priv->base.contexts = fifo->nr; - priv->base.context_base = dma_fence_context_alloc(priv->base.contexts); priv->base.uevent = true; mutex_init(&priv->mutex); @@ -207,7 +204,7 @@ nv84_fence_create(struct nouveau_drm *drm) * will lose CPU/GPU coherency! */ TTM_PL_FLAG_TT | TTM_PL_FLAG_UNCACHED; - ret = nouveau_bo_new(&drm->client, 16 * priv->base.contexts, 0, + ret = nouveau_bo_new(&drm->client, 16 * drm->chan.nr, 0, domain, 0, 0, NULL, NULL, &priv->bo); if (ret == 0) { ret = nouveau_bo_pin(priv->bo, domain, false); diff --git a/drivers/gpu/drm/nouveau/nvif/Kbuild b/drivers/gpu/drm/nouveau/nvif/Kbuild index f1675a4ab6fa..42e8c85caa33 100644 --- a/drivers/gpu/drm/nouveau/nvif/Kbuild +++ b/drivers/gpu/drm/nouveau/nvif/Kbuild @@ -1,8 +1,14 @@ nvif-y := nvif/object.o nvif-y += nvif/client.o nvif-y += nvif/device.o +nvif-y += nvif/disp.o nvif-y += nvif/driver.o +nvif-y += nvif/fifo.o nvif-y += nvif/mem.o nvif-y += nvif/mmu.o nvif-y += nvif/notify.o nvif-y += nvif/vmm.o + +# Usermode classes +nvif-y += nvif/user.o +nvif-y += nvif/userc361.o diff --git a/drivers/gpu/drm/nouveau/nvif/device.c b/drivers/gpu/drm/nouveau/nvif/device.c index 252d8c33215b..1ec101ba3b42 100644 --- a/drivers/gpu/drm/nouveau/nvif/device.c +++ b/drivers/gpu/drm/nouveau/nvif/device.c @@ -37,6 +37,9 @@ nvif_device_time(struct nvif_device *device) void nvif_device_fini(struct nvif_device *device) { + nvif_user_fini(device); + kfree(device->runlist); + device->runlist = NULL; nvif_object_fini(&device->object); } @@ -46,6 +49,8 @@ nvif_device_init(struct nvif_object *parent, u32 handle, s32 oclass, { int ret = nvif_object_init(parent, handle, oclass, data, size, &device->object); + device->runlist = NULL; + device->user.func = NULL; if (ret == 0) { device->info.version = 0; ret = nvif_object_mthd(&device->object, NV_DEVICE_V0_INFO, diff --git a/drivers/gpu/drm/nouveau/nvif/disp.c b/drivers/gpu/drm/nouveau/nvif/disp.c new file mode 100644 index 000000000000..18c7d064f75c --- /dev/null +++ b/drivers/gpu/drm/nouveau/nvif/disp.c @@ -0,0 +1,60 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include <nvif/disp.h> +#include <nvif/device.h> + +#include <nvif/class.h> + +void +nvif_disp_dtor(struct nvif_disp *disp) +{ + nvif_object_fini(&disp->object); +} + +int +nvif_disp_ctor(struct nvif_device *device, s32 oclass, struct nvif_disp *disp) +{ + static const struct nvif_mclass disps[] = { + { GV100_DISP, -1 }, + { GP102_DISP, -1 }, + { GP100_DISP, -1 }, + { GM200_DISP, -1 }, + { GM107_DISP, -1 }, + { GK110_DISP, -1 }, + { GK104_DISP, -1 }, + { GF110_DISP, -1 }, + { GT214_DISP, -1 }, + { GT206_DISP, -1 }, + { GT200_DISP, -1 }, + { G82_DISP, -1 }, + { NV50_DISP, -1 }, + { NV04_DISP, -1 }, + {} + }; + int cid = nvif_sclass(&device->object, disps, oclass); + disp->object.client = NULL; + if (cid < 0) + return cid; + + return nvif_object_init(&device->object, 0, disps[cid].oclass, + NULL, 0, &disp->object); +} diff --git a/drivers/gpu/drm/nouveau/nvif/fifo.c b/drivers/gpu/drm/nouveau/nvif/fifo.c new file mode 100644 index 000000000000..99d4fd17543c --- /dev/null +++ b/drivers/gpu/drm/nouveau/nvif/fifo.c @@ -0,0 +1,99 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include <nvif/fifo.h> + +static int +nvif_fifo_runlists(struct nvif_device *device) +{ + struct nvif_object *object = &device->object; + struct { + struct nv_device_info_v1 m; + struct { + struct nv_device_info_v1_data runlists; + struct nv_device_info_v1_data runlist[64]; + } v; + } *a; + int ret, i; + + if (device->runlist) + return 0; + + if (!(a = kmalloc(sizeof(*a), GFP_KERNEL))) + return -ENOMEM; + a->m.version = 1; + a->m.count = sizeof(a->v) / sizeof(a->v.runlists); + a->v.runlists.mthd = NV_DEVICE_FIFO_RUNLISTS; + for (i = 0; i < ARRAY_SIZE(a->v.runlist); i++) + a->v.runlist[i].mthd = NV_DEVICE_FIFO_RUNLIST_ENGINES(i); + + ret = nvif_object_mthd(object, NV_DEVICE_V0_INFO, a, sizeof(*a)); + if (ret) + goto done; + + device->runlists = fls64(a->v.runlists.data); + device->runlist = kzalloc(sizeof(*device->runlist) * + device->runlists, GFP_KERNEL); + if (!device->runlist) { + ret = -ENOMEM; + goto done; + } + + for (i = 0; i < device->runlists; i++) { + if (a->v.runlists.data & BIT_ULL(i)) + device->runlist[i].engines = a->v.runlist[i].data; + } + +done: + kfree(a); + return ret; +} + +u64 +nvif_fifo_runlist(struct nvif_device *device, u64 engine) +{ + struct nvif_object *object = &device->object; + struct { + struct nv_device_info_v1 m; + struct { + struct nv_device_info_v1_data engine; + } v; + } a = { + .m.version = 1, + .m.count = sizeof(a.v) / sizeof(a.v.engine), + .v.engine.mthd = engine, + }; + u64 runm = 0; + int ret, i; + + if ((ret = nvif_fifo_runlists(device))) + return runm; + + ret = nvif_object_mthd(object, NV_DEVICE_V0_INFO, &a, sizeof(a)); + if (ret == 0) { + for (i = 0; i < device->runlists; i++) { + if (device->runlist[i].engines & a.v.engine.data) + runm |= BIT_ULL(i); + } + } + + return runm; +} diff --git a/drivers/gpu/drm/nouveau/nvif/mem.c b/drivers/gpu/drm/nouveau/nvif/mem.c index 0f9382c60145..b6ebb3b58673 100644 --- a/drivers/gpu/drm/nouveau/nvif/mem.c +++ b/drivers/gpu/drm/nouveau/nvif/mem.c @@ -24,6 +24,19 @@ #include <nvif/if000a.h> +int +nvif_mem_init_map(struct nvif_mmu *mmu, u8 type, u64 size, struct nvif_mem *mem) +{ + int ret = nvif_mem_init(mmu, mmu->mem, NVIF_MEM_MAPPABLE | type, 0, + size, NULL, 0, mem); + if (ret == 0) { + ret = nvif_object_map(&mem->object, NULL, 0); + if (ret) + nvif_mem_fini(mem); + } + return ret; +} + void nvif_mem_fini(struct nvif_mem *mem) { diff --git a/drivers/gpu/drm/nouveau/nvif/mmu.c b/drivers/gpu/drm/nouveau/nvif/mmu.c index 15d0dcbf7ab4..358ac4f3cf91 100644 --- a/drivers/gpu/drm/nouveau/nvif/mmu.c +++ b/drivers/gpu/drm/nouveau/nvif/mmu.c @@ -36,6 +36,12 @@ nvif_mmu_fini(struct nvif_mmu *mmu) int nvif_mmu_init(struct nvif_object *parent, s32 oclass, struct nvif_mmu *mmu) { + static const struct nvif_mclass mems[] = { + { NVIF_CLASS_MEM_GF100, -1 }, + { NVIF_CLASS_MEM_NV50 , -1 }, + { NVIF_CLASS_MEM_NV04 , -1 }, + {} + }; struct nvif_mmu_v0 args; int ret, i; @@ -54,6 +60,11 @@ nvif_mmu_init(struct nvif_object *parent, s32 oclass, struct nvif_mmu *mmu) mmu->type_nr = args.type_nr; mmu->kind_nr = args.kind_nr; + ret = nvif_mclass(&mmu->object, mems); + if (ret < 0) + goto done; + mmu->mem = mems[ret].oclass; + mmu->heap = kmalloc(sizeof(*mmu->heap) * mmu->heap_nr, GFP_KERNEL); mmu->type = kmalloc(sizeof(*mmu->type) * mmu->type_nr, GFP_KERNEL); if (ret = -ENOMEM, !mmu->heap || !mmu->type) diff --git a/drivers/gpu/drm/nouveau/nvif/user.c b/drivers/gpu/drm/nouveau/nvif/user.c new file mode 100644 index 000000000000..10da3cdca647 --- /dev/null +++ b/drivers/gpu/drm/nouveau/nvif/user.c @@ -0,0 +1,64 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include <nvif/user.h> +#include <nvif/device.h> + +#include <nvif/class.h> + +void +nvif_user_fini(struct nvif_device *device) +{ + if (device->user.func) { + nvif_object_fini(&device->user.object); + device->user.func = NULL; + } +} + +int +nvif_user_init(struct nvif_device *device) +{ + struct { + s32 oclass; + int version; + const struct nvif_user_func *func; + } users[] = { + { VOLTA_USERMODE_A, -1, &nvif_userc361 }, + {} + }; + int cid, ret; + + if (device->user.func) + return 0; + + cid = nvif_mclass(&device->object, users); + if (cid < 0) + return cid; + + ret = nvif_object_init(&device->object, 0, users[cid].oclass, NULL, 0, + &device->user.object); + if (ret) + return ret; + + nvif_object_map(&device->user.object, NULL, 0); + device->user.func = users[cid].func; + return 0; +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gpfifogp100.c b/drivers/gpu/drm/nouveau/nvif/userc361.c index 1530a9217aea..19f9958e7e01 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gpfifogp100.c +++ b/drivers/gpu/drm/nouveau/nvif/userc361.c @@ -1,5 +1,5 @@ /* - * Copyright 2016 Red Hat Inc. + * Copyright 2018 Red Hat Inc. * * Permission is hereby granted, free of charge, to any person obtaining a * copy of this software and associated documentation files (the "Software"), @@ -18,17 +18,16 @@ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR * OTHER DEALINGS IN THE SOFTWARE. - * - * Authors: Ben Skeggs */ -#include "changk104.h" +#include <nvif/user.h> -#include <nvif/class.h> +static void +nvif_userc361_doorbell(struct nvif_user *user, u32 token) +{ + nvif_wr32(&user->object, 0x90, token); +} -const struct nvkm_fifo_chan_oclass -gp100_fifo_gpfifo_oclass = { - .base.oclass = PASCAL_CHANNEL_GPFIFO_A, - .base.minver = 0, - .base.maxver = 0, - .ctor = gk104_fifo_gpfifo_new, +const struct nvif_user_func +nvif_userc361 = { + .doorbell = nvif_userc361_doorbell, }; diff --git a/drivers/gpu/drm/nouveau/nvif/vmm.c b/drivers/gpu/drm/nouveau/nvif/vmm.c index 31cdb2d2e1ff..191832be6c65 100644 --- a/drivers/gpu/drm/nouveau/nvif/vmm.c +++ b/drivers/gpu/drm/nouveau/nvif/vmm.c @@ -37,7 +37,7 @@ nvif_vmm_map(struct nvif_vmm *vmm, u64 addr, u64 size, void *argv, u32 argc, struct nvif_mem *mem, u64 offset) { struct nvif_vmm_map_v0 *args; - u8 stack[16]; + u8 stack[48]; int ret; if (sizeof(*args) + argc > sizeof(stack)) { diff --git a/drivers/gpu/drm/nouveau/nvkm/core/engine.c b/drivers/gpu/drm/nouveau/nvkm/core/engine.c index 657231c3c098..d0322ce85172 100644 --- a/drivers/gpu/drm/nouveau/nvkm/core/engine.c +++ b/drivers/gpu/drm/nouveau/nvkm/core/engine.c @@ -83,6 +83,20 @@ nvkm_engine_intr(struct nvkm_subdev *subdev) } static int +nvkm_engine_info(struct nvkm_subdev *subdev, u64 mthd, u64 *data) +{ + struct nvkm_engine *engine = nvkm_engine(subdev); + if (engine->func->info) { + if ((engine = nvkm_engine_ref(engine))) { + int ret = engine->func->info(engine, mthd, data); + nvkm_engine_unref(&engine); + return ret; + } + } + return -ENOSYS; +} + +static int nvkm_engine_fini(struct nvkm_subdev *subdev, bool suspend) { struct nvkm_engine *engine = nvkm_engine(subdev); @@ -150,6 +164,7 @@ nvkm_engine_func = { .preinit = nvkm_engine_preinit, .init = nvkm_engine_init, .fini = nvkm_engine_fini, + .info = nvkm_engine_info, .intr = nvkm_engine_intr, }; diff --git a/drivers/gpu/drm/nouveau/nvkm/core/subdev.c b/drivers/gpu/drm/nouveau/nvkm/core/subdev.c index a134d225f958..03f676c18aad 100644 --- a/drivers/gpu/drm/nouveau/nvkm/core/subdev.c +++ b/drivers/gpu/drm/nouveau/nvkm/core/subdev.c @@ -35,6 +35,7 @@ nvkm_subdev_name[NVKM_SUBDEV_NR] = { [NVKM_SUBDEV_BUS ] = "bus", [NVKM_SUBDEV_CLK ] = "clk", [NVKM_SUBDEV_DEVINIT ] = "devinit", + [NVKM_SUBDEV_FAULT ] = "fault", [NVKM_SUBDEV_FB ] = "fb", [NVKM_SUBDEV_FUSE ] = "fuse", [NVKM_SUBDEV_GPIO ] = "gpio", @@ -60,6 +61,9 @@ nvkm_subdev_name[NVKM_SUBDEV_NR] = { [NVKM_ENGINE_CE3 ] = "ce3", [NVKM_ENGINE_CE4 ] = "ce4", [NVKM_ENGINE_CE5 ] = "ce5", + [NVKM_ENGINE_CE6 ] = "ce6", + [NVKM_ENGINE_CE7 ] = "ce7", + [NVKM_ENGINE_CE8 ] = "ce8", [NVKM_ENGINE_CIPHER ] = "cipher", [NVKM_ENGINE_DISP ] = "disp", [NVKM_ENGINE_DMAOBJ ] = "dma", @@ -92,6 +96,14 @@ nvkm_subdev_intr(struct nvkm_subdev *subdev) } int +nvkm_subdev_info(struct nvkm_subdev *subdev, u64 mthd, u64 *data) +{ + if (subdev->func->info) + return subdev->func->info(subdev, mthd, data); + return -ENOSYS; +} + +int nvkm_subdev_fini(struct nvkm_subdev *subdev, bool suspend) { struct nvkm_device *device = subdev->device; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/ce/Kbuild b/drivers/gpu/drm/nouveau/nvkm/engine/ce/Kbuild index 255d81ccf916..80d784441904 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/ce/Kbuild +++ b/drivers/gpu/drm/nouveau/nvkm/engine/ce/Kbuild @@ -5,3 +5,4 @@ nvkm-y += nvkm/engine/ce/gm107.o nvkm-y += nvkm/engine/ce/gm200.o nvkm-y += nvkm/engine/ce/gp100.o nvkm-y += nvkm/engine/ce/gp102.o +nvkm-y += nvkm/engine/ce/gv100.o diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/cursg84.c b/drivers/gpu/drm/nouveau/nvkm/engine/ce/gv100.c index fa781b5a7e07..fcda3de45857 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/cursg84.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/ce/gv100.c @@ -1,5 +1,5 @@ /* - * Copyright 2015 Red Hat Inc. + * Copyright 2018 Red Hat Inc. * * Permission is hereby granted, free of charge, to any person obtaining a * copy of this software and associated documentation files (the "Software"), @@ -18,20 +18,23 @@ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR * OTHER DEALINGS IN THE SOFTWARE. - * - * Authors: Ben Skeggs <bskeggs@redhat.com> */ -#include "channv50.h" -#include "rootnv50.h" +#include "priv.h" #include <nvif/class.h> -const struct nv50_disp_pioc_oclass -g84_disp_curs_oclass = { - .base.oclass = G82_DISP_CURSOR, - .base.minver = 0, - .base.maxver = 0, - .ctor = nv50_disp_curs_new, - .func = &nv50_disp_pioc_func, - .chid = { 7, 7 }, +static const struct nvkm_engine_func +gv100_ce = { + .intr = gp100_ce_intr, + .sclass = { + { -1, -1, VOLTA_DMA_COPY_A }, + {} + } }; + +int +gv100_ce_new(struct nvkm_device *device, int index, + struct nvkm_engine **pengine) +{ + return nvkm_engine_new_(&gv100_ce, device, index, true, pengine); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c b/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c index 05cd674326a6..e294013426ce 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c @@ -2161,6 +2161,7 @@ nv130_chipset = { .bios = nvkm_bios_new, .bus = gf100_bus_new, .devinit = gm200_devinit_new, + .fault = gp100_fault_new, .fb = gp100_fb_new, .fuse = gm107_fuse_new, .gpio = gk104_gpio_new, @@ -2196,13 +2197,14 @@ nv132_chipset = { .bios = nvkm_bios_new, .bus = gf100_bus_new, .devinit = gm200_devinit_new, + .fault = gp100_fault_new, .fb = gp102_fb_new, .fuse = gm107_fuse_new, .gpio = gk104_gpio_new, .i2c = gm200_i2c_new, .ibus = gm200_ibus_new, .imem = nv50_instmem_new, - .ltc = gp100_ltc_new, + .ltc = gp102_ltc_new, .mc = gp100_mc_new, .mmu = gp100_mmu_new, .therm = gp100_therm_new, @@ -2231,13 +2233,14 @@ nv134_chipset = { .bios = nvkm_bios_new, .bus = gf100_bus_new, .devinit = gm200_devinit_new, + .fault = gp100_fault_new, .fb = gp102_fb_new, .fuse = gm107_fuse_new, .gpio = gk104_gpio_new, .i2c = gm200_i2c_new, .ibus = gm200_ibus_new, .imem = nv50_instmem_new, - .ltc = gp100_ltc_new, + .ltc = gp102_ltc_new, .mc = gp100_mc_new, .mmu = gp100_mmu_new, .therm = gp100_therm_new, @@ -2253,7 +2256,7 @@ nv134_chipset = { .disp = gp102_disp_new, .dma = gf119_dma_new, .fifo = gp100_fifo_new, - .gr = gp102_gr_new, + .gr = gp104_gr_new, .nvdec = gp102_nvdec_new, .sec2 = gp102_sec2_new, .sw = gf100_sw_new, @@ -2266,13 +2269,14 @@ nv136_chipset = { .bios = nvkm_bios_new, .bus = gf100_bus_new, .devinit = gm200_devinit_new, + .fault = gp100_fault_new, .fb = gp102_fb_new, .fuse = gm107_fuse_new, .gpio = gk104_gpio_new, .i2c = gm200_i2c_new, .ibus = gm200_ibus_new, .imem = nv50_instmem_new, - .ltc = gp100_ltc_new, + .ltc = gp102_ltc_new, .mc = gp100_mc_new, .mmu = gp100_mmu_new, .therm = gp100_therm_new, @@ -2288,7 +2292,7 @@ nv136_chipset = { .disp = gp102_disp_new, .dma = gf119_dma_new, .fifo = gp100_fifo_new, - .gr = gp102_gr_new, + .gr = gp104_gr_new, .nvdec = gp102_nvdec_new, .sec2 = gp102_sec2_new, .sw = gf100_sw_new, @@ -2301,13 +2305,14 @@ nv137_chipset = { .bios = nvkm_bios_new, .bus = gf100_bus_new, .devinit = gm200_devinit_new, + .fault = gp100_fault_new, .fb = gp102_fb_new, .fuse = gm107_fuse_new, .gpio = gk104_gpio_new, .i2c = gm200_i2c_new, .ibus = gm200_ibus_new, .imem = nv50_instmem_new, - .ltc = gp100_ltc_new, + .ltc = gp102_ltc_new, .mc = gp100_mc_new, .mmu = gp100_mmu_new, .therm = gp100_therm_new, @@ -2336,13 +2341,14 @@ nv138_chipset = { .bios = nvkm_bios_new, .bus = gf100_bus_new, .devinit = gm200_devinit_new, + .fault = gp100_fault_new, .fb = gp102_fb_new, .fuse = gm107_fuse_new, .gpio = gk104_gpio_new, .i2c = gm200_i2c_new, .ibus = gm200_ibus_new, .imem = nv50_instmem_new, - .ltc = gp100_ltc_new, + .ltc = gp102_ltc_new, .mc = gp100_mc_new, .mmu = gp100_mmu_new, .therm = gp100_therm_new, @@ -2369,11 +2375,12 @@ nv13b_chipset = { .name = "GP10B", .bar = gm20b_bar_new, .bus = gf100_bus_new, + .fault = gp100_fault_new, .fb = gp10b_fb_new, .fuse = gm107_fuse_new, .ibus = gp10b_ibus_new, .imem = gk20a_instmem_new, - .ltc = gp100_ltc_new, + .ltc = gp102_ltc_new, .mc = gp10b_mc_new, .mmu = gp10b_mmu_new, .secboot = gp10b_secboot_new, @@ -2387,6 +2394,46 @@ nv13b_chipset = { .sw = gf100_sw_new, }; +static const struct nvkm_device_chip +nv140_chipset = { + .name = "GV100", + .bar = gm107_bar_new, + .bios = nvkm_bios_new, + .bus = gf100_bus_new, + .devinit = gv100_devinit_new, + .fault = gv100_fault_new, + .fb = gv100_fb_new, + .fuse = gm107_fuse_new, + .gpio = gk104_gpio_new, + .i2c = gm200_i2c_new, + .ibus = gm200_ibus_new, + .imem = nv50_instmem_new, + .ltc = gp102_ltc_new, + .mc = gp100_mc_new, + .mmu = gv100_mmu_new, + .pci = gp100_pci_new, + .pmu = gp102_pmu_new, + .secboot = gp108_secboot_new, + .therm = gp100_therm_new, + .timer = gk20a_timer_new, + .top = gk104_top_new, + .disp = gv100_disp_new, + .ce[0] = gv100_ce_new, + .ce[1] = gv100_ce_new, + .ce[2] = gv100_ce_new, + .ce[3] = gv100_ce_new, + .ce[4] = gv100_ce_new, + .ce[5] = gv100_ce_new, + .ce[6] = gv100_ce_new, + .ce[7] = gv100_ce_new, + .ce[8] = gv100_ce_new, + .dma = gv100_dma_new, + .fifo = gv100_fifo_new, + .gr = gv100_gr_new, + .nvdec = gp102_nvdec_new, + .sec2 = gp102_sec2_new, +}; + static int nvkm_device_event_ctor(struct nvkm_object *object, void *data, u32 size, struct nvkm_notify *notify) @@ -2420,6 +2467,7 @@ nvkm_device_subdev(struct nvkm_device *device, int index) _(BUS , device->bus , &device->bus->subdev); _(CLK , device->clk , &device->clk->subdev); _(DEVINIT , device->devinit , &device->devinit->subdev); + _(FAULT , device->fault , &device->fault->subdev); _(FB , device->fb , &device->fb->subdev); _(FUSE , device->fuse , &device->fuse->subdev); _(GPIO , device->gpio , &device->gpio->subdev); @@ -2463,6 +2511,9 @@ nvkm_device_engine(struct nvkm_device *device, int index) _(CE3 , device->ce[3] , device->ce[3]); _(CE4 , device->ce[4] , device->ce[4]); _(CE5 , device->ce[5] , device->ce[5]); + _(CE6 , device->ce[6] , device->ce[6]); + _(CE7 , device->ce[7] , device->ce[7]); + _(CE8 , device->ce[8] , device->ce[8]); _(CIPHER , device->cipher , device->cipher); _(DISP , device->disp , &device->disp->engine); _(DMAOBJ , device->dma , &device->dma->engine); @@ -2739,6 +2790,7 @@ nvkm_device_ctor(const struct nvkm_device_func *func, case 0x110: case 0x120: device->card_type = GM100; break; case 0x130: device->card_type = GP100; break; + case 0x140: device->card_type = GV100; break; default: break; } @@ -2830,6 +2882,7 @@ nvkm_device_ctor(const struct nvkm_device_func *func, case 0x137: device->chip = &nv137_chipset; break; case 0x138: device->chip = &nv138_chipset; break; case 0x13b: device->chip = &nv13b_chipset; break; + case 0x140: device->chip = &nv140_chipset; break; default: nvdev_error(device, "unknown chipset (%08x)\n", boot0); goto done; @@ -2891,6 +2944,7 @@ nvkm_device_ctor(const struct nvkm_device_func *func, _(NVKM_SUBDEV_BUS , bus); _(NVKM_SUBDEV_CLK , clk); _(NVKM_SUBDEV_DEVINIT , devinit); + _(NVKM_SUBDEV_FAULT , fault); _(NVKM_SUBDEV_FB , fb); _(NVKM_SUBDEV_FUSE , fuse); _(NVKM_SUBDEV_GPIO , gpio); @@ -2916,6 +2970,9 @@ nvkm_device_ctor(const struct nvkm_device_func *func, _(NVKM_ENGINE_CE3 , ce[3]); _(NVKM_ENGINE_CE4 , ce[4]); _(NVKM_ENGINE_CE5 , ce[5]); + _(NVKM_ENGINE_CE6 , ce[6]); + _(NVKM_ENGINE_CE7 , ce[7]); + _(NVKM_ENGINE_CE8 , ce[8]); _(NVKM_ENGINE_CIPHER , cipher); _(NVKM_ENGINE_DISP , disp); _(NVKM_ENGINE_DMAOBJ , dma); diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/device/priv.h b/drivers/gpu/drm/nouveau/nvkm/engine/device/priv.h index 08d0bf605722..253ab914a8ef 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/device/priv.h +++ b/drivers/gpu/drm/nouveau/nvkm/engine/device/priv.h @@ -8,6 +8,7 @@ #include <subdev/bus.h> #include <subdev/clk.h> #include <subdev/devinit.h> +#include <subdev/fault.h> #include <subdev/fb.h> #include <subdev/fuse.h> #include <subdev/gpio.h> diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/device/user.c b/drivers/gpu/drm/nouveau/nvkm/engine/device/user.c index 17adcb4e8854..dde6bbafa709 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/device/user.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/device/user.c @@ -40,6 +40,66 @@ struct nvkm_udevice { }; static int +nvkm_udevice_info_subdev(struct nvkm_device *device, u64 mthd, u64 *data) +{ + struct nvkm_subdev *subdev; + enum nvkm_devidx subidx; + + switch (mthd & NV_DEVICE_INFO_UNIT) { + case NV_DEVICE_FIFO(0): subidx = NVKM_ENGINE_FIFO; break; + default: + return -EINVAL; + } + + subdev = nvkm_device_subdev(device, subidx); + if (subdev) + return nvkm_subdev_info(subdev, mthd, data); + return -ENODEV; +} + +static void +nvkm_udevice_info_v1(struct nvkm_device *device, + struct nv_device_info_v1_data *args) +{ + if (args->mthd & NV_DEVICE_INFO_UNIT) { + if (nvkm_udevice_info_subdev(device, args->mthd, &args->data)) + args->mthd = NV_DEVICE_INFO_INVALID; + return; + } + + switch (args->mthd) { +#define ENGINE__(A,B,C) NV_DEVICE_INFO_ENGINE_##A: { int _i; \ + for (_i = (B), args->data = 0ULL; _i <= (C); _i++) { \ + if (nvkm_device_engine(device, _i)) \ + args->data |= BIT_ULL(_i); \ + } \ +} +#define ENGINE_A(A) ENGINE__(A, NVKM_ENGINE_##A , NVKM_ENGINE_##A) +#define ENGINE_B(A) ENGINE__(A, NVKM_ENGINE_##A##0, NVKM_ENGINE_##A##_LAST) + case ENGINE_A(SW ); break; + case ENGINE_A(GR ); break; + case ENGINE_A(MPEG ); break; + case ENGINE_A(ME ); break; + case ENGINE_A(CIPHER); break; + case ENGINE_A(BSP ); break; + case ENGINE_A(VP ); break; + case ENGINE_B(CE ); break; + case ENGINE_A(SEC ); break; + case ENGINE_A(MSVLD ); break; + case ENGINE_A(MSPDEC); break; + case ENGINE_A(MSPPP ); break; + case ENGINE_A(MSENC ); break; + case ENGINE_A(VIC ); break; + case ENGINE_A(SEC2 ); break; + case ENGINE_A(NVDEC ); break; + case ENGINE_B(NVENC ); break; + default: + args->mthd = NV_DEVICE_INFO_INVALID; + break; + } +} + +static int nvkm_udevice_info(struct nvkm_udevice *udev, void *data, u32 size) { struct nvkm_object *object = &udev->object; @@ -48,10 +108,21 @@ nvkm_udevice_info(struct nvkm_udevice *udev, void *data, u32 size) struct nvkm_instmem *imem = device->imem; union { struct nv_device_info_v0 v0; + struct nv_device_info_v1 v1; } *args = data; - int ret = -ENOSYS; + int ret = -ENOSYS, i; nvif_ioctl(object, "device info size %d\n", size); + if (!(ret = nvif_unpack(ret, &data, &size, args->v1, 1, 1, true))) { + nvif_ioctl(object, "device info vers %d count %d\n", + args->v1.version, args->v1.count); + if (args->v1.count * sizeof(args->v1.data[0]) == size) { + for (i = 0; i < args->v1.count; i++) + nvkm_udevice_info_v1(device, &args->v1.data[i]); + return 0; + } + return -EINVAL; + } else if (!(ret = nvif_unpack(ret, &data, &size, args->v0, 0, 0, false))) { nvif_ioctl(object, "device info vers %d\n", args->v0.version); } else @@ -103,6 +174,7 @@ nvkm_udevice_info(struct nvkm_udevice *udev, void *data, u32 size) case NV_E0: args->v0.family = NV_DEVICE_INFO_V0_KEPLER; break; case GM100: args->v0.family = NV_DEVICE_INFO_V0_MAXWELL; break; case GP100: args->v0.family = NV_DEVICE_INFO_V0_PASCAL; break; + case GV100: args->v0.family = NV_DEVICE_INFO_V0_VOLTA; break; default: args->v0.family = 0; break; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/Kbuild b/drivers/gpu/drm/nouveau/nvkm/engine/disp/Kbuild index 48ce6699183e..3d485dbf310a 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/Kbuild +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/Kbuild @@ -14,12 +14,14 @@ nvkm-y += nvkm/engine/disp/gm107.o nvkm-y += nvkm/engine/disp/gm200.o nvkm-y += nvkm/engine/disp/gp100.o nvkm-y += nvkm/engine/disp/gp102.o +nvkm-y += nvkm/engine/disp/gv100.o nvkm-y += nvkm/engine/disp/vga.o nvkm-y += nvkm/engine/disp/head.o nvkm-y += nvkm/engine/disp/headnv04.o nvkm-y += nvkm/engine/disp/headnv50.o nvkm-y += nvkm/engine/disp/headgf119.o +nvkm-y += nvkm/engine/disp/headgv100.o nvkm-y += nvkm/engine/disp/ior.o nvkm-y += nvkm/engine/disp/dacnv50.o @@ -35,6 +37,7 @@ nvkm-y += nvkm/engine/disp/sorgf119.o nvkm-y += nvkm/engine/disp/sorgk104.o nvkm-y += nvkm/engine/disp/sorgm107.o nvkm-y += nvkm/engine/disp/sorgm200.o +nvkm-y += nvkm/engine/disp/sorgv100.o nvkm-y += nvkm/engine/disp/outp.o nvkm-y += nvkm/engine/disp/dp.o @@ -47,6 +50,7 @@ nvkm-y += nvkm/engine/disp/hdmig84.o nvkm-y += nvkm/engine/disp/hdmigt215.o nvkm-y += nvkm/engine/disp/hdmigf119.o nvkm-y += nvkm/engine/disp/hdmigk104.o +nvkm-y += nvkm/engine/disp/hdmigv100.o nvkm-y += nvkm/engine/disp/conn.o @@ -63,57 +67,49 @@ nvkm-y += nvkm/engine/disp/rootgm107.o nvkm-y += nvkm/engine/disp/rootgm200.o nvkm-y += nvkm/engine/disp/rootgp100.o nvkm-y += nvkm/engine/disp/rootgp102.o +nvkm-y += nvkm/engine/disp/rootgv100.o nvkm-y += nvkm/engine/disp/channv50.o nvkm-y += nvkm/engine/disp/changf119.o +nvkm-y += nvkm/engine/disp/changv100.o nvkm-y += nvkm/engine/disp/dmacnv50.o nvkm-y += nvkm/engine/disp/dmacgf119.o nvkm-y += nvkm/engine/disp/dmacgp102.o +nvkm-y += nvkm/engine/disp/dmacgv100.o nvkm-y += nvkm/engine/disp/basenv50.o nvkm-y += nvkm/engine/disp/baseg84.o -nvkm-y += nvkm/engine/disp/basegt200.o -nvkm-y += nvkm/engine/disp/basegt215.o nvkm-y += nvkm/engine/disp/basegf119.o -nvkm-y += nvkm/engine/disp/basegk104.o -nvkm-y += nvkm/engine/disp/basegk110.o nvkm-y += nvkm/engine/disp/basegp102.o nvkm-y += nvkm/engine/disp/corenv50.o nvkm-y += nvkm/engine/disp/coreg84.o nvkm-y += nvkm/engine/disp/coreg94.o -nvkm-y += nvkm/engine/disp/coregt200.o -nvkm-y += nvkm/engine/disp/coregt215.o nvkm-y += nvkm/engine/disp/coregf119.o nvkm-y += nvkm/engine/disp/coregk104.o -nvkm-y += nvkm/engine/disp/coregk110.o -nvkm-y += nvkm/engine/disp/coregm107.o -nvkm-y += nvkm/engine/disp/coregm200.o -nvkm-y += nvkm/engine/disp/coregp100.o nvkm-y += nvkm/engine/disp/coregp102.o +nvkm-y += nvkm/engine/disp/coregv100.o nvkm-y += nvkm/engine/disp/ovlynv50.o nvkm-y += nvkm/engine/disp/ovlyg84.o nvkm-y += nvkm/engine/disp/ovlygt200.o -nvkm-y += nvkm/engine/disp/ovlygt215.o nvkm-y += nvkm/engine/disp/ovlygf119.o nvkm-y += nvkm/engine/disp/ovlygk104.o nvkm-y += nvkm/engine/disp/ovlygp102.o +nvkm-y += nvkm/engine/disp/wimmgv100.o + +nvkm-y += nvkm/engine/disp/wndwgv100.o + nvkm-y += nvkm/engine/disp/piocnv50.o nvkm-y += nvkm/engine/disp/piocgf119.o nvkm-y += nvkm/engine/disp/cursnv50.o -nvkm-y += nvkm/engine/disp/cursg84.o -nvkm-y += nvkm/engine/disp/cursgt215.o nvkm-y += nvkm/engine/disp/cursgf119.o -nvkm-y += nvkm/engine/disp/cursgk104.o nvkm-y += nvkm/engine/disp/cursgp102.o +nvkm-y += nvkm/engine/disp/cursgv100.o nvkm-y += nvkm/engine/disp/oimmnv50.o -nvkm-y += nvkm/engine/disp/oimmg84.o -nvkm-y += nvkm/engine/disp/oimmgt215.o nvkm-y += nvkm/engine/disp/oimmgf119.o -nvkm-y += nvkm/engine/disp/oimmgk104.o nvkm-y += nvkm/engine/disp/oimmgp102.o diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/base.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/base.c index 93a75e5b2791..32fa94a9773f 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/base.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/base.c @@ -220,6 +220,9 @@ nvkm_disp_fini(struct nvkm_engine *engine, bool suspend) struct nvkm_conn *conn; struct nvkm_outp *outp; + if (disp->func->fini) + disp->func->fini(disp); + list_for_each_entry(outp, &disp->outp, head) { nvkm_outp_fini(outp); } @@ -237,6 +240,7 @@ nvkm_disp_init(struct nvkm_engine *engine) struct nvkm_disp *disp = nvkm_disp(engine); struct nvkm_conn *conn; struct nvkm_outp *outp; + struct nvkm_ior *ior; list_for_each_entry(conn, &disp->conn, head) { nvkm_conn_init(conn); @@ -246,6 +250,19 @@ nvkm_disp_init(struct nvkm_engine *engine) nvkm_outp_init(outp); } + if (disp->func->init) { + int ret = disp->func->init(disp); + if (ret) + return ret; + } + + /* Set 'normal' (ie. when it's attached to a head) state for + * each output resource to 'fully enabled'. + */ + list_for_each_entry(ior, &disp->ior, head) { + ior->func->power(ior, true, true, true, true, true); + } + return 0; } @@ -376,6 +393,12 @@ nvkm_disp_oneinit(struct nvkm_engine *engine) if (ret) return ret; + if (disp->func->oneinit) { + ret = disp->func->oneinit(disp); + if (ret) + return ret; + } + i = 0; list_for_each_entry(head, &disp->head, head) i = max(i, head->id + 1); diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/baseg84.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/baseg84.c index 6d17630a3dee..01253f4a9946 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/baseg84.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/baseg84.c @@ -21,10 +21,7 @@ * * Authors: Ben Skeggs */ -#include "dmacnv50.h" -#include "rootnv50.h" - -#include <nvif/class.h> +#include "channv50.h" static const struct nv50_disp_mthd_list g84_disp_base_mthd_base = { @@ -56,8 +53,8 @@ g84_disp_base_mthd_base = { } }; -const struct nv50_disp_chan_mthd -g84_disp_base_chan_mthd = { +static const struct nv50_disp_chan_mthd +g84_disp_base_mthd = { .name = "Base", .addr = 0x000540, .prev = 0x000004, @@ -68,13 +65,10 @@ g84_disp_base_chan_mthd = { } }; -const struct nv50_disp_dmac_oclass -g84_disp_base_oclass = { - .base.oclass = G82_DISP_BASE_CHANNEL_DMA, - .base.minver = 0, - .base.maxver = 0, - .ctor = nv50_disp_base_new, - .func = &nv50_disp_dmac_func, - .mthd = &g84_disp_base_chan_mthd, - .chid = 1, -}; +int +g84_disp_base_new(const struct nvkm_oclass *oclass, void *argv, u32 argc, + struct nv50_disp *disp, struct nvkm_object **pobject) +{ + return nv50_disp_base_new_(&nv50_disp_dmac_func, &g84_disp_base_mthd, + disp, 1, oclass, argv, argc, pobject); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/basegf119.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/basegf119.c index ebcb925e9d90..389e19dfc514 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/basegf119.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/basegf119.c @@ -21,10 +21,7 @@ * * Authors: Ben Skeggs */ -#include "dmacnv50.h" -#include "rootnv50.h" - -#include <nvif/class.h> +#include "channv50.h" static const struct nv50_disp_mthd_list gf119_disp_base_mthd_base = { @@ -91,7 +88,7 @@ gf119_disp_base_mthd_image = { }; const struct nv50_disp_chan_mthd -gf119_disp_base_chan_mthd = { +gf119_disp_base_mthd = { .name = "Base", .addr = 0x001000, .prev = -0x020000, @@ -102,13 +99,10 @@ gf119_disp_base_chan_mthd = { } }; -const struct nv50_disp_dmac_oclass -gf119_disp_base_oclass = { - .base.oclass = GF110_DISP_BASE_CHANNEL_DMA, - .base.minver = 0, - .base.maxver = 0, - .ctor = nv50_disp_base_new, - .func = &gf119_disp_dmac_func, - .mthd = &gf119_disp_base_chan_mthd, - .chid = 1, -}; +int +gf119_disp_base_new(const struct nvkm_oclass *oclass, void *argv, u32 argc, + struct nv50_disp *disp, struct nvkm_object **pobject) +{ + return nv50_disp_base_new_(&gf119_disp_dmac_func, &gf119_disp_base_mthd, + disp, 1, oclass, argv, argc, pobject); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/basegp102.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/basegp102.c index 8a3cdeef8d2c..0cb23d673aa0 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/basegp102.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/basegp102.c @@ -21,18 +21,12 @@ * * Authors: Ben Skeggs <bskeggs@redhat.com> */ -#include "dmacnv50.h" -#include "rootnv50.h" +#include "channv50.h" -#include <nvif/class.h> - -const struct nv50_disp_dmac_oclass -gp102_disp_base_oclass = { - .base.oclass = GK110_DISP_BASE_CHANNEL_DMA, - .base.minver = 0, - .base.maxver = 0, - .ctor = nv50_disp_base_new, - .func = &gp102_disp_dmac_func, - .mthd = &gf119_disp_base_chan_mthd, - .chid = 1, -}; +int +gp102_disp_base_new(const struct nvkm_oclass *oclass, void *argv, u32 argc, + struct nv50_disp *disp, struct nvkm_object **pobject) +{ + return nv50_disp_base_new_(&gp102_disp_dmac_func, &gf119_disp_base_mthd, + disp, 1, oclass, argv, argc, pobject); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/basenv50.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/basenv50.c index f1d6b820d482..19eb7dde01f2 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/basenv50.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/basenv50.c @@ -21,33 +21,30 @@ * * Authors: Ben Skeggs */ -#include "dmacnv50.h" +#include "channv50.h" #include "head.h" -#include "rootnv50.h" #include <core/client.h> -#include <nvif/class.h> #include <nvif/cl507c.h> #include <nvif/unpack.h> int -nv50_disp_base_new(const struct nv50_disp_dmac_func *func, - const struct nv50_disp_chan_mthd *mthd, - struct nv50_disp_root *root, int chid, - const struct nvkm_oclass *oclass, void *data, u32 size, - struct nvkm_object **pobject) +nv50_disp_base_new_(const struct nv50_disp_chan_func *func, + const struct nv50_disp_chan_mthd *mthd, + struct nv50_disp *disp, int chid, + const struct nvkm_oclass *oclass, void *argv, u32 argc, + struct nvkm_object **pobject) { union { struct nv50_disp_base_channel_dma_v0 v0; - } *args = data; + } *args = argv; struct nvkm_object *parent = oclass->parent; - struct nv50_disp *disp = root->disp; int head, ret = -ENOSYS; u64 push; - nvif_ioctl(parent, "create disp base channel dma size %d\n", size); - if (!(ret = nvif_unpack(ret, &data, &size, args->v0, 0, 0, false))) { + nvif_ioctl(parent, "create disp base channel dma size %d\n", argc); + if (!(ret = nvif_unpack(ret, &argv, &argc, args->v0, 0, 0, false))) { nvif_ioctl(parent, "create disp base channel dma vers %d " "pushbuf %016llx head %d\n", args->v0.version, args->v0.pushbuf, args->v0.head); @@ -58,7 +55,7 @@ nv50_disp_base_new(const struct nv50_disp_dmac_func *func, } else return ret; - return nv50_disp_dmac_new_(func, mthd, root, chid + head, + return nv50_disp_dmac_new_(func, mthd, disp, chid + head, head, push, oclass, pobject); } @@ -102,7 +99,7 @@ nv50_disp_base_mthd_image = { }; static const struct nv50_disp_chan_mthd -nv50_disp_base_chan_mthd = { +nv50_disp_base_mthd = { .name = "Base", .addr = 0x000540, .prev = 0x000004, @@ -113,13 +110,10 @@ nv50_disp_base_chan_mthd = { } }; -const struct nv50_disp_dmac_oclass -nv50_disp_base_oclass = { - .base.oclass = NV50_DISP_BASE_CHANNEL_DMA, - .base.minver = 0, - .base.maxver = 0, - .ctor = nv50_disp_base_new, - .func = &nv50_disp_dmac_func, - .mthd = &nv50_disp_base_chan_mthd, - .chid = 1, -}; +int +nv50_disp_base_new(const struct nvkm_oclass *oclass, void *argv, u32 argc, + struct nv50_disp *disp, struct nvkm_object **pobject) +{ + return nv50_disp_base_new_(&nv50_disp_dmac_func, &nv50_disp_base_mthd, + disp, 1, oclass, argv, argc, pobject); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/changf119.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/changf119.c index 17a3d835cb42..29e6dd58ac48 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/changf119.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/changf119.c @@ -47,3 +47,16 @@ gf119_disp_chan_uevent = { .init = gf119_disp_chan_uevent_init, .fini = gf119_disp_chan_uevent_fini, }; + +void +gf119_disp_chan_intr(struct nv50_disp_chan *chan, bool en) +{ + struct nvkm_device *device = chan->disp->base.engine.subdev.device; + const u64 mask = 0x00000001 << chan->chid.user; + if (!en) { + nvkm_mask(device, 0x610090, mask, 0x00000000); + nvkm_mask(device, 0x6100a0, mask, 0x00000000); + } else { + nvkm_mask(device, 0x6100a0, mask, mask); + } +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gpfifogm200.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/changv100.c index a13315147391..75247c9c7e10 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gpfifogm200.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/changv100.c @@ -1,5 +1,5 @@ /* - * Copyright 2015 Red Hat Inc. + * Copyright 2018 Red Hat Inc. * * Permission is hereby granted, free of charge, to any person obtaining a * copy of this software and associated documentation files (the "Software"), @@ -18,17 +18,17 @@ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR * OTHER DEALINGS IN THE SOFTWARE. - * - * Authors: Ben Skeggs */ -#include "changk104.h" - -#include <nvif/class.h> +#include "channv50.h" -const struct nvkm_fifo_chan_oclass -gm200_fifo_gpfifo_oclass = { - .base.oclass = MAXWELL_CHANNEL_GPFIFO_A, - .base.minver = 0, - .base.maxver = 0, - .ctor = gk104_fifo_gpfifo_new, +const struct nvkm_event_func +gv100_disp_chan_uevent = { + .ctor = nv50_disp_chan_uevent_ctor, }; + +u64 +gv100_disp_chan_user(struct nv50_disp_chan *chan, u64 *psize) +{ + *psize = 0x1000; + return 0x690000 + ((chan->chid.user - 1) * 0x1000); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/channv50.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/channv50.c index 723dcbde2ac2..57719f675eec 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/channv50.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/channv50.c @@ -26,6 +26,7 @@ #include <core/client.h> #include <core/notify.h> +#include <core/oproxy.h> #include <core/ramht.h> #include <engine/dma.h> @@ -65,7 +66,7 @@ nv50_disp_mthd_list(struct nv50_disp *disp, int debug, u32 base, int c, void nv50_disp_chan_mthd(struct nv50_disp_chan *chan, int debug) { - struct nv50_disp *disp = chan->root->disp; + struct nv50_disp *disp = chan->disp; struct nvkm_subdev *subdev = &disp->base.engine.subdev; const struct nv50_disp_chan_mthd *mthd = chan->mthd; const struct nv50_disp_mthd_list *list; @@ -154,13 +155,29 @@ nv50_disp_chan_uevent = { .fini = nv50_disp_chan_uevent_fini, }; +u64 +nv50_disp_chan_user(struct nv50_disp_chan *chan, u64 *psize) +{ + *psize = 0x1000; + return 0x640000 + (chan->chid.user * 0x1000); +} + +void +nv50_disp_chan_intr(struct nv50_disp_chan *chan, bool en) +{ + struct nvkm_device *device = chan->disp->base.engine.subdev.device; + const u64 mask = 0x00010001 << chan->chid.user; + const u64 data = en ? 0x00010000 : 0x00000000; + nvkm_mask(device, 0x610028, mask, data); +} + static int nv50_disp_chan_rd32(struct nvkm_object *object, u64 addr, u32 *data) { struct nv50_disp_chan *chan = nv50_disp_chan(object); - struct nv50_disp *disp = chan->root->disp; - struct nvkm_device *device = disp->base.engine.subdev.device; - *data = nvkm_rd32(device, 0x640000 + (chan->chid.user * 0x1000) + addr); + struct nvkm_device *device = chan->disp->base.engine.subdev.device; + u64 size, base = chan->func->user(chan, &size); + *data = nvkm_rd32(device, base + addr); return 0; } @@ -168,9 +185,9 @@ static int nv50_disp_chan_wr32(struct nvkm_object *object, u64 addr, u32 data) { struct nv50_disp_chan *chan = nv50_disp_chan(object); - struct nv50_disp *disp = chan->root->disp; - struct nvkm_device *device = disp->base.engine.subdev.device; - nvkm_wr32(device, 0x640000 + (chan->chid.user * 0x1000) + addr, data); + struct nvkm_device *device = chan->disp->base.engine.subdev.device; + u64 size, base = chan->func->user(chan, &size); + nvkm_wr32(device, base + addr, data); return 0; } @@ -179,7 +196,7 @@ nv50_disp_chan_ntfy(struct nvkm_object *object, u32 type, struct nvkm_event **pevent) { struct nv50_disp_chan *chan = nv50_disp_chan(object); - struct nv50_disp *disp = chan->root->disp; + struct nv50_disp *disp = chan->disp; switch (type) { case NV50_DISP_CORE_CHANNEL_DMA_V0_NTFY_UEVENT: *pevent = &disp->uevent; @@ -195,34 +212,83 @@ nv50_disp_chan_map(struct nvkm_object *object, void *argv, u32 argc, enum nvkm_object_map *type, u64 *addr, u64 *size) { struct nv50_disp_chan *chan = nv50_disp_chan(object); - struct nv50_disp *disp = chan->root->disp; - struct nvkm_device *device = disp->base.engine.subdev.device; + struct nvkm_device *device = chan->disp->base.engine.subdev.device; + const u64 base = device->func->resource_addr(device, 0); *type = NVKM_OBJECT_MAP_IO; - *addr = device->func->resource_addr(device, 0) + - 0x640000 + (chan->chid.user * 0x1000); - *size = 0x001000; + *addr = base + chan->func->user(chan, size); return 0; } +struct nv50_disp_chan_object { + struct nvkm_oproxy oproxy; + struct nv50_disp *disp; + int hash; +}; + +static void +nv50_disp_chan_child_del_(struct nvkm_oproxy *base) +{ + struct nv50_disp_chan_object *object = + container_of(base, typeof(*object), oproxy); + nvkm_ramht_remove(object->disp->ramht, object->hash); +} + +static const struct nvkm_oproxy_func +nv50_disp_chan_child_func_ = { + .dtor[0] = nv50_disp_chan_child_del_, +}; + static int nv50_disp_chan_child_new(const struct nvkm_oclass *oclass, - void *data, u32 size, struct nvkm_object **pobject) + void *argv, u32 argc, struct nvkm_object **pobject) { struct nv50_disp_chan *chan = nv50_disp_chan(oclass->parent); - return chan->func->child_new(chan, oclass, data, size, pobject); + struct nv50_disp *disp = chan->disp; + struct nvkm_device *device = disp->base.engine.subdev.device; + const struct nvkm_device_oclass *sclass = oclass->priv; + struct nv50_disp_chan_object *object; + int ret; + + if (!(object = kzalloc(sizeof(*object), GFP_KERNEL))) + return -ENOMEM; + nvkm_oproxy_ctor(&nv50_disp_chan_child_func_, oclass, &object->oproxy); + object->disp = disp; + *pobject = &object->oproxy.base; + + ret = sclass->ctor(device, oclass, argv, argc, &object->oproxy.object); + if (ret) + return ret; + + object->hash = chan->func->bind(chan, object->oproxy.object, + oclass->handle); + if (object->hash < 0) + return object->hash; + + return 0; } static int nv50_disp_chan_child_get(struct nvkm_object *object, int index, - struct nvkm_oclass *oclass) + struct nvkm_oclass *sclass) { struct nv50_disp_chan *chan = nv50_disp_chan(object); - if (chan->func->child_get) { - int ret = chan->func->child_get(chan, index, oclass); - if (ret == 0) - oclass->ctor = nv50_disp_chan_child_new; - return ret; + struct nvkm_device *device = chan->disp->base.engine.subdev.device; + const struct nvkm_device_oclass *oclass = NULL; + + if (chan->func->bind) + sclass->engine = nvkm_device_engine(device, NVKM_ENGINE_DMAOBJ); + else + sclass->engine = NULL; + + if (sclass->engine && sclass->engine->func->base.sclass) { + sclass->engine->func->base.sclass(sclass, index, &oclass); + if (oclass) { + sclass->ctor = nv50_disp_chan_child_new, + sclass->priv = oclass; + return 0; + } } + return -EINVAL; } @@ -231,6 +297,7 @@ nv50_disp_chan_fini(struct nvkm_object *object, bool suspend) { struct nv50_disp_chan *chan = nv50_disp_chan(object); chan->func->fini(chan); + chan->func->intr(chan, false); return 0; } @@ -238,6 +305,7 @@ static int nv50_disp_chan_init(struct nvkm_object *object) { struct nv50_disp_chan *chan = nv50_disp_chan(object); + chan->func->intr(chan, true); return chan->func->init(chan); } @@ -245,10 +313,11 @@ static void * nv50_disp_chan_dtor(struct nvkm_object *object) { struct nv50_disp_chan *chan = nv50_disp_chan(object); - struct nv50_disp *disp = chan->root->disp; + struct nv50_disp *disp = chan->disp; if (chan->chid.user >= 0) disp->chan[chan->chid.user] = NULL; - return chan->func->dtor ? chan->func->dtor(chan) : chan; + nvkm_memory_unref(&chan->memory); + return chan; } static const struct nvkm_object_func @@ -264,18 +333,22 @@ nv50_disp_chan = { }; int -nv50_disp_chan_ctor(const struct nv50_disp_chan_func *func, +nv50_disp_chan_new_(const struct nv50_disp_chan_func *func, const struct nv50_disp_chan_mthd *mthd, - struct nv50_disp_root *root, int ctrl, int user, int head, + struct nv50_disp *disp, int ctrl, int user, int head, const struct nvkm_oclass *oclass, - struct nv50_disp_chan *chan) + struct nvkm_object **pobject) { - struct nv50_disp *disp = root->disp; + struct nv50_disp_chan *chan; + + if (!(chan = kzalloc(sizeof(*chan), GFP_KERNEL))) + return -ENOMEM; + *pobject = &chan->object; nvkm_object_ctor(&nv50_disp_chan, oclass, &chan->object); chan->func = func; chan->mthd = mthd; - chan->root = root; + chan->disp = disp; chan->chid.ctrl = ctrl; chan->chid.user = user; chan->head = head; @@ -287,20 +360,3 @@ nv50_disp_chan_ctor(const struct nv50_disp_chan_func *func, disp->chan[chan->chid.user] = chan; return 0; } - -int -nv50_disp_chan_new_(const struct nv50_disp_chan_func *func, - const struct nv50_disp_chan_mthd *mthd, - struct nv50_disp_root *root, int ctrl, int user, int head, - const struct nvkm_oclass *oclass, - struct nvkm_object **pobject) -{ - struct nv50_disp_chan *chan; - - if (!(chan = kzalloc(sizeof(*chan), GFP_KERNEL))) - return -ENOMEM; - *pobject = &chan->object; - - return nv50_disp_chan_ctor(func, mthd, root, ctrl, user, - head, oclass, chan); -} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/channv50.h b/drivers/gpu/drm/nouveau/nvkm/engine/disp/channv50.h index 40681db91a02..adc9d76d09cc 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/channv50.h +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/channv50.h @@ -4,11 +4,12 @@ #define nv50_disp_chan(p) container_of((p), struct nv50_disp_chan, object) #include <core/object.h> #include "nv50.h" +struct nv50_disp_root; struct nv50_disp_chan { const struct nv50_disp_chan_func *func; const struct nv50_disp_chan_mthd *mthd; - struct nv50_disp_root *root; + struct nv50_disp *disp; struct { int ctrl; @@ -17,36 +18,133 @@ struct nv50_disp_chan { int head; struct nvkm_object object; + + struct nvkm_memory *memory; + u64 push; }; struct nv50_disp_chan_func { - void *(*dtor)(struct nv50_disp_chan *); int (*init)(struct nv50_disp_chan *); void (*fini)(struct nv50_disp_chan *); - int (*child_get)(struct nv50_disp_chan *, int index, - struct nvkm_oclass *); - int (*child_new)(struct nv50_disp_chan *, const struct nvkm_oclass *, - void *data, u32 size, struct nvkm_object **); + void (*intr)(struct nv50_disp_chan *, bool en); + u64 (*user)(struct nv50_disp_chan *, u64 *size); + int (*bind)(struct nv50_disp_chan *, struct nvkm_object *, u32 handle); }; -int nv50_disp_chan_ctor(const struct nv50_disp_chan_func *, - const struct nv50_disp_chan_mthd *, - struct nv50_disp_root *, int ctrl, int user, int head, - const struct nvkm_oclass *, struct nv50_disp_chan *); int nv50_disp_chan_new_(const struct nv50_disp_chan_func *, const struct nv50_disp_chan_mthd *, - struct nv50_disp_root *, int ctrl, int user, int head, + struct nv50_disp *, int ctrl, int user, int head, + const struct nvkm_oclass *, struct nvkm_object **); +int nv50_disp_dmac_new_(const struct nv50_disp_chan_func *, + const struct nv50_disp_chan_mthd *, + struct nv50_disp *, int chid, int head, u64 push, const struct nvkm_oclass *, struct nvkm_object **); +void nv50_disp_chan_intr(struct nv50_disp_chan *, bool); +u64 nv50_disp_chan_user(struct nv50_disp_chan *, u64 *); extern const struct nv50_disp_chan_func nv50_disp_pioc_func; -extern const struct nv50_disp_chan_func gf119_disp_pioc_func; - -extern const struct nvkm_event_func nv50_disp_chan_uevent; -int nv50_disp_chan_uevent_ctor(struct nvkm_object *, void *, u32, - struct nvkm_notify *); -void nv50_disp_chan_uevent_send(struct nv50_disp *, int); +extern const struct nv50_disp_chan_func nv50_disp_dmac_func; +int nv50_disp_dmac_bind(struct nv50_disp_chan *, struct nvkm_object *, u32); +extern const struct nv50_disp_chan_func nv50_disp_core_func; -extern const struct nvkm_event_func gf119_disp_chan_uevent; +void gf119_disp_chan_intr(struct nv50_disp_chan *, bool); +extern const struct nv50_disp_chan_func gf119_disp_pioc_func; +extern const struct nv50_disp_chan_func gf119_disp_dmac_func; +void gf119_disp_dmac_fini(struct nv50_disp_chan *); +int gf119_disp_dmac_bind(struct nv50_disp_chan *, struct nvkm_object *, u32); +extern const struct nv50_disp_chan_func gf119_disp_core_func; +void gf119_disp_core_fini(struct nv50_disp_chan *); + +extern const struct nv50_disp_chan_func gp102_disp_dmac_func; + +u64 gv100_disp_chan_user(struct nv50_disp_chan *, u64 *); +int gv100_disp_dmac_init(struct nv50_disp_chan *); +void gv100_disp_dmac_fini(struct nv50_disp_chan *); +int gv100_disp_dmac_bind(struct nv50_disp_chan *, struct nvkm_object *, u32); + +int nv50_disp_curs_new_(const struct nv50_disp_chan_func *, + struct nv50_disp *, int ctrl, int user, + const struct nvkm_oclass *, void *argv, u32 argc, + struct nvkm_object **); +int nv50_disp_oimm_new_(const struct nv50_disp_chan_func *, + struct nv50_disp *, int ctrl, int user, + const struct nvkm_oclass *, void *argv, u32 argc, + struct nvkm_object **); +int nv50_disp_base_new_(const struct nv50_disp_chan_func *, + const struct nv50_disp_chan_mthd *, + struct nv50_disp *, int chid, + const struct nvkm_oclass *, void *argv, u32 argc, + struct nvkm_object **); +int nv50_disp_core_new_(const struct nv50_disp_chan_func *, + const struct nv50_disp_chan_mthd *, + struct nv50_disp *, int chid, + const struct nvkm_oclass *oclass, void *argv, u32 argc, + struct nvkm_object **); +int nv50_disp_ovly_new_(const struct nv50_disp_chan_func *, + const struct nv50_disp_chan_mthd *, + struct nv50_disp *, int chid, + const struct nvkm_oclass *, void *argv, u32 argc, + struct nvkm_object **); + +int nv50_disp_curs_new(const struct nvkm_oclass *, void *, u32, + struct nv50_disp *, struct nvkm_object **); +int nv50_disp_oimm_new(const struct nvkm_oclass *, void *, u32, + struct nv50_disp *, struct nvkm_object **); +int nv50_disp_base_new(const struct nvkm_oclass *, void *, u32, + struct nv50_disp *, struct nvkm_object **); +int nv50_disp_core_new(const struct nvkm_oclass *, void *, u32, + struct nv50_disp *, struct nvkm_object **); +int nv50_disp_ovly_new(const struct nvkm_oclass *, void *, u32, + struct nv50_disp *, struct nvkm_object **); + +int g84_disp_base_new(const struct nvkm_oclass *, void *, u32, + struct nv50_disp *, struct nvkm_object **); +int g84_disp_core_new(const struct nvkm_oclass *, void *, u32, + struct nv50_disp *, struct nvkm_object **); +int g84_disp_ovly_new(const struct nvkm_oclass *, void *, u32, + struct nv50_disp *, struct nvkm_object **); + +int g94_disp_core_new(const struct nvkm_oclass *, void *, u32, + struct nv50_disp *, struct nvkm_object **); + +int gt200_disp_ovly_new(const struct nvkm_oclass *, void *, u32, + struct nv50_disp *, struct nvkm_object **); + +int gf119_disp_curs_new(const struct nvkm_oclass *, void *, u32, + struct nv50_disp *, struct nvkm_object **); +int gf119_disp_oimm_new(const struct nvkm_oclass *, void *, u32, + struct nv50_disp *, struct nvkm_object **); +int gf119_disp_base_new(const struct nvkm_oclass *, void *, u32, + struct nv50_disp *, struct nvkm_object **); +int gf119_disp_core_new(const struct nvkm_oclass *, void *, u32, + struct nv50_disp *, struct nvkm_object **); +int gf119_disp_ovly_new(const struct nvkm_oclass *, void *, u32, + struct nv50_disp *, struct nvkm_object **); + +int gk104_disp_core_new(const struct nvkm_oclass *, void *, u32, + struct nv50_disp *, struct nvkm_object **); +int gk104_disp_ovly_new(const struct nvkm_oclass *, void *, u32, + struct nv50_disp *, struct nvkm_object **); + +int gp102_disp_curs_new(const struct nvkm_oclass *, void *, u32, + struct nv50_disp *, struct nvkm_object **); +int gp102_disp_oimm_new(const struct nvkm_oclass *, void *, u32, + struct nv50_disp *, struct nvkm_object **); +int gp102_disp_base_new(const struct nvkm_oclass *, void *, u32, + struct nv50_disp *, struct nvkm_object **); +int gp102_disp_core_new(const struct nvkm_oclass *, void *, u32, + struct nv50_disp *, struct nvkm_object **); +int gp102_disp_ovly_new(const struct nvkm_oclass *, void *, u32, + struct nv50_disp *, struct nvkm_object **); + +int gv100_disp_curs_new(const struct nvkm_oclass *, void *, u32, + struct nv50_disp *, struct nvkm_object **); +int gv100_disp_wimm_new(const struct nvkm_oclass *, void *, u32, + struct nv50_disp *, struct nvkm_object **); +int gv100_disp_core_new(const struct nvkm_oclass *, void *, u32, + struct nv50_disp *, struct nvkm_object **); +int gv100_disp_wndw_new(const struct nvkm_oclass *, void *, u32, + struct nv50_disp *, struct nvkm_object **); struct nv50_disp_mthd_list { u32 mthd; @@ -76,64 +174,18 @@ extern const struct nv50_disp_mthd_list nv50_disp_core_mthd_sor; extern const struct nv50_disp_mthd_list nv50_disp_core_mthd_pior; extern const struct nv50_disp_mthd_list nv50_disp_base_mthd_image; -extern const struct nv50_disp_chan_mthd g84_disp_core_chan_mthd; +extern const struct nv50_disp_chan_mthd g84_disp_core_mthd; extern const struct nv50_disp_mthd_list g84_disp_core_mthd_dac; extern const struct nv50_disp_mthd_list g84_disp_core_mthd_head; -extern const struct nv50_disp_chan_mthd g84_disp_base_chan_mthd; -extern const struct nv50_disp_chan_mthd g84_disp_ovly_chan_mthd; -extern const struct nv50_disp_chan_mthd g94_disp_core_chan_mthd; +extern const struct nv50_disp_chan_mthd g94_disp_core_mthd; extern const struct nv50_disp_mthd_list gf119_disp_core_mthd_base; extern const struct nv50_disp_mthd_list gf119_disp_core_mthd_dac; extern const struct nv50_disp_mthd_list gf119_disp_core_mthd_sor; extern const struct nv50_disp_mthd_list gf119_disp_core_mthd_pior; -extern const struct nv50_disp_chan_mthd gf119_disp_base_chan_mthd; - -extern const struct nv50_disp_chan_mthd gk104_disp_core_chan_mthd; -extern const struct nv50_disp_chan_mthd gk104_disp_ovly_chan_mthd; - -struct nv50_disp_pioc_oclass { - int (*ctor)(const struct nv50_disp_chan_func *, - const struct nv50_disp_chan_mthd *, - struct nv50_disp_root *, int ctrl, int user, - const struct nvkm_oclass *, void *data, u32 size, - struct nvkm_object **); - struct nvkm_sclass base; - const struct nv50_disp_chan_func *func; - const struct nv50_disp_chan_mthd *mthd; - struct { - int ctrl; - int user; - } chid; -}; - -extern const struct nv50_disp_pioc_oclass nv50_disp_oimm_oclass; -extern const struct nv50_disp_pioc_oclass nv50_disp_curs_oclass; - -extern const struct nv50_disp_pioc_oclass g84_disp_oimm_oclass; -extern const struct nv50_disp_pioc_oclass g84_disp_curs_oclass; - -extern const struct nv50_disp_pioc_oclass gt215_disp_oimm_oclass; -extern const struct nv50_disp_pioc_oclass gt215_disp_curs_oclass; - -extern const struct nv50_disp_pioc_oclass gf119_disp_oimm_oclass; -extern const struct nv50_disp_pioc_oclass gf119_disp_curs_oclass; - -extern const struct nv50_disp_pioc_oclass gk104_disp_oimm_oclass; -extern const struct nv50_disp_pioc_oclass gk104_disp_curs_oclass; - -extern const struct nv50_disp_pioc_oclass gp102_disp_oimm_oclass; -extern const struct nv50_disp_pioc_oclass gp102_disp_curs_oclass; +extern const struct nv50_disp_chan_mthd gf119_disp_base_mthd; -int nv50_disp_curs_new(const struct nv50_disp_chan_func *, - const struct nv50_disp_chan_mthd *, - struct nv50_disp_root *, int ctrl, int user, - const struct nvkm_oclass *, void *data, u32 size, - struct nvkm_object **); -int nv50_disp_oimm_new(const struct nv50_disp_chan_func *, - const struct nv50_disp_chan_mthd *, - struct nv50_disp_root *, int ctrl, int user, - const struct nvkm_oclass *, void *data, u32 size, - struct nvkm_object **); +extern const struct nv50_disp_chan_mthd gk104_disp_core_mthd; +extern const struct nv50_disp_chan_mthd gk104_disp_ovly_mthd; #endif diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/coreg84.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/coreg84.c index 1baa5c34b327..cfc54aad3e7c 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/coreg84.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/coreg84.c @@ -21,10 +21,7 @@ * * Authors: Ben Skeggs */ -#include "dmacnv50.h" -#include "rootnv50.h" - -#include <nvif/class.h> +#include "channv50.h" const struct nv50_disp_mthd_list g84_disp_core_mthd_dac = { @@ -91,7 +88,7 @@ g84_disp_core_mthd_head = { }; const struct nv50_disp_chan_mthd -g84_disp_core_chan_mthd = { +g84_disp_core_mthd = { .name = "Core", .addr = 0x000000, .prev = 0x000004, @@ -105,13 +102,10 @@ g84_disp_core_chan_mthd = { } }; -const struct nv50_disp_dmac_oclass -g84_disp_core_oclass = { - .base.oclass = G82_DISP_CORE_CHANNEL_DMA, - .base.minver = 0, - .base.maxver = 0, - .ctor = nv50_disp_core_new, - .func = &nv50_disp_core_func, - .mthd = &g84_disp_core_chan_mthd, - .chid = 0, -}; +int +g84_disp_core_new(const struct nvkm_oclass *oclass, void *argv, u32 argc, + struct nv50_disp *disp, struct nvkm_object **pobject) +{ + return nv50_disp_core_new_(&nv50_disp_core_func, &g84_disp_core_mthd, + disp, 0, oclass, argv, argc, pobject); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/coreg94.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/coreg94.c index c65c9f3ff69f..e911925f1182 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/coreg94.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/coreg94.c @@ -21,10 +21,7 @@ * * Authors: Ben Skeggs */ -#include "dmacnv50.h" -#include "rootnv50.h" - -#include <nvif/class.h> +#include "channv50.h" static const struct nv50_disp_mthd_list g94_disp_core_mthd_sor = { @@ -37,7 +34,7 @@ g94_disp_core_mthd_sor = { }; const struct nv50_disp_chan_mthd -g94_disp_core_chan_mthd = { +g94_disp_core_mthd = { .name = "Core", .addr = 0x000000, .prev = 0x000004, @@ -51,13 +48,10 @@ g94_disp_core_chan_mthd = { } }; -const struct nv50_disp_dmac_oclass -g94_disp_core_oclass = { - .base.oclass = GT206_DISP_CORE_CHANNEL_DMA, - .base.minver = 0, - .base.maxver = 0, - .ctor = nv50_disp_core_new, - .func = &nv50_disp_core_func, - .mthd = &g94_disp_core_chan_mthd, - .chid = 0, -}; +int +g94_disp_core_new(const struct nvkm_oclass *oclass, void *argv, u32 argc, + struct nv50_disp *disp, struct nvkm_object **pobject) +{ + return nv50_disp_core_new_(&nv50_disp_core_func, &g94_disp_core_mthd, + disp, 0, oclass, argv, argc, pobject); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/coregf119.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/coregf119.c index 21fbf89b6319..d162b9cf4eac 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/coregf119.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/coregf119.c @@ -21,15 +21,10 @@ * * Authors: Ben Skeggs */ -#include "dmacnv50.h" -#include "rootnv50.h" +#include "channv50.h" -#include <core/client.h> #include <subdev/timer.h> -#include <nvif/class.h> -#include <nvif/unpack.h> - const struct nv50_disp_mthd_list gf119_disp_core_mthd_base = { .mthd = 0x0000, @@ -157,7 +152,7 @@ gf119_disp_core_mthd_head = { }; static const struct nv50_disp_chan_mthd -gf119_disp_core_chan_mthd = { +gf119_disp_core_mthd = { .name = "Core", .addr = 0x000000, .prev = -0x020000, @@ -172,10 +167,9 @@ gf119_disp_core_chan_mthd = { }; void -gf119_disp_core_fini(struct nv50_disp_dmac *chan) +gf119_disp_core_fini(struct nv50_disp_chan *chan) { - struct nv50_disp *disp = chan->base.root->disp; - struct nvkm_subdev *subdev = &disp->base.engine.subdev; + struct nvkm_subdev *subdev = &chan->disp->base.engine.subdev; struct nvkm_device *device = subdev->device; /* deactivate channel */ @@ -188,22 +182,14 @@ gf119_disp_core_fini(struct nv50_disp_dmac *chan) nvkm_error(subdev, "core fini: %08x\n", nvkm_rd32(device, 0x610490)); } - - /* disable error reporting and completion notification */ - nvkm_mask(device, 0x610090, 0x00000001, 0x00000000); - nvkm_mask(device, 0x6100a0, 0x00000001, 0x00000000); } static int -gf119_disp_core_init(struct nv50_disp_dmac *chan) +gf119_disp_core_init(struct nv50_disp_chan *chan) { - struct nv50_disp *disp = chan->base.root->disp; - struct nvkm_subdev *subdev = &disp->base.engine.subdev; + struct nvkm_subdev *subdev = &chan->disp->base.engine.subdev; struct nvkm_device *device = subdev->device; - /* enable error reporting */ - nvkm_mask(device, 0x6100a0, 0x00000001, 0x00000001); - /* initialise channel for dma command submission */ nvkm_wr32(device, 0x610494, chan->push); nvkm_wr32(device, 0x610498, 0x00010000); @@ -225,20 +211,19 @@ gf119_disp_core_init(struct nv50_disp_dmac *chan) return 0; } -const struct nv50_disp_dmac_func +const struct nv50_disp_chan_func gf119_disp_core_func = { .init = gf119_disp_core_init, .fini = gf119_disp_core_fini, + .intr = gf119_disp_chan_intr, + .user = nv50_disp_chan_user, .bind = gf119_disp_dmac_bind, }; -const struct nv50_disp_dmac_oclass -gf119_disp_core_oclass = { - .base.oclass = GF110_DISP_CORE_CHANNEL_DMA, - .base.minver = 0, - .base.maxver = 0, - .ctor = nv50_disp_core_new, - .func = &gf119_disp_core_func, - .mthd = &gf119_disp_core_chan_mthd, - .chid = 0, -}; +int +gf119_disp_core_new(const struct nvkm_oclass *oclass, void *argv, u32 argc, + struct nv50_disp *disp, struct nvkm_object **pobject) +{ + return nv50_disp_core_new_(&gf119_disp_core_func, &gf119_disp_core_mthd, + disp, 0, oclass, argv, argc, pobject); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/coregk104.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/coregk104.c index 088ab222e823..5c800174e079 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/coregk104.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/coregk104.c @@ -21,10 +21,7 @@ * * Authors: Ben Skeggs */ -#include "dmacnv50.h" -#include "rootnv50.h" - -#include <nvif/class.h> +#include "channv50.h" static const struct nv50_disp_mthd_list gk104_disp_core_mthd_head = { @@ -106,7 +103,7 @@ gk104_disp_core_mthd_head = { }; const struct nv50_disp_chan_mthd -gk104_disp_core_chan_mthd = { +gk104_disp_core_mthd = { .name = "Core", .addr = 0x000000, .prev = -0x020000, @@ -120,13 +117,10 @@ gk104_disp_core_chan_mthd = { } }; -const struct nv50_disp_dmac_oclass -gk104_disp_core_oclass = { - .base.oclass = GK104_DISP_CORE_CHANNEL_DMA, - .base.minver = 0, - .base.maxver = 0, - .ctor = nv50_disp_core_new, - .func = &gf119_disp_core_func, - .mthd = &gk104_disp_core_chan_mthd, - .chid = 0, -}; +int +gk104_disp_core_new(const struct nvkm_oclass *oclass, void *argv, u32 argc, + struct nv50_disp *disp, struct nvkm_object **pobject) +{ + return nv50_disp_core_new_(&gf119_disp_core_func, &gk104_disp_core_mthd, + disp, 0, oclass, argv, argc, pobject); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/coregk110.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/coregk110.c deleted file mode 100644 index df0f45c20108..000000000000 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/coregk110.c +++ /dev/null @@ -1,38 +0,0 @@ -/* - * Copyright 2015 Red Hat Inc. - * - * Permission is hereby granted, free of charge, to any person obtaining a - * copy of this software and associated documentation files (the "Software"), - * to deal in the Software without restriction, including without limitation - * the rights to use, copy, modify, merge, publish, distribute, sublicense, - * and/or sell copies of the Software, and to permit persons to whom the - * Software is furnished to do so, subject to the following conditions: - * - * The above copyright notice and this permission notice shall be included in - * all copies or substantial portions of the Software. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL - * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR - * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, - * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR - * OTHER DEALINGS IN THE SOFTWARE. - * - * Authors: Ben Skeggs <bskeggs@redhat.com> - */ -#include "dmacnv50.h" -#include "rootnv50.h" - -#include <nvif/class.h> - -const struct nv50_disp_dmac_oclass -gk110_disp_core_oclass = { - .base.oclass = GK110_DISP_CORE_CHANNEL_DMA, - .base.minver = 0, - .base.maxver = 0, - .ctor = nv50_disp_core_new, - .func = &gf119_disp_core_func, - .mthd = &gk104_disp_core_chan_mthd, - .chid = 0, -}; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/coregm107.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/coregm107.c deleted file mode 100644 index 9e27f8fd98b6..000000000000 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/coregm107.c +++ /dev/null @@ -1,38 +0,0 @@ -/* - * Copyright 2015 Red Hat Inc. - * - * Permission is hereby granted, free of charge, to any person obtaining a - * copy of this software and associated documentation files (the "Software"), - * to deal in the Software without restriction, including without limitation - * the rights to use, copy, modify, merge, publish, distribute, sublicense, - * and/or sell copies of the Software, and to permit persons to whom the - * Software is furnished to do so, subject to the following conditions: - * - * The above copyright notice and this permission notice shall be included in - * all copies or substantial portions of the Software. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL - * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR - * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, - * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR - * OTHER DEALINGS IN THE SOFTWARE. - * - * Authors: Ben Skeggs <bskeggs@redhat.com> - */ -#include "dmacnv50.h" -#include "rootnv50.h" - -#include <nvif/class.h> - -const struct nv50_disp_dmac_oclass -gm107_disp_core_oclass = { - .base.oclass = GM107_DISP_CORE_CHANNEL_DMA, - .base.minver = 0, - .base.maxver = 0, - .ctor = nv50_disp_core_new, - .func = &gf119_disp_core_func, - .mthd = &gk104_disp_core_chan_mthd, - .chid = 0, -}; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/coregm200.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/coregm200.c deleted file mode 100644 index bb23a8658ac0..000000000000 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/coregm200.c +++ /dev/null @@ -1,38 +0,0 @@ -/* - * Copyright 2015 Red Hat Inc. - * - * Permission is hereby granted, free of charge, to any person obtaining a - * copy of this software and associated documentation files (the "Software"), - * to deal in the Software without restriction, including without limitation - * the rights to use, copy, modify, merge, publish, distribute, sublicense, - * and/or sell copies of the Software, and to permit persons to whom the - * Software is furnished to do so, subject to the following conditions: - * - * The above copyright notice and this permission notice shall be included in - * all copies or substantial portions of the Software. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL - * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR - * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, - * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR - * OTHER DEALINGS IN THE SOFTWARE. - * - * Authors: Ben Skeggs <bskeggs@redhat.com> - */ -#include "dmacnv50.h" -#include "rootnv50.h" - -#include <nvif/class.h> - -const struct nv50_disp_dmac_oclass -gm200_disp_core_oclass = { - .base.oclass = GM200_DISP_CORE_CHANNEL_DMA, - .base.minver = 0, - .base.maxver = 0, - .ctor = nv50_disp_core_new, - .func = &gf119_disp_core_func, - .mthd = &gk104_disp_core_chan_mthd, - .chid = 0, -}; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/coregp100.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/coregp100.c deleted file mode 100644 index d5dff6619d4d..000000000000 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/coregp100.c +++ /dev/null @@ -1,38 +0,0 @@ -/* - * Copyright 2015 Red Hat Inc. - * - * Permission is hereby granted, free of charge, to any person obtaining a - * copy of this software and associated documentation files (the "Software"), - * to deal in the Software without restriction, including without limitation - * the rights to use, copy, modify, merge, publish, distribute, sublicense, - * and/or sell copies of the Software, and to permit persons to whom the - * Software is furnished to do so, subject to the following conditions: - * - * The above copyright notice and this permission notice shall be included in - * all copies or substantial portions of the Software. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL - * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR - * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, - * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR - * OTHER DEALINGS IN THE SOFTWARE. - * - * Authors: Ben Skeggs <bskeggs@redhat.com> - */ -#include "dmacnv50.h" -#include "rootnv50.h" - -#include <nvif/class.h> - -const struct nv50_disp_dmac_oclass -gp100_disp_core_oclass = { - .base.oclass = GP100_DISP_CORE_CHANNEL_DMA, - .base.minver = 0, - .base.maxver = 0, - .ctor = nv50_disp_core_new, - .func = &gf119_disp_core_func, - .mthd = &gk104_disp_core_chan_mthd, - .chid = 0, -}; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/coregp102.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/coregp102.c index b0df4b752b8c..5b7f993c73c7 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/coregp102.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/coregp102.c @@ -21,23 +21,16 @@ * * Authors: Ben Skeggs <bskeggs@redhat.com> */ -#include "dmacnv50.h" -#include "rootnv50.h" +#include "channv50.h" #include <subdev/timer.h> -#include <nvif/class.h> - static int -gp102_disp_core_init(struct nv50_disp_dmac *chan) +gp102_disp_core_init(struct nv50_disp_chan *chan) { - struct nv50_disp *disp = chan->base.root->disp; - struct nvkm_subdev *subdev = &disp->base.engine.subdev; + struct nvkm_subdev *subdev = &chan->disp->base.engine.subdev; struct nvkm_device *device = subdev->device; - /* enable error reporting */ - nvkm_mask(device, 0x6100a0, 0x00000001, 0x00000001); - /* initialise channel for dma command submission */ nvkm_wr32(device, 0x611494, chan->push); nvkm_wr32(device, 0x611498, 0x00010000); @@ -59,20 +52,19 @@ gp102_disp_core_init(struct nv50_disp_dmac *chan) return 0; } -static const struct nv50_disp_dmac_func +static const struct nv50_disp_chan_func gp102_disp_core_func = { .init = gp102_disp_core_init, .fini = gf119_disp_core_fini, + .intr = gf119_disp_chan_intr, + .user = nv50_disp_chan_user, .bind = gf119_disp_dmac_bind, }; -const struct nv50_disp_dmac_oclass -gp102_disp_core_oclass = { - .base.oclass = GP102_DISP_CORE_CHANNEL_DMA, - .base.minver = 0, - .base.maxver = 0, - .ctor = nv50_disp_core_new, - .func = &gp102_disp_core_func, - .mthd = &gk104_disp_core_chan_mthd, - .chid = 0, -}; +int +gp102_disp_core_new(const struct nvkm_oclass *oclass, void *argv, u32 argc, + struct nv50_disp *disp, struct nvkm_object **pobject) +{ + return nv50_disp_core_new_(&gp102_disp_core_func, &gk104_disp_core_mthd, + disp, 0, oclass, argv, argc, pobject); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/coregt215.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/coregt215.c deleted file mode 100644 index 8f5ba2018975..000000000000 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/coregt215.c +++ /dev/null @@ -1,38 +0,0 @@ -/* - * Copyright 2015 Red Hat Inc. - * - * Permission is hereby granted, free of charge, to any person obtaining a - * copy of this software and associated documentation files (the "Software"), - * to deal in the Software without restriction, including without limitation - * the rights to use, copy, modify, merge, publish, distribute, sublicense, - * and/or sell copies of the Software, and to permit persons to whom the - * Software is furnished to do so, subject to the following conditions: - * - * The above copyright notice and this permission notice shall be included in - * all copies or substantial portions of the Software. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL - * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR - * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, - * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR - * OTHER DEALINGS IN THE SOFTWARE. - * - * Authors: Ben Skeggs <bskeggs@redhat.com> - */ -#include "dmacnv50.h" -#include "rootnv50.h" - -#include <nvif/class.h> - -const struct nv50_disp_dmac_oclass -gt215_disp_core_oclass = { - .base.oclass = GT214_DISP_CORE_CHANNEL_DMA, - .base.minver = 0, - .base.maxver = 0, - .ctor = nv50_disp_core_new, - .func = &nv50_disp_core_func, - .mthd = &g94_disp_core_chan_mthd, - .chid = 0, -}; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/coregv100.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/coregv100.c new file mode 100644 index 000000000000..4592d0e69fec --- /dev/null +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/coregv100.c @@ -0,0 +1,204 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "channv50.h" + +#include <subdev/timer.h> + +const struct nv50_disp_mthd_list +gv100_disp_core_mthd_base = { + .mthd = 0x0000, + .addr = 0x000000, + .data = { + { 0x0200, 0x680200 }, + { 0x0208, 0x680208 }, + { 0x020c, 0x68020c }, + { 0x0210, 0x680210 }, + { 0x0214, 0x680214 }, + { 0x0218, 0x680218 }, + { 0x021c, 0x68021c }, + {} + } +}; + +const struct nv50_disp_mthd_list +gv100_disp_core_mthd_sor = { + .mthd = 0x0020, + .addr = 0x000020, + .data = { + { 0x0300, 0x680300 }, + { 0x0304, 0x680304 }, + { 0x0308, 0x680308 }, + { 0x030c, 0x68030c }, + {} + } +}; + +static const struct nv50_disp_mthd_list +gv100_disp_core_mthd_wndw = { + .mthd = 0x0080, + .addr = 0x000080, + .data = { + { 0x1000, 0x681000 }, + { 0x1004, 0x681004 }, + { 0x1008, 0x681008 }, + { 0x100c, 0x68100c }, + { 0x1010, 0x681010 }, + {} + } +}; + +static const struct nv50_disp_mthd_list +gv100_disp_core_mthd_head = { + .mthd = 0x0400, + .addr = 0x000400, + .data = { + { 0x2000, 0x682000 }, + { 0x2004, 0x682004 }, + { 0x2008, 0x682008 }, + { 0x200c, 0x68200c }, + { 0x2014, 0x682014 }, + { 0x2018, 0x682018 }, + { 0x201c, 0x68201c }, + { 0x2020, 0x682020 }, + { 0x2028, 0x682028 }, + { 0x202c, 0x68202c }, + { 0x2030, 0x682030 }, + { 0x2038, 0x682038 }, + { 0x203c, 0x68203c }, + { 0x2048, 0x682048 }, + { 0x204c, 0x68204c }, + { 0x2050, 0x682050 }, + { 0x2054, 0x682054 }, + { 0x2058, 0x682058 }, + { 0x205c, 0x68205c }, + { 0x2060, 0x682060 }, + { 0x2064, 0x682064 }, + { 0x2068, 0x682068 }, + { 0x206c, 0x68206c }, + { 0x2070, 0x682070 }, + { 0x2074, 0x682074 }, + { 0x2078, 0x682078 }, + { 0x207c, 0x68207c }, + { 0x2080, 0x682080 }, + { 0x2088, 0x682088 }, + { 0x2090, 0x682090 }, + { 0x209c, 0x68209c }, + { 0x20a0, 0x6820a0 }, + { 0x20a4, 0x6820a4 }, + { 0x20a8, 0x6820a8 }, + { 0x20ac, 0x6820ac }, + { 0x218c, 0x68218c }, + { 0x2194, 0x682194 }, + { 0x2198, 0x682198 }, + { 0x219c, 0x68219c }, + { 0x21a0, 0x6821a0 }, + { 0x21a4, 0x6821a4 }, + { 0x2214, 0x682214 }, + { 0x2218, 0x682218 }, + {} + } +}; + +static const struct nv50_disp_chan_mthd +gv100_disp_core_mthd = { + .name = "Core", + .addr = 0x000000, + .prev = 0x008000, + .data = { + { "Global", 1, &gv100_disp_core_mthd_base }, + { "SOR", 4, &gv100_disp_core_mthd_sor }, + { "WINDOW", 8, &gv100_disp_core_mthd_wndw }, + { "HEAD", 4, &gv100_disp_core_mthd_head }, + {} + } +}; + +static int +gv100_disp_core_idle(struct nv50_disp_chan *chan) +{ + struct nvkm_device *device = chan->disp->base.engine.subdev.device; + nvkm_msec(device, 2000, + u32 stat = nvkm_rd32(device, 0x610630); + if ((stat & 0x001f0000) == 0x000b0000) + return 0; + ); + return -EBUSY; +} + +static u64 +gv100_disp_core_user(struct nv50_disp_chan *chan, u64 *psize) +{ + *psize = 0x10000; + return 0x680000; +} + +static void +gv100_disp_core_intr(struct nv50_disp_chan *chan, bool en) +{ + struct nvkm_device *device = chan->disp->base.engine.subdev.device; + const u32 mask = 0x00000001; + const u32 data = en ? mask : 0; + nvkm_mask(device, 0x611dac, mask, data); +} + +static void +gv100_disp_core_fini(struct nv50_disp_chan *chan) +{ + struct nvkm_device *device = chan->disp->base.engine.subdev.device; + nvkm_mask(device, 0x6104e0, 0x00000010, 0x00000000); + gv100_disp_core_idle(chan); + nvkm_mask(device, 0x6104e0, 0x00000002, 0x00000000); +} + +static int +gv100_disp_core_init(struct nv50_disp_chan *chan) +{ + struct nvkm_subdev *subdev = &chan->disp->base.engine.subdev; + struct nvkm_device *device = subdev->device; + + nvkm_wr32(device, 0x610b24, lower_32_bits(chan->push)); + nvkm_wr32(device, 0x610b20, upper_32_bits(chan->push)); + nvkm_wr32(device, 0x610b28, 0x00000001); + nvkm_wr32(device, 0x610b2c, 0x00000040); + + nvkm_mask(device, 0x6104e0, 0x00000010, 0x00000010); + nvkm_wr32(device, 0x680000, 0x00000000); + nvkm_wr32(device, 0x6104e0, 0x00000013); + return gv100_disp_core_idle(chan); +} + +static const struct nv50_disp_chan_func +gv100_disp_core = { + .init = gv100_disp_core_init, + .fini = gv100_disp_core_fini, + .intr = gv100_disp_core_intr, + .user = gv100_disp_core_user, + .bind = gv100_disp_dmac_bind, +}; + +int +gv100_disp_core_new(const struct nvkm_oclass *oclass, void *argv, u32 argc, + struct nv50_disp *disp, struct nvkm_object **pobject) +{ + return nv50_disp_core_new_(&gv100_disp_core, &gv100_disp_core_mthd, + disp, 0, oclass, argv, argc, pobject); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/corenv50.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/corenv50.c index b547c8b833ca..55db9a22b4be 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/corenv50.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/corenv50.c @@ -21,32 +21,30 @@ * * Authors: Ben Skeggs */ -#include "dmacnv50.h" -#include "rootnv50.h" +#include "channv50.h" #include <core/client.h> #include <subdev/timer.h> -#include <nvif/class.h> #include <nvif/cl507d.h> #include <nvif/unpack.h> int -nv50_disp_core_new(const struct nv50_disp_dmac_func *func, - const struct nv50_disp_chan_mthd *mthd, - struct nv50_disp_root *root, int chid, - const struct nvkm_oclass *oclass, void *data, u32 size, - struct nvkm_object **pobject) +nv50_disp_core_new_(const struct nv50_disp_chan_func *func, + const struct nv50_disp_chan_mthd *mthd, + struct nv50_disp *disp, int chid, + const struct nvkm_oclass *oclass, void *argv, u32 argc, + struct nvkm_object **pobject) { union { struct nv50_disp_core_channel_dma_v0 v0; - } *args = data; + } *args = argv; struct nvkm_object *parent = oclass->parent; u64 push; int ret = -ENOSYS; - nvif_ioctl(parent, "create disp core channel dma size %d\n", size); - if (!(ret = nvif_unpack(ret, &data, &size, args->v0, 0, 0, false))) { + nvif_ioctl(parent, "create disp core channel dma size %d\n", argc); + if (!(ret = nvif_unpack(ret, &argv, &argc, args->v0, 0, 0, false))) { nvif_ioctl(parent, "create disp core channel dma vers %d " "pushbuf %016llx\n", args->v0.version, args->v0.pushbuf); @@ -54,7 +52,7 @@ nv50_disp_core_new(const struct nv50_disp_dmac_func *func, } else return ret; - return nv50_disp_dmac_new_(func, mthd, root, chid, 0, + return nv50_disp_dmac_new_(func, mthd, disp, chid, 0, push, oclass, pobject); } @@ -151,7 +149,7 @@ nv50_disp_core_mthd_head = { }; static const struct nv50_disp_chan_mthd -nv50_disp_core_chan_mthd = { +nv50_disp_core_mthd = { .name = "Core", .addr = 0x000000, .prev = 0x000004, @@ -166,10 +164,9 @@ nv50_disp_core_chan_mthd = { }; static void -nv50_disp_core_fini(struct nv50_disp_dmac *chan) +nv50_disp_core_fini(struct nv50_disp_chan *chan) { - struct nv50_disp *disp = chan->base.root->disp; - struct nvkm_subdev *subdev = &disp->base.engine.subdev; + struct nvkm_subdev *subdev = &chan->disp->base.engine.subdev; struct nvkm_device *device = subdev->device; /* deactivate channel */ @@ -182,21 +179,14 @@ nv50_disp_core_fini(struct nv50_disp_dmac *chan) nvkm_error(subdev, "core fini: %08x\n", nvkm_rd32(device, 0x610200)); } - - /* disable error reporting and completion notifications */ - nvkm_mask(device, 0x610028, 0x00010001, 0x00000000); } static int -nv50_disp_core_init(struct nv50_disp_dmac *chan) +nv50_disp_core_init(struct nv50_disp_chan *chan) { - struct nv50_disp *disp = chan->base.root->disp; - struct nvkm_subdev *subdev = &disp->base.engine.subdev; + struct nvkm_subdev *subdev = &chan->disp->base.engine.subdev; struct nvkm_device *device = subdev->device; - /* enable error reporting */ - nvkm_mask(device, 0x610028, 0x00010000, 0x00010000); - /* attempt to unstick channel from some unknown state */ if ((nvkm_rd32(device, 0x610200) & 0x009f0000) == 0x00020000) nvkm_mask(device, 0x610200, 0x00800000, 0x00800000); @@ -224,20 +214,19 @@ nv50_disp_core_init(struct nv50_disp_dmac *chan) return 0; } -const struct nv50_disp_dmac_func +const struct nv50_disp_chan_func nv50_disp_core_func = { .init = nv50_disp_core_init, .fini = nv50_disp_core_fini, + .intr = nv50_disp_chan_intr, + .user = nv50_disp_chan_user, .bind = nv50_disp_dmac_bind, }; -const struct nv50_disp_dmac_oclass -nv50_disp_core_oclass = { - .base.oclass = NV50_DISP_CORE_CHANNEL_DMA, - .base.minver = 0, - .base.maxver = 0, - .ctor = nv50_disp_core_new, - .func = &nv50_disp_core_func, - .mthd = &nv50_disp_core_chan_mthd, - .chid = 0, -}; +int +nv50_disp_core_new(const struct nvkm_oclass *oclass, void *argv, u32 argc, + struct nv50_disp *disp, struct nvkm_object **pobject) +{ + return nv50_disp_core_new_(&nv50_disp_core_func, &nv50_disp_core_mthd, + disp, 0, oclass, argv, argc, pobject); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/cursgf119.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/cursgf119.c index 2be6fb052c65..cdda3658dcb3 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/cursgf119.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/cursgf119.c @@ -22,16 +22,11 @@ * Authors: Ben Skeggs */ #include "channv50.h" -#include "rootnv50.h" -#include <nvif/class.h> - -const struct nv50_disp_pioc_oclass -gf119_disp_curs_oclass = { - .base.oclass = GF110_DISP_CURSOR, - .base.minver = 0, - .base.maxver = 0, - .ctor = nv50_disp_curs_new, - .func = &gf119_disp_pioc_func, - .chid = { 13, 13 }, -}; +int +gf119_disp_curs_new(const struct nvkm_oclass *oclass, void *argv, u32 argc, + struct nv50_disp *disp, struct nvkm_object **pobject) +{ + return nv50_disp_curs_new_(&gf119_disp_pioc_func, disp, 13, 13, + oclass, argv, argc, pobject); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/cursgp102.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/cursgp102.c index e958210d8105..1a4601f975e6 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/cursgp102.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/cursgp102.c @@ -22,16 +22,11 @@ * Authors: Ben Skeggs <bskeggs@redhat.com> */ #include "channv50.h" -#include "rootnv50.h" -#include <nvif/class.h> - -const struct nv50_disp_pioc_oclass -gp102_disp_curs_oclass = { - .base.oclass = GK104_DISP_CURSOR, - .base.minver = 0, - .base.maxver = 0, - .ctor = nv50_disp_curs_new, - .func = &gf119_disp_pioc_func, - .chid = { 13, 17 }, -}; +int +gp102_disp_curs_new(const struct nvkm_oclass *oclass, void *argv, u32 argc, + struct nv50_disp *disp, struct nvkm_object **pobject) +{ + return nv50_disp_curs_new_(&gf119_disp_pioc_func, disp, 13, 17, + oclass, argv, argc, pobject); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/cursgv100.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/cursgv100.c new file mode 100644 index 000000000000..a3e4f6900245 --- /dev/null +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/cursgv100.c @@ -0,0 +1,81 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "channv50.h" + +#include <subdev/timer.h> + +static int +gv100_disp_curs_idle(struct nv50_disp_chan *chan) +{ + struct nvkm_device *device = chan->disp->base.engine.subdev.device; + const u32 soff = (chan->chid.ctrl - 1) * 0x04; + nvkm_msec(device, 2000, + u32 stat = nvkm_rd32(device, 0x610664 + soff); + if ((stat & 0x00070000) == 0x00040000) + return 0; + ); + return -EBUSY; +} + +static void +gv100_disp_curs_intr(struct nv50_disp_chan *chan, bool en) +{ + struct nvkm_device *device = chan->disp->base.engine.subdev.device; + const u32 mask = 0x00010000 << chan->head; + const u32 data = en ? mask : 0; + nvkm_mask(device, 0x611dac, mask, data); +} + +static void +gv100_disp_curs_fini(struct nv50_disp_chan *chan) +{ + struct nvkm_device *device = chan->disp->base.engine.subdev.device; + const u32 hoff = chan->chid.ctrl * 4; + nvkm_mask(device, 0x6104e0 + hoff, 0x00000010, 0x00000010); + gv100_disp_curs_idle(chan); + nvkm_mask(device, 0x6104e0 + hoff, 0x00000001, 0x00000000); +} + +static int +gv100_disp_curs_init(struct nv50_disp_chan *chan) +{ + struct nvkm_subdev *subdev = &chan->disp->base.engine.subdev; + struct nvkm_device *device = subdev->device; + nvkm_wr32(device, 0x6104e0 + chan->chid.ctrl * 4, 0x00000001); + return gv100_disp_curs_idle(chan); +} + +static const struct nv50_disp_chan_func +gv100_disp_curs = { + .init = gv100_disp_curs_init, + .fini = gv100_disp_curs_fini, + .intr = gv100_disp_curs_intr, + .user = gv100_disp_chan_user, +}; + +int +gv100_disp_curs_new(const struct nvkm_oclass *oclass, void *argv, u32 argc, + struct nv50_disp *disp, struct nvkm_object **pobject) +{ + return nv50_disp_curs_new_(&gv100_disp_curs, disp, 73, 73, + oclass, argv, argc, pobject); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/cursnv50.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/cursnv50.c index ab51121b7982..d29758504a5f 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/cursnv50.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/cursnv50.c @@ -23,30 +23,26 @@ */ #include "channv50.h" #include "head.h" -#include "rootnv50.h" #include <core/client.h> -#include <nvif/class.h> #include <nvif/cl507a.h> #include <nvif/unpack.h> int -nv50_disp_curs_new(const struct nv50_disp_chan_func *func, - const struct nv50_disp_chan_mthd *mthd, - struct nv50_disp_root *root, int ctrl, int user, - const struct nvkm_oclass *oclass, void *data, u32 size, - struct nvkm_object **pobject) +nv50_disp_curs_new_(const struct nv50_disp_chan_func *func, + struct nv50_disp *disp, int ctrl, int user, + const struct nvkm_oclass *oclass, void *argv, u32 argc, + struct nvkm_object **pobject) { union { struct nv50_disp_cursor_v0 v0; - } *args = data; + } *args = argv; struct nvkm_object *parent = oclass->parent; - struct nv50_disp *disp = root->disp; int head, ret = -ENOSYS; - nvif_ioctl(parent, "create disp cursor size %d\n", size); - if (!(ret = nvif_unpack(ret, &data, &size, args->v0, 0, 0, false))) { + nvif_ioctl(parent, "create disp cursor size %d\n", argc); + if (!(ret = nvif_unpack(ret, &argv, &argc, args->v0, 0, 0, false))) { nvif_ioctl(parent, "create disp cursor vers %d head %d\n", args->v0.version, args->v0.head); if (!nvkm_head_find(&disp->base, args->v0.head)) @@ -55,16 +51,14 @@ nv50_disp_curs_new(const struct nv50_disp_chan_func *func, } else return ret; - return nv50_disp_chan_new_(func, mthd, root, ctrl + head, user + head, + return nv50_disp_chan_new_(func, NULL, disp, ctrl + head, user + head, head, oclass, pobject); } -const struct nv50_disp_pioc_oclass -nv50_disp_curs_oclass = { - .base.oclass = NV50_DISP_CURSOR, - .base.minver = 0, - .base.maxver = 0, - .ctor = nv50_disp_curs_new, - .func = &nv50_disp_pioc_func, - .chid = { 7, 7 }, -}; +int +nv50_disp_curs_new(const struct nvkm_oclass *oclass, void *argv, u32 argc, + struct nv50_disp *disp, struct nvkm_object **pobject) +{ + return nv50_disp_curs_new_(&nv50_disp_pioc_func, disp, 7, 7, + oclass, argv, argc, pobject); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/dacgf119.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/dacgf119.c index dbd032ef352a..71a94777ea2e 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/dacgf119.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/dacgf119.c @@ -58,8 +58,13 @@ gf119_dac = { int gf119_dac_new(struct nvkm_disp *disp, int id) { - struct nvkm_device *device = disp->engine.subdev.device; - if (!(nvkm_rd32(device, 0x612004) & (0x00000010 << id))) - return 0; return nvkm_ior_new_(&gf119_dac, disp, DAC, id); } + +int +gf119_dac_cnt(struct nvkm_disp *disp, unsigned long *pmask) +{ + struct nvkm_device *device = disp->engine.subdev.device; + *pmask = (nvkm_rd32(device, 0x612004) & 0x000000f0) >> 4; + return 4; +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/dacnv50.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/dacnv50.c index 85e692b12260..558012db35f8 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/dacnv50.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/dacnv50.c @@ -109,8 +109,13 @@ nv50_dac = { int nv50_dac_new(struct nvkm_disp *disp, int id) { - struct nvkm_device *device = disp->engine.subdev.device; - if (!(nvkm_rd32(device, 0x610184) & (0x00100000 << id))) - return 0; return nvkm_ior_new_(&nv50_dac, disp, DAC, id); } + +int +nv50_dac_cnt(struct nvkm_disp *disp, unsigned long *pmask) +{ + struct nvkm_device *device = disp->engine.subdev.device; + *pmask = (nvkm_rd32(device, 0x610184) & 0x00700000) >> 20; + return 3; +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/dmacgf119.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/dmacgf119.c index ce7cd74fbd5d..edf7dd0d931d 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/dmacgf119.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/dmacgf119.c @@ -21,29 +21,27 @@ * * Authors: Ben Skeggs */ -#include "dmacnv50.h" -#include "rootnv50.h" +#include "channv50.h" #include <core/ramht.h> #include <subdev/timer.h> int -gf119_disp_dmac_bind(struct nv50_disp_dmac *chan, +gf119_disp_dmac_bind(struct nv50_disp_chan *chan, struct nvkm_object *object, u32 handle) { - return nvkm_ramht_insert(chan->base.root->ramht, object, - chan->base.chid.user, -9, handle, - chan->base.chid.user << 27 | 0x00000001); + return nvkm_ramht_insert(chan->disp->ramht, object, + chan->chid.user, -9, handle, + chan->chid.user << 27 | 0x00000001); } void -gf119_disp_dmac_fini(struct nv50_disp_dmac *chan) +gf119_disp_dmac_fini(struct nv50_disp_chan *chan) { - struct nv50_disp *disp = chan->base.root->disp; - struct nvkm_subdev *subdev = &disp->base.engine.subdev; + struct nvkm_subdev *subdev = &chan->disp->base.engine.subdev; struct nvkm_device *device = subdev->device; - int ctrl = chan->base.chid.ctrl; - int user = chan->base.chid.user; + int ctrl = chan->chid.ctrl; + int user = chan->chid.user; /* deactivate channel */ nvkm_mask(device, 0x610490 + (ctrl * 0x0010), 0x00001010, 0x00001000); @@ -55,23 +53,15 @@ gf119_disp_dmac_fini(struct nv50_disp_dmac *chan) nvkm_error(subdev, "ch %d fini: %08x\n", user, nvkm_rd32(device, 0x610490 + (ctrl * 0x10))); } - - /* disable error reporting and completion notification */ - nvkm_mask(device, 0x610090, 0x00000001 << user, 0x00000000); - nvkm_mask(device, 0x6100a0, 0x00000001 << user, 0x00000000); } static int -gf119_disp_dmac_init(struct nv50_disp_dmac *chan) +gf119_disp_dmac_init(struct nv50_disp_chan *chan) { - struct nv50_disp *disp = chan->base.root->disp; - struct nvkm_subdev *subdev = &disp->base.engine.subdev; + struct nvkm_subdev *subdev = &chan->disp->base.engine.subdev; struct nvkm_device *device = subdev->device; - int ctrl = chan->base.chid.ctrl; - int user = chan->base.chid.user; - - /* enable error reporting */ - nvkm_mask(device, 0x6100a0, 0x00000001 << user, 0x00000001 << user); + int ctrl = chan->chid.ctrl; + int user = chan->chid.user; /* initialise channel for dma command submission */ nvkm_wr32(device, 0x610494 + (ctrl * 0x0010), chan->push); @@ -94,9 +84,11 @@ gf119_disp_dmac_init(struct nv50_disp_dmac *chan) return 0; } -const struct nv50_disp_dmac_func +const struct nv50_disp_chan_func gf119_disp_dmac_func = { .init = gf119_disp_dmac_init, .fini = gf119_disp_dmac_fini, + .intr = gf119_disp_chan_intr, + .user = nv50_disp_chan_user, .bind = gf119_disp_dmac_bind, }; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/dmacgp102.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/dmacgp102.c index cdead9500343..f21a433199aa 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/dmacgp102.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/dmacgp102.c @@ -21,22 +21,17 @@ * * Authors: Ben Skeggs <bskeggs@redhat.com> */ -#include "dmacnv50.h" -#include "rootnv50.h" +#include "channv50.h" #include <subdev/timer.h> static int -gp102_disp_dmac_init(struct nv50_disp_dmac *chan) +gp102_disp_dmac_init(struct nv50_disp_chan *chan) { - struct nv50_disp *disp = chan->base.root->disp; - struct nvkm_subdev *subdev = &disp->base.engine.subdev; + struct nvkm_subdev *subdev = &chan->disp->base.engine.subdev; struct nvkm_device *device = subdev->device; - int ctrl = chan->base.chid.ctrl; - int user = chan->base.chid.user; - - /* enable error reporting */ - nvkm_mask(device, 0x6100a0, 0x00000001 << user, 0x00000001 << user); + int ctrl = chan->chid.ctrl; + int user = chan->chid.user; /* initialise channel for dma command submission */ nvkm_wr32(device, 0x611494 + (ctrl * 0x0010), chan->push); @@ -59,9 +54,11 @@ gp102_disp_dmac_init(struct nv50_disp_dmac *chan) return 0; } -const struct nv50_disp_dmac_func +const struct nv50_disp_chan_func gp102_disp_dmac_func = { .init = gp102_disp_dmac_init, .fini = gf119_disp_dmac_fini, + .intr = gf119_disp_chan_intr, + .user = nv50_disp_chan_user, .bind = gf119_disp_dmac_bind, }; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/dmacgv100.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/dmacgv100.c new file mode 100644 index 000000000000..eac0e42da354 --- /dev/null +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/dmacgv100.c @@ -0,0 +1,77 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "channv50.h" + +#include <core/ramht.h> +#include <subdev/timer.h> + +static int +gv100_disp_dmac_idle(struct nv50_disp_chan *chan) +{ + struct nvkm_device *device = chan->disp->base.engine.subdev.device; + const u32 soff = (chan->chid.ctrl - 1) * 0x04; + nvkm_msec(device, 2000, + u32 stat = nvkm_rd32(device, 0x610664 + soff); + if ((stat & 0x000f0000) == 0x00040000) + return 0; + ); + return -EBUSY; +} + +int +gv100_disp_dmac_bind(struct nv50_disp_chan *chan, + struct nvkm_object *object, u32 handle) +{ + return nvkm_ramht_insert(chan->disp->ramht, object, + chan->chid.user, -9, handle, + chan->chid.user << 25 | 0x00000040); +} + +void +gv100_disp_dmac_fini(struct nv50_disp_chan *chan) +{ + struct nvkm_device *device = chan->disp->base.engine.subdev.device; + const u32 coff = chan->chid.ctrl * 0x04; + nvkm_mask(device, 0x6104e0 + coff, 0x00000010, 0x00000000); + gv100_disp_dmac_idle(chan); + nvkm_mask(device, 0x6104e0 + coff, 0x00000002, 0x00000000); +} + +int +gv100_disp_dmac_init(struct nv50_disp_chan *chan) +{ + struct nvkm_subdev *subdev = &chan->disp->base.engine.subdev; + struct nvkm_device *device = subdev->device; + const u32 uoff = (chan->chid.ctrl - 1) * 0x1000; + const u32 poff = chan->chid.ctrl * 0x10; + const u32 coff = chan->chid.ctrl * 0x04; + + nvkm_wr32(device, 0x610b24 + poff, lower_32_bits(chan->push)); + nvkm_wr32(device, 0x610b20 + poff, upper_32_bits(chan->push)); + nvkm_wr32(device, 0x610b28 + poff, 0x00000001); + nvkm_wr32(device, 0x610b2c + poff, 0x00000040); + + nvkm_mask(device, 0x6104e0 + coff, 0x00000010, 0x00000010); + nvkm_wr32(device, 0x690000 + uoff, 0x00000000); + nvkm_wr32(device, 0x6104e0 + coff, 0x00000013); + return gv100_disp_dmac_idle(chan); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/dmacnv50.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/dmacnv50.c index 070ec5e18fdb..9e8a9d7a9b68 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/dmacnv50.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/dmacnv50.c @@ -21,176 +21,68 @@ * * Authors: Ben Skeggs */ -#include "dmacnv50.h" -#include "rootnv50.h" +#include "channv50.h" #include <core/client.h> -#include <core/oproxy.h> #include <core/ramht.h> #include <subdev/fb.h> +#include <subdev/mmu.h> #include <subdev/timer.h> #include <engine/dma.h> -struct nv50_disp_dmac_object { - struct nvkm_oproxy oproxy; - struct nv50_disp_root *root; - int hash; -}; - -static void -nv50_disp_dmac_child_del_(struct nvkm_oproxy *base) -{ - struct nv50_disp_dmac_object *object = - container_of(base, typeof(*object), oproxy); - nvkm_ramht_remove(object->root->ramht, object->hash); -} - -static const struct nvkm_oproxy_func -nv50_disp_dmac_child_func_ = { - .dtor[0] = nv50_disp_dmac_child_del_, -}; - -static int -nv50_disp_dmac_child_new_(struct nv50_disp_chan *base, - const struct nvkm_oclass *oclass, - void *data, u32 size, struct nvkm_object **pobject) -{ - struct nv50_disp_dmac *chan = nv50_disp_dmac(base); - struct nv50_disp_root *root = chan->base.root; - struct nvkm_device *device = root->disp->base.engine.subdev.device; - const struct nvkm_device_oclass *sclass = oclass->priv; - struct nv50_disp_dmac_object *object; - int ret; - - if (!(object = kzalloc(sizeof(*object), GFP_KERNEL))) - return -ENOMEM; - nvkm_oproxy_ctor(&nv50_disp_dmac_child_func_, oclass, &object->oproxy); - object->root = root; - *pobject = &object->oproxy.base; - - ret = sclass->ctor(device, oclass, data, size, &object->oproxy.object); - if (ret) - return ret; - - object->hash = chan->func->bind(chan, object->oproxy.object, - oclass->handle); - if (object->hash < 0) - return object->hash; - - return 0; -} - -static int -nv50_disp_dmac_child_get_(struct nv50_disp_chan *base, int index, - struct nvkm_oclass *sclass) -{ - struct nv50_disp_dmac *chan = nv50_disp_dmac(base); - struct nv50_disp *disp = chan->base.root->disp; - struct nvkm_device *device = disp->base.engine.subdev.device; - const struct nvkm_device_oclass *oclass = NULL; - - sclass->engine = nvkm_device_engine(device, NVKM_ENGINE_DMAOBJ); - if (sclass->engine && sclass->engine->func->base.sclass) { - sclass->engine->func->base.sclass(sclass, index, &oclass); - if (oclass) { - sclass->priv = oclass; - return 0; - } - } - - return -EINVAL; -} - -static void -nv50_disp_dmac_fini_(struct nv50_disp_chan *base) -{ - struct nv50_disp_dmac *chan = nv50_disp_dmac(base); - chan->func->fini(chan); -} - -static int -nv50_disp_dmac_init_(struct nv50_disp_chan *base) -{ - struct nv50_disp_dmac *chan = nv50_disp_dmac(base); - return chan->func->init(chan); -} - -static void * -nv50_disp_dmac_dtor_(struct nv50_disp_chan *base) -{ - return nv50_disp_dmac(base); -} - -static const struct nv50_disp_chan_func -nv50_disp_dmac_func_ = { - .dtor = nv50_disp_dmac_dtor_, - .init = nv50_disp_dmac_init_, - .fini = nv50_disp_dmac_fini_, - .child_get = nv50_disp_dmac_child_get_, - .child_new = nv50_disp_dmac_child_new_, -}; - int -nv50_disp_dmac_new_(const struct nv50_disp_dmac_func *func, +nv50_disp_dmac_new_(const struct nv50_disp_chan_func *func, const struct nv50_disp_chan_mthd *mthd, - struct nv50_disp_root *root, int chid, int head, u64 push, + struct nv50_disp *disp, int chid, int head, u64 push, const struct nvkm_oclass *oclass, struct nvkm_object **pobject) { struct nvkm_client *client = oclass->client; - struct nvkm_dmaobj *dmaobj; - struct nv50_disp_dmac *chan; + struct nv50_disp_chan *chan; int ret; - if (!(chan = kzalloc(sizeof(*chan), GFP_KERNEL))) - return -ENOMEM; - *pobject = &chan->base.object; - chan->func = func; - - ret = nv50_disp_chan_ctor(&nv50_disp_dmac_func_, mthd, root, - chid, chid, head, oclass, &chan->base); + ret = nv50_disp_chan_new_(func, mthd, disp, chid, chid, head, oclass, + pobject); + chan = nv50_disp_chan(*pobject); if (ret) return ret; - dmaobj = nvkm_dmaobj_search(client, push); - if (IS_ERR(dmaobj)) - return PTR_ERR(dmaobj); + chan->memory = nvkm_umem_search(client, push); + if (IS_ERR(chan->memory)) + return PTR_ERR(chan->memory); - if (dmaobj->limit - dmaobj->start != 0xfff) + if (nvkm_memory_size(chan->memory) < 0x1000) return -EINVAL; - switch (dmaobj->target) { - case NV_MEM_TARGET_VRAM: - chan->push = 0x00000001 | dmaobj->start >> 8; - break; - case NV_MEM_TARGET_PCI_NOSNOOP: - chan->push = 0x00000003 | dmaobj->start >> 8; - break; + switch (nvkm_memory_target(chan->memory)) { + case NVKM_MEM_TARGET_VRAM: chan->push = 0x00000001; break; + case NVKM_MEM_TARGET_NCOH: chan->push = 0x00000002; break; + case NVKM_MEM_TARGET_HOST: chan->push = 0x00000003; break; default: return -EINVAL; } + chan->push |= nvkm_memory_addr(chan->memory) >> 8; return 0; } int -nv50_disp_dmac_bind(struct nv50_disp_dmac *chan, +nv50_disp_dmac_bind(struct nv50_disp_chan *chan, struct nvkm_object *object, u32 handle) { - return nvkm_ramht_insert(chan->base.root->ramht, object, - chan->base.chid.user, -10, handle, - chan->base.chid.user << 28 | - chan->base.chid.user); + return nvkm_ramht_insert(chan->disp->ramht, object, + chan->chid.user, -10, handle, + chan->chid.user << 28 | + chan->chid.user); } static void -nv50_disp_dmac_fini(struct nv50_disp_dmac *chan) +nv50_disp_dmac_fini(struct nv50_disp_chan *chan) { - struct nv50_disp *disp = chan->base.root->disp; - struct nvkm_subdev *subdev = &disp->base.engine.subdev; + struct nvkm_subdev *subdev = &chan->disp->base.engine.subdev; struct nvkm_device *device = subdev->device; - int ctrl = chan->base.chid.ctrl; - int user = chan->base.chid.user; + int ctrl = chan->chid.ctrl; + int user = chan->chid.user; /* deactivate channel */ nvkm_mask(device, 0x610200 + (ctrl * 0x0010), 0x00001010, 0x00001000); @@ -202,22 +94,15 @@ nv50_disp_dmac_fini(struct nv50_disp_dmac *chan) nvkm_error(subdev, "ch %d fini timeout, %08x\n", user, nvkm_rd32(device, 0x610200 + (ctrl * 0x10))); } - - /* disable error reporting and completion notifications */ - nvkm_mask(device, 0x610028, 0x00010001 << user, 0x00000000 << user); } static int -nv50_disp_dmac_init(struct nv50_disp_dmac *chan) +nv50_disp_dmac_init(struct nv50_disp_chan *chan) { - struct nv50_disp *disp = chan->base.root->disp; - struct nvkm_subdev *subdev = &disp->base.engine.subdev; + struct nvkm_subdev *subdev = &chan->disp->base.engine.subdev; struct nvkm_device *device = subdev->device; - int ctrl = chan->base.chid.ctrl; - int user = chan->base.chid.user; - - /* enable error reporting */ - nvkm_mask(device, 0x610028, 0x00010000 << user, 0x00010000 << user); + int ctrl = chan->chid.ctrl; + int user = chan->chid.user; /* initialise channel for dma command submission */ nvkm_wr32(device, 0x610204 + (ctrl * 0x0010), chan->push); @@ -240,9 +125,11 @@ nv50_disp_dmac_init(struct nv50_disp_dmac *chan) return 0; } -const struct nv50_disp_dmac_func +const struct nv50_disp_chan_func nv50_disp_dmac_func = { .init = nv50_disp_dmac_init, .fini = nv50_disp_dmac_fini, + .intr = nv50_disp_chan_intr, + .user = nv50_disp_chan_user, .bind = nv50_disp_dmac_bind, }; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/dmacnv50.h b/drivers/gpu/drm/nouveau/nvkm/engine/disp/dmacnv50.h deleted file mode 100644 index f9b98211da6a..000000000000 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/dmacnv50.h +++ /dev/null @@ -1,102 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0 */ -#ifndef __NV50_DISP_DMAC_H__ -#define __NV50_DISP_DMAC_H__ -#define nv50_disp_dmac(p) container_of((p), struct nv50_disp_dmac, base) -#include "channv50.h" - -struct nv50_disp_dmac { - const struct nv50_disp_dmac_func *func; - struct nv50_disp_chan base; - u32 push; -}; - -struct nv50_disp_dmac_func { - int (*init)(struct nv50_disp_dmac *); - void (*fini)(struct nv50_disp_dmac *); - int (*bind)(struct nv50_disp_dmac *, struct nvkm_object *, u32 handle); -}; - -int nv50_disp_dmac_new_(const struct nv50_disp_dmac_func *, - const struct nv50_disp_chan_mthd *, - struct nv50_disp_root *, int chid, int head, u64 push, - const struct nvkm_oclass *, struct nvkm_object **); - -extern const struct nv50_disp_dmac_func nv50_disp_dmac_func; -int nv50_disp_dmac_bind(struct nv50_disp_dmac *, struct nvkm_object *, u32); -extern const struct nv50_disp_dmac_func nv50_disp_core_func; - -extern const struct nv50_disp_dmac_func gf119_disp_dmac_func; -void gf119_disp_dmac_fini(struct nv50_disp_dmac *); -int gf119_disp_dmac_bind(struct nv50_disp_dmac *, struct nvkm_object *, u32); -extern const struct nv50_disp_dmac_func gf119_disp_core_func; -void gf119_disp_core_fini(struct nv50_disp_dmac *); - -extern const struct nv50_disp_dmac_func gp102_disp_dmac_func; - -struct nv50_disp_dmac_oclass { - int (*ctor)(const struct nv50_disp_dmac_func *, - const struct nv50_disp_chan_mthd *, - struct nv50_disp_root *, int chid, - const struct nvkm_oclass *, void *data, u32 size, - struct nvkm_object **); - struct nvkm_sclass base; - const struct nv50_disp_dmac_func *func; - const struct nv50_disp_chan_mthd *mthd; - int chid; -}; - -int nv50_disp_core_new(const struct nv50_disp_dmac_func *, - const struct nv50_disp_chan_mthd *, - struct nv50_disp_root *, int chid, - const struct nvkm_oclass *oclass, void *data, u32 size, - struct nvkm_object **); -int nv50_disp_base_new(const struct nv50_disp_dmac_func *, - const struct nv50_disp_chan_mthd *, - struct nv50_disp_root *, int chid, - const struct nvkm_oclass *oclass, void *data, u32 size, - struct nvkm_object **); -int nv50_disp_ovly_new(const struct nv50_disp_dmac_func *, - const struct nv50_disp_chan_mthd *, - struct nv50_disp_root *, int chid, - const struct nvkm_oclass *oclass, void *data, u32 size, - struct nvkm_object **); - -extern const struct nv50_disp_dmac_oclass nv50_disp_core_oclass; -extern const struct nv50_disp_dmac_oclass nv50_disp_base_oclass; -extern const struct nv50_disp_dmac_oclass nv50_disp_ovly_oclass; - -extern const struct nv50_disp_dmac_oclass g84_disp_core_oclass; -extern const struct nv50_disp_dmac_oclass g84_disp_base_oclass; -extern const struct nv50_disp_dmac_oclass g84_disp_ovly_oclass; - -extern const struct nv50_disp_dmac_oclass g94_disp_core_oclass; - -extern const struct nv50_disp_dmac_oclass gt200_disp_core_oclass; -extern const struct nv50_disp_dmac_oclass gt200_disp_base_oclass; -extern const struct nv50_disp_dmac_oclass gt200_disp_ovly_oclass; - -extern const struct nv50_disp_dmac_oclass gt215_disp_core_oclass; -extern const struct nv50_disp_dmac_oclass gt215_disp_base_oclass; -extern const struct nv50_disp_dmac_oclass gt215_disp_ovly_oclass; - -extern const struct nv50_disp_dmac_oclass gf119_disp_core_oclass; -extern const struct nv50_disp_dmac_oclass gf119_disp_base_oclass; -extern const struct nv50_disp_dmac_oclass gf119_disp_ovly_oclass; - -extern const struct nv50_disp_dmac_oclass gk104_disp_core_oclass; -extern const struct nv50_disp_dmac_oclass gk104_disp_base_oclass; -extern const struct nv50_disp_dmac_oclass gk104_disp_ovly_oclass; - -extern const struct nv50_disp_dmac_oclass gk110_disp_core_oclass; -extern const struct nv50_disp_dmac_oclass gk110_disp_base_oclass; - -extern const struct nv50_disp_dmac_oclass gm107_disp_core_oclass; - -extern const struct nv50_disp_dmac_oclass gm200_disp_core_oclass; - -extern const struct nv50_disp_dmac_oclass gp100_disp_core_oclass; - -extern const struct nv50_disp_dmac_oclass gp102_disp_core_oclass; -extern const struct nv50_disp_dmac_oclass gp102_disp_base_oclass; -extern const struct nv50_disp_dmac_oclass gp102_disp_ovly_oclass; -#endif diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/g84.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/g84.c index 842e1b72ee42..731f188fc1ee 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/g84.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/g84.c @@ -28,18 +28,20 @@ static const struct nv50_disp_func g84_disp = { + .init = nv50_disp_init, + .fini = nv50_disp_fini, .intr = nv50_disp_intr, .uevent = &nv50_disp_chan_uevent, .super = nv50_disp_super, .root = &g84_disp_root_oclass, - .head.new = nv50_head_new, - .dac = { .nr = 3, .new = nv50_dac_new }, - .sor = { .nr = 2, .new = g84_sor_new }, - .pior = { .nr = 3, .new = nv50_pior_new }, + .head = { .cnt = nv50_head_cnt, .new = nv50_head_new }, + .dac = { .cnt = nv50_dac_cnt, .new = nv50_dac_new }, + .sor = { .cnt = nv50_sor_cnt, .new = g84_sor_new }, + .pior = { .cnt = nv50_pior_cnt, .new = nv50_pior_new }, }; int g84_disp_new(struct nvkm_device *device, int index, struct nvkm_disp **pdisp) { - return nv50_disp_new_(&g84_disp, device, index, 2, pdisp); + return nv50_disp_new_(&g84_disp, device, index, pdisp); } diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/g94.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/g94.c index d184e6ab8918..def54fe1951e 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/g94.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/g94.c @@ -28,18 +28,20 @@ static const struct nv50_disp_func g94_disp = { + .init = nv50_disp_init, + .fini = nv50_disp_fini, .intr = nv50_disp_intr, .uevent = &nv50_disp_chan_uevent, .super = nv50_disp_super, .root = &g94_disp_root_oclass, - .head.new = nv50_head_new, - .dac = { .nr = 3, .new = nv50_dac_new }, - .sor = { .nr = 4, .new = g94_sor_new }, - .pior = { .nr = 3, .new = nv50_pior_new }, + .head = { .cnt = nv50_head_cnt, .new = nv50_head_new }, + .dac = { .cnt = nv50_dac_cnt, .new = nv50_dac_new }, + .sor = { .cnt = g94_sor_cnt, .new = g94_sor_new }, + .pior = { .cnt = nv50_pior_cnt, .new = nv50_pior_new }, }; int g94_disp_new(struct nvkm_device *device, int index, struct nvkm_disp **pdisp) { - return nv50_disp_new_(&g94_disp, device, index, 2, pdisp); + return nv50_disp_new_(&g94_disp, device, index, pdisp); } diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/gf119.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/gf119.c index d8765b57180b..794e90982641 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/gf119.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/gf119.c @@ -24,8 +24,12 @@ #include "nv50.h" #include "head.h" #include "ior.h" +#include "channv50.h" #include "rootnv50.h" +#include <core/ramht.h> +#include <subdev/timer.h> + void gf119_disp_super(struct work_struct *work) { @@ -164,28 +168,99 @@ gf119_disp_intr(struct nv50_disp *disp) } } +void +gf119_disp_fini(struct nv50_disp *disp) +{ + struct nvkm_device *device = disp->base.engine.subdev.device; + /* disable all interrupts */ + nvkm_wr32(device, 0x6100b0, 0x00000000); +} + int -gf119_disp_new_(const struct nv50_disp_func *func, struct nvkm_device *device, - int index, struct nvkm_disp **pdisp) +gf119_disp_init(struct nv50_disp *disp) { - u32 heads = nvkm_rd32(device, 0x022448); - return nv50_disp_new_(func, device, index, heads, pdisp); + struct nvkm_device *device = disp->base.engine.subdev.device; + struct nvkm_head *head; + u32 tmp; + int i; + + /* The below segments of code copying values from one register to + * another appear to inform EVO of the display capabilities or + * something similar. + */ + + /* ... CRTC caps */ + list_for_each_entry(head, &disp->base.head, head) { + const u32 hoff = head->id * 0x800; + tmp = nvkm_rd32(device, 0x616104 + hoff); + nvkm_wr32(device, 0x6101b4 + hoff, tmp); + tmp = nvkm_rd32(device, 0x616108 + hoff); + nvkm_wr32(device, 0x6101b8 + hoff, tmp); + tmp = nvkm_rd32(device, 0x61610c + hoff); + nvkm_wr32(device, 0x6101bc + hoff, tmp); + } + + /* ... DAC caps */ + for (i = 0; i < disp->dac.nr; i++) { + tmp = nvkm_rd32(device, 0x61a000 + (i * 0x800)); + nvkm_wr32(device, 0x6101c0 + (i * 0x800), tmp); + } + + /* ... SOR caps */ + for (i = 0; i < disp->sor.nr; i++) { + tmp = nvkm_rd32(device, 0x61c000 + (i * 0x800)); + nvkm_wr32(device, 0x6301c4 + (i * 0x800), tmp); + } + + /* steal display away from vbios, or something like that */ + if (nvkm_rd32(device, 0x6100ac) & 0x00000100) { + nvkm_wr32(device, 0x6100ac, 0x00000100); + nvkm_mask(device, 0x6194e8, 0x00000001, 0x00000000); + if (nvkm_msec(device, 2000, + if (!(nvkm_rd32(device, 0x6194e8) & 0x00000002)) + break; + ) < 0) + return -EBUSY; + } + + /* point at display engine memory area (hash table, objects) */ + nvkm_wr32(device, 0x610010, (disp->inst->addr >> 8) | 9); + + /* enable supervisor interrupts, disable everything else */ + nvkm_wr32(device, 0x610090, 0x00000000); + nvkm_wr32(device, 0x6100a0, 0x00000000); + nvkm_wr32(device, 0x6100b0, 0x00000307); + + /* disable underflow reporting, preventing an intermittent issue + * on some gk104 boards where the production vbios left this + * setting enabled by default. + * + * ftp://download.nvidia.com/open-gpu-doc/gk104-disable-underflow-reporting/1/gk104-disable-underflow-reporting.txt + */ + list_for_each_entry(head, &disp->base.head, head) { + const u32 hoff = head->id * 0x800; + nvkm_mask(device, 0x616308 + hoff, 0x00000111, 0x00000010); + } + + return 0; } static const struct nv50_disp_func gf119_disp = { + .init = gf119_disp_init, + .fini = gf119_disp_fini, .intr = gf119_disp_intr, .intr_error = gf119_disp_intr_error, .uevent = &gf119_disp_chan_uevent, .super = gf119_disp_super, .root = &gf119_disp_root_oclass, - .head.new = gf119_head_new, - .dac = { .nr = 3, .new = gf119_dac_new }, - .sor = { .nr = 4, .new = gf119_sor_new }, + .head = { .cnt = gf119_head_cnt, .new = gf119_head_new }, + .dac = { .cnt = gf119_dac_cnt, .new = gf119_dac_new }, + .sor = { .cnt = gf119_sor_cnt, .new = gf119_sor_new }, }; int gf119_disp_new(struct nvkm_device *device, int index, struct nvkm_disp **pdisp) { - return gf119_disp_new_(&gf119_disp, device, index, pdisp); + return nv50_disp_new_(&gf119_disp, device, index, pdisp); } diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/gk104.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/gk104.c index e8fe9f315d64..4c3439b1a62d 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/gk104.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/gk104.c @@ -28,18 +28,20 @@ static const struct nv50_disp_func gk104_disp = { + .init = gf119_disp_init, + .fini = gf119_disp_fini, .intr = gf119_disp_intr, .intr_error = gf119_disp_intr_error, .uevent = &gf119_disp_chan_uevent, .super = gf119_disp_super, .root = &gk104_disp_root_oclass, - .head.new = gf119_head_new, - .dac = { .nr = 3, .new = gf119_dac_new }, - .sor = { .nr = 4, .new = gk104_sor_new }, + .head = { .cnt = gf119_head_cnt, .new = gf119_head_new }, + .dac = { .cnt = gf119_dac_cnt, .new = gf119_dac_new }, + .sor = { .cnt = gf119_sor_cnt, .new = gk104_sor_new }, }; int gk104_disp_new(struct nvkm_device *device, int index, struct nvkm_disp **pdisp) { - return gf119_disp_new_(&gk104_disp, device, index, pdisp); + return nv50_disp_new_(&gk104_disp, device, index, pdisp); } diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/gk110.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/gk110.c index 769687502e7a..bc6f4750c942 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/gk110.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/gk110.c @@ -28,18 +28,20 @@ static const struct nv50_disp_func gk110_disp = { + .init = gf119_disp_init, + .fini = gf119_disp_fini, .intr = gf119_disp_intr, .intr_error = gf119_disp_intr_error, .uevent = &gf119_disp_chan_uevent, .super = gf119_disp_super, .root = &gk110_disp_root_oclass, - .head.new = gf119_head_new, - .dac = { .nr = 3, .new = gf119_dac_new }, - .sor = { .nr = 4, .new = gk104_sor_new }, + .head = { .cnt = gf119_head_cnt, .new = gf119_head_new }, + .dac = { .cnt = gf119_dac_cnt, .new = gf119_dac_new }, + .sor = { .cnt = gf119_sor_cnt, .new = gk104_sor_new }, }; int gk110_disp_new(struct nvkm_device *device, int index, struct nvkm_disp **pdisp) { - return gf119_disp_new_(&gk110_disp, device, index, pdisp); + return nv50_disp_new_(&gk110_disp, device, index, pdisp); } diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/gm107.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/gm107.c index ede70e5d188e..031cf6b03a76 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/gm107.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/gm107.c @@ -28,18 +28,20 @@ static const struct nv50_disp_func gm107_disp = { + .init = gf119_disp_init, + .fini = gf119_disp_fini, .intr = gf119_disp_intr, .intr_error = gf119_disp_intr_error, .uevent = &gf119_disp_chan_uevent, .super = gf119_disp_super, .root = &gm107_disp_root_oclass, - .head.new = gf119_head_new, - .dac = { .nr = 3, .new = gf119_dac_new }, - .sor = { .nr = 4, .new = gm107_sor_new }, + .head = { .cnt = gf119_head_cnt, .new = gf119_head_new }, + .dac = { .cnt = gf119_dac_cnt, .new = gf119_dac_new }, + .sor = { .cnt = gf119_sor_cnt, .new = gm107_sor_new }, }; int gm107_disp_new(struct nvkm_device *device, int index, struct nvkm_disp **pdisp) { - return gf119_disp_new_(&gm107_disp, device, index, pdisp); + return nv50_disp_new_(&gm107_disp, device, index, pdisp); } diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/gm200.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/gm200.c index 292d3b5f9704..ec9c33a5162d 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/gm200.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/gm200.c @@ -28,18 +28,20 @@ static const struct nv50_disp_func gm200_disp = { + .init = gf119_disp_init, + .fini = gf119_disp_fini, .intr = gf119_disp_intr, .intr_error = gf119_disp_intr_error, .uevent = &gf119_disp_chan_uevent, .super = gf119_disp_super, .root = &gm200_disp_root_oclass, - .head.new = gf119_head_new, - .dac = { .nr = 3, .new = gf119_dac_new }, - .sor = { .nr = 4, .new = gm200_sor_new }, + .head = { .cnt = gf119_head_cnt, .new = gf119_head_new }, + .dac = { .cnt = gf119_dac_cnt, .new = gf119_dac_new }, + .sor = { .cnt = gf119_sor_cnt, .new = gm200_sor_new }, }; int gm200_disp_new(struct nvkm_device *device, int index, struct nvkm_disp **pdisp) { - return gf119_disp_new_(&gm200_disp, device, index, pdisp); + return nv50_disp_new_(&gm200_disp, device, index, pdisp); } diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/gp100.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/gp100.c index 39eb98b2c3a2..fd6216684f6d 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/gp100.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/gp100.c @@ -28,17 +28,19 @@ static const struct nv50_disp_func gp100_disp = { + .init = gf119_disp_init, + .fini = gf119_disp_fini, .intr = gf119_disp_intr, .intr_error = gf119_disp_intr_error, .uevent = &gf119_disp_chan_uevent, .super = gf119_disp_super, .root = &gp100_disp_root_oclass, - .head.new = gf119_head_new, - .sor = { .nr = 4, .new = gm200_sor_new }, + .head = { .cnt = gf119_head_cnt, .new = gf119_head_new }, + .sor = { .cnt = gf119_sor_cnt, .new = gm200_sor_new }, }; int gp100_disp_new(struct nvkm_device *device, int index, struct nvkm_disp **pdisp) { - return gf119_disp_new_(&gp100_disp, device, index, pdisp); + return nv50_disp_new_(&gp100_disp, device, index, pdisp); } diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/gp102.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/gp102.c index 91d70fe18275..3468ddec1270 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/gp102.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/gp102.c @@ -24,6 +24,7 @@ #include "nv50.h" #include "head.h" #include "ior.h" +#include "channv50.h" #include "rootnv50.h" static void @@ -54,17 +55,19 @@ gp102_disp_intr_error(struct nv50_disp *disp, int chid) static const struct nv50_disp_func gp102_disp = { + .init = gf119_disp_init, + .fini = gf119_disp_fini, .intr = gf119_disp_intr, .intr_error = gp102_disp_intr_error, .uevent = &gf119_disp_chan_uevent, .super = gf119_disp_super, .root = &gp102_disp_root_oclass, - .head.new = gf119_head_new, - .sor = { .nr = 4, .new = gm200_sor_new }, + .head = { .cnt = gf119_head_cnt, .new = gf119_head_new }, + .sor = { .cnt = gf119_sor_cnt, .new = gm200_sor_new }, }; int gp102_disp_new(struct nvkm_device *device, int index, struct nvkm_disp **pdisp) { - return gf119_disp_new_(&gp102_disp, device, index, pdisp); + return nv50_disp_new_(&gp102_disp, device, index, pdisp); } diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/gt200.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/gt200.c index bf00c4e3be3a..f80183701f44 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/gt200.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/gt200.c @@ -28,18 +28,20 @@ static const struct nv50_disp_func gt200_disp = { + .init = nv50_disp_init, + .fini = nv50_disp_fini, .intr = nv50_disp_intr, .uevent = &nv50_disp_chan_uevent, .super = nv50_disp_super, .root = >200_disp_root_oclass, - .head.new = nv50_head_new, - .dac = { .nr = 3, .new = nv50_dac_new }, - .sor = { .nr = 2, .new = g84_sor_new }, - .pior = { .nr = 3, .new = nv50_pior_new }, + .head = { .cnt = nv50_head_cnt, .new = nv50_head_new }, + .dac = { .cnt = nv50_dac_cnt, .new = nv50_dac_new }, + .sor = { .cnt = nv50_sor_cnt, .new = g84_sor_new }, + .pior = { .cnt = nv50_pior_cnt, .new = nv50_pior_new }, }; int gt200_disp_new(struct nvkm_device *device, int index, struct nvkm_disp **pdisp) { - return nv50_disp_new_(>200_disp, device, index, 2, pdisp); + return nv50_disp_new_(>200_disp, device, index, pdisp); } diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/gt215.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/gt215.c index 2cdd4d7a98d3..7581efc1357e 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/gt215.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/gt215.c @@ -28,18 +28,20 @@ static const struct nv50_disp_func gt215_disp = { + .init = nv50_disp_init, + .fini = nv50_disp_fini, .intr = nv50_disp_intr, .uevent = &nv50_disp_chan_uevent, .super = nv50_disp_super, .root = >215_disp_root_oclass, - .head.new = nv50_head_new, - .dac = { .nr = 3, .new = nv50_dac_new }, - .sor = { .nr = 4, .new = gt215_sor_new }, - .pior = { .nr = 3, .new = nv50_pior_new }, + .head = { .cnt = nv50_head_cnt, .new = nv50_head_new }, + .dac = { .cnt = nv50_dac_cnt, .new = nv50_dac_new }, + .sor = { .cnt = g94_sor_cnt, .new = gt215_sor_new }, + .pior = { .cnt = nv50_pior_cnt, .new = nv50_pior_new }, }; int gt215_disp_new(struct nvkm_device *device, int index, struct nvkm_disp **pdisp) { - return nv50_disp_new_(>215_disp, device, index, 2, pdisp); + return nv50_disp_new_(>215_disp, device, index, pdisp); } diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/gv100.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/gv100.c new file mode 100644 index 000000000000..d0a7e3456da1 --- /dev/null +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/gv100.c @@ -0,0 +1,427 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "nv50.h" +#include "head.h" +#include "ior.h" +#include "channv50.h" +#include "rootnv50.h" + +#include <core/gpuobj.h> +#include <subdev/timer.h> + +static int +gv100_disp_wndw_cnt(struct nvkm_disp *disp, unsigned long *pmask) +{ + struct nvkm_device *device = disp->engine.subdev.device; + *pmask = nvkm_rd32(device, 0x610064); + return (nvkm_rd32(device, 0x610074) & 0x03f00000) >> 20; +} + +static void +gv100_disp_super(struct work_struct *work) +{ + struct nv50_disp *disp = + container_of(work, struct nv50_disp, supervisor); + struct nvkm_subdev *subdev = &disp->base.engine.subdev; + struct nvkm_device *device = subdev->device; + struct nvkm_head *head; + u32 stat = nvkm_rd32(device, 0x6107a8); + u32 mask[4]; + + nvkm_debug(subdev, "supervisor %d: %08x\n", ffs(disp->super), stat); + list_for_each_entry(head, &disp->base.head, head) { + mask[head->id] = nvkm_rd32(device, 0x6107ac + (head->id * 4)); + HEAD_DBG(head, "%08x", mask[head->id]); + } + + if (disp->super & 0x00000001) { + nv50_disp_chan_mthd(disp->chan[0], NV_DBG_DEBUG); + nv50_disp_super_1(disp); + list_for_each_entry(head, &disp->base.head, head) { + if (!(mask[head->id] & 0x00001000)) + continue; + nv50_disp_super_1_0(disp, head); + } + } else + if (disp->super & 0x00000002) { + list_for_each_entry(head, &disp->base.head, head) { + if (!(mask[head->id] & 0x00001000)) + continue; + nv50_disp_super_2_0(disp, head); + } + nvkm_outp_route(&disp->base); + list_for_each_entry(head, &disp->base.head, head) { + if (!(mask[head->id] & 0x00010000)) + continue; + nv50_disp_super_2_1(disp, head); + } + list_for_each_entry(head, &disp->base.head, head) { + if (!(mask[head->id] & 0x00001000)) + continue; + nv50_disp_super_2_2(disp, head); + } + } else + if (disp->super & 0x00000004) { + list_for_each_entry(head, &disp->base.head, head) { + if (!(mask[head->id] & 0x00001000)) + continue; + nv50_disp_super_3_0(disp, head); + } + } + + list_for_each_entry(head, &disp->base.head, head) + nvkm_wr32(device, 0x6107ac + (head->id * 4), 0x00000000); + nvkm_wr32(device, 0x6107a8, 0x80000000); +} + +static void +gv100_disp_exception(struct nv50_disp *disp, int chid) +{ + struct nvkm_subdev *subdev = &disp->base.engine.subdev; + struct nvkm_device *device = subdev->device; + u32 stat = nvkm_rd32(device, 0x611020 + (chid * 12)); + u32 type = (stat & 0x00007000) >> 12; + u32 mthd = (stat & 0x00000fff) << 2; + u32 data = nvkm_rd32(device, 0x611024 + (chid * 12)); + u32 code = nvkm_rd32(device, 0x611028 + (chid * 12)); + + nvkm_error(subdev, "chid %d %08x [type %d mthd %04x] " + "data %08x code %08x\n", + chid, stat, type, mthd, data, code); + + if (chid < ARRAY_SIZE(disp->chan) && disp->chan[chid]) { + switch (mthd) { + case 0x0200: + nv50_disp_chan_mthd(disp->chan[chid], NV_DBG_ERROR); + break; + default: + break; + } + } + + nvkm_wr32(device, 0x611020 + (chid * 12), 0x90000000); +} + +static void +gv100_disp_intr_ctrl_disp(struct nv50_disp *disp) +{ + struct nvkm_subdev *subdev = &disp->base.engine.subdev; + struct nvkm_device *device = subdev->device; + u32 stat = nvkm_rd32(device, 0x611c30); + + if (stat & 0x00000007) { + disp->super = (stat & 0x00000007); + queue_work(disp->wq, &disp->supervisor); + nvkm_wr32(device, 0x611860, disp->super); + stat &= ~0x00000007; + } + + /*TODO: I would guess this is VBIOS_RELEASE, however, NFI how to + * ACK it, nor does RM appear to bother. + */ + if (stat & 0x00000008) + stat &= ~0x00000008; + + if (stat & 0x00000100) { + unsigned long wndws = nvkm_rd32(device, 0x611858); + unsigned long other = nvkm_rd32(device, 0x61185c); + int wndw; + + nvkm_wr32(device, 0x611858, wndws); + nvkm_wr32(device, 0x61185c, other); + + /* AWAKEN_OTHER_CORE. */ + if (other & 0x00000001) + nv50_disp_chan_uevent_send(disp, 0); + + /* AWAKEN_WIN_CH(n). */ + for_each_set_bit(wndw, &wndws, disp->wndw.nr) { + nv50_disp_chan_uevent_send(disp, 1 + wndw); + } + } + + if (stat) + nvkm_warn(subdev, "ctrl %08x\n", stat); +} + +static void +gv100_disp_intr_exc_other(struct nv50_disp *disp) +{ + struct nvkm_subdev *subdev = &disp->base.engine.subdev; + struct nvkm_device *device = subdev->device; + u32 stat = nvkm_rd32(device, 0x611854); + unsigned long mask; + int head; + + if (stat & 0x00000001) { + nvkm_wr32(device, 0x611854, 0x00000001); + gv100_disp_exception(disp, 0); + stat &= ~0x00000001; + } + + if ((mask = (stat & 0x00ff0000) >> 16)) { + for_each_set_bit(head, &mask, disp->wndw.nr) { + nvkm_wr32(device, 0x611854, 0x00010000 << head); + gv100_disp_exception(disp, 73 + head); + stat &= ~(0x00010000 << head); + } + } + + if (stat) { + nvkm_warn(subdev, "exception %08x\n", stat); + nvkm_wr32(device, 0x611854, stat); + } +} + +static void +gv100_disp_intr_exc_winim(struct nv50_disp *disp) +{ + struct nvkm_subdev *subdev = &disp->base.engine.subdev; + struct nvkm_device *device = subdev->device; + unsigned long stat = nvkm_rd32(device, 0x611850); + int wndw; + + for_each_set_bit(wndw, &stat, disp->wndw.nr) { + nvkm_wr32(device, 0x611850, BIT(wndw)); + gv100_disp_exception(disp, 33 + wndw); + stat &= ~BIT(wndw); + } + + if (stat) { + nvkm_warn(subdev, "wimm %08x\n", (u32)stat); + nvkm_wr32(device, 0x611850, stat); + } +} + +static void +gv100_disp_intr_exc_win(struct nv50_disp *disp) +{ + struct nvkm_subdev *subdev = &disp->base.engine.subdev; + struct nvkm_device *device = subdev->device; + unsigned long stat = nvkm_rd32(device, 0x61184c); + int wndw; + + for_each_set_bit(wndw, &stat, disp->wndw.nr) { + nvkm_wr32(device, 0x61184c, BIT(wndw)); + gv100_disp_exception(disp, 1 + wndw); + stat &= ~BIT(wndw); + } + + if (stat) { + nvkm_warn(subdev, "wndw %08x\n", (u32)stat); + nvkm_wr32(device, 0x61184c, stat); + } +} + +static void +gv100_disp_intr_head_timing(struct nv50_disp *disp, int head) +{ + struct nvkm_subdev *subdev = &disp->base.engine.subdev; + struct nvkm_device *device = subdev->device; + u32 stat = nvkm_rd32(device, 0x611800 + (head * 0x04)); + + /* LAST_DATA, LOADV. */ + if (stat & 0x00000003) { + nvkm_wr32(device, 0x611800 + (head * 0x04), stat & 0x00000003); + stat &= ~0x00000003; + } + + if (stat & 0x00000004) { + nvkm_disp_vblank(&disp->base, head); + nvkm_wr32(device, 0x611800 + (head * 0x04), 0x00000004); + stat &= ~0x00000004; + } + + if (stat) { + nvkm_warn(subdev, "head %08x\n", stat); + nvkm_wr32(device, 0x611800 + (head * 0x04), stat); + } +} + +static void +gv100_disp_intr(struct nv50_disp *disp) +{ + struct nvkm_subdev *subdev = &disp->base.engine.subdev; + struct nvkm_device *device = subdev->device; + u32 stat = nvkm_rd32(device, 0x611ec0); + unsigned long mask; + int head; + + if ((mask = (stat & 0x000000ff))) { + for_each_set_bit(head, &mask, 8) { + gv100_disp_intr_head_timing(disp, head); + stat &= ~BIT(head); + } + } + + if (stat & 0x00000200) { + gv100_disp_intr_exc_win(disp); + stat &= ~0x00000200; + } + + if (stat & 0x00000400) { + gv100_disp_intr_exc_winim(disp); + stat &= ~0x00000400; + } + + if (stat & 0x00000800) { + gv100_disp_intr_exc_other(disp); + stat &= ~0x00000800; + } + + if (stat & 0x00001000) { + gv100_disp_intr_ctrl_disp(disp); + stat &= ~0x00001000; + } + + if (stat) + nvkm_warn(subdev, "intr %08x\n", stat); +} + +static void +gv100_disp_fini(struct nv50_disp *disp) +{ + struct nvkm_device *device = disp->base.engine.subdev.device; + nvkm_wr32(device, 0x611db0, 0x00000000); +} + +static int +gv100_disp_init(struct nv50_disp *disp) +{ + struct nvkm_device *device = disp->base.engine.subdev.device; + struct nvkm_head *head; + int i, j; + u32 tmp; + + /* Claim ownership of display. */ + if (nvkm_rd32(device, 0x6254e8) & 0x00000002) { + nvkm_mask(device, 0x6254e8, 0x00000001, 0x00000000); + if (nvkm_msec(device, 2000, + if (!(nvkm_rd32(device, 0x6254e8) & 0x00000002)) + break; + ) < 0) + return -EBUSY; + } + + /* Lock pin capabilities. */ + tmp = nvkm_rd32(device, 0x610068); + nvkm_wr32(device, 0x640008, tmp); + + /* SOR capabilities. */ + for (i = 0; i < disp->sor.nr; i++) { + tmp = nvkm_rd32(device, 0x61c000 + (i * 0x800)); + nvkm_mask(device, 0x640000, 0x00000100 << i, 0x00000100 << i); + nvkm_wr32(device, 0x640144 + (i * 0x08), tmp); + } + + /* Head capabilities. */ + list_for_each_entry(head, &disp->base.head, head) { + const int id = head->id; + + /* RG. */ + tmp = nvkm_rd32(device, 0x616300 + (id * 0x800)); + nvkm_wr32(device, 0x640048 + (id * 0x020), tmp); + + /* POSTCOMP. */ + for (j = 0; j < 6 * 4; j += 4) { + tmp = nvkm_rd32(device, 0x616100 + (id * 0x800) + j); + nvkm_wr32(device, 0x640030 + (id * 0x20) + j, tmp); + } + } + + /* Window capabilities. */ + for (i = 0; i < disp->wndw.nr; i++) { + nvkm_mask(device, 0x640004, 1 << i, 1 << i); + for (j = 0; j < 6 * 4; j += 4) { + tmp = nvkm_rd32(device, 0x630050 + (i * 0x800) + j); + nvkm_wr32(device, 0x6401e4 + (i * 0x20) + j, tmp); + } + } + + /* IHUB capabilities. */ + for (i = 0; i < 4; i++) { + tmp = nvkm_rd32(device, 0x62e000 + (i * 0x04)); + nvkm_wr32(device, 0x640010 + (i * 0x04), tmp); + } + + nvkm_mask(device, 0x610078, 0x00000001, 0x00000001); + + /* Setup instance memory. */ + switch (nvkm_memory_target(disp->inst->memory)) { + case NVKM_MEM_TARGET_VRAM: tmp = 0x00000001; break; + case NVKM_MEM_TARGET_NCOH: tmp = 0x00000002; break; + case NVKM_MEM_TARGET_HOST: tmp = 0x00000003; break; + default: + break; + } + nvkm_wr32(device, 0x610010, 0x00000008 | tmp); + nvkm_wr32(device, 0x610014, disp->inst->addr >> 16); + + /* CTRL_DISP: AWAKEN, ERROR, SUPERVISOR[1-3]. */ + nvkm_wr32(device, 0x611cf0, 0x00000187); /* MSK. */ + nvkm_wr32(device, 0x611db0, 0x00000187); /* EN. */ + + /* EXC_OTHER: CURSn, CORE. */ + nvkm_wr32(device, 0x611cec, disp->head.mask << 16 | + 0x00000001); /* MSK. */ + nvkm_wr32(device, 0x611dac, 0x00000000); /* EN. */ + + /* EXC_WINIM. */ + nvkm_wr32(device, 0x611ce8, disp->wndw.mask); /* MSK. */ + nvkm_wr32(device, 0x611da8, 0x00000000); /* EN. */ + + /* EXC_WIN. */ + nvkm_wr32(device, 0x611ce4, disp->wndw.mask); /* MSK. */ + nvkm_wr32(device, 0x611da4, 0x00000000); /* EN. */ + + /* HEAD_TIMING(n): VBLANK. */ + list_for_each_entry(head, &disp->base.head, head) { + const u32 hoff = head->id * 4; + nvkm_wr32(device, 0x611cc0 + hoff, 0x00000004); /* MSK. */ + nvkm_wr32(device, 0x611d80 + hoff, 0x00000000); /* EN. */ + } + + /* OR. */ + nvkm_wr32(device, 0x611cf4, 0x00000000); /* MSK. */ + nvkm_wr32(device, 0x611db4, 0x00000000); /* EN. */ + return 0; +} + +static const struct nv50_disp_func +gv100_disp = { + .init = gv100_disp_init, + .fini = gv100_disp_fini, + .intr = gv100_disp_intr, + .uevent = &gv100_disp_chan_uevent, + .super = gv100_disp_super, + .root = &gv100_disp_root_oclass, + .wndw = { .cnt = gv100_disp_wndw_cnt }, + .head = { .cnt = gv100_head_cnt, .new = gv100_head_new }, + .sor = { .cnt = gv100_sor_cnt, .new = gv100_sor_new }, + .ramht_size = 0x2000, +}; + +int +gv100_disp_new(struct nvkm_device *device, int index, struct nvkm_disp **pdisp) +{ + return nv50_disp_new_(&gv100_disp, device, index, pdisp); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/hdmigv100.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/hdmigv100.c new file mode 100644 index 000000000000..6e3c450eaace --- /dev/null +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/hdmigv100.c @@ -0,0 +1,85 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "hdmi.h" + +void +gv100_hdmi_ctrl(struct nvkm_ior *ior, int head, bool enable, u8 max_ac_packet, + u8 rekey, u8 *avi, u8 avi_size, u8 *vendor, u8 vendor_size) +{ + struct nvkm_device *device = ior->disp->engine.subdev.device; + const u32 ctrl = 0x40000000 * enable | + max_ac_packet << 16 | + rekey; + const u32 hoff = head * 0x800; + const u32 hdmi = head * 0x400; + struct packed_hdmi_infoframe avi_infoframe; + struct packed_hdmi_infoframe vendor_infoframe; + + pack_hdmi_infoframe(&avi_infoframe, avi, avi_size); + pack_hdmi_infoframe(&vendor_infoframe, vendor, vendor_size); + + if (!(ctrl & 0x40000000)) { + nvkm_mask(device, 0x6165c0 + hoff, 0x40000000, 0x00000000); + nvkm_mask(device, 0x6f0100 + hdmi, 0x00000001, 0x00000000); + nvkm_mask(device, 0x6f00c0 + hdmi, 0x00000001, 0x00000000); + nvkm_mask(device, 0x6f0000 + hdmi, 0x00000001, 0x00000000); + return; + } + + /* AVI InfoFrame (AVI). */ + nvkm_mask(device, 0x6f0000 + hdmi, 0x00000001, 0x00000000); + if (avi_size) { + nvkm_wr32(device, 0x6f0008 + hdmi, avi_infoframe.header); + nvkm_wr32(device, 0x6f000c + hdmi, avi_infoframe.subpack0_low); + nvkm_wr32(device, 0x6f0010 + hdmi, avi_infoframe.subpack0_high); + nvkm_wr32(device, 0x6f0014 + hdmi, avi_infoframe.subpack1_low); + nvkm_wr32(device, 0x6f0018 + hdmi, avi_infoframe.subpack1_high); + nvkm_mask(device, 0x6f0000 + hdmi, 0x00000001, 0x00000001); + } + + /* Vendor-specific InfoFrame (VSI). */ + nvkm_mask(device, 0x6f0100 + hdmi, 0x00010001, 0x00000000); + if (vendor_size) { + nvkm_wr32(device, 0x6f0108 + hdmi, vendor_infoframe.header); + nvkm_wr32(device, 0x6f010c + hdmi, vendor_infoframe.subpack0_low); + nvkm_wr32(device, 0x6f0110 + hdmi, vendor_infoframe.subpack0_high); + nvkm_wr32(device, 0x6f0110 + hdmi, 0x00000000); + nvkm_wr32(device, 0x6f0114 + hdmi, 0x00000000); + nvkm_wr32(device, 0x6f0118 + hdmi, 0x00000000); + nvkm_wr32(device, 0x6f011c + hdmi, 0x00000000); + nvkm_wr32(device, 0x6f0120 + hdmi, 0x00000000); + nvkm_wr32(device, 0x6f0124 + hdmi, 0x00000000); + nvkm_mask(device, 0x6f0100 + hdmi, 0x00000001, 0x00000001); + } + + + /* General Control (GCP). */ + nvkm_mask(device, 0x6f00c0 + hdmi, 0x00000001, 0x00000000); + nvkm_wr32(device, 0x6f00cc + hdmi, 0x00000010); + nvkm_mask(device, 0x6f00c0 + hdmi, 0x00000001, 0x00000001); + + /* Audio Clock Regeneration (ACR). */ + nvkm_wr32(device, 0x6f0080 + hdmi, 0x82000000); + + /* NV_PDISP_SF_HDMI_CTRL. */ + nvkm_mask(device, 0x6165c0 + hoff, 0x401f007f, ctrl); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/head.h b/drivers/gpu/drm/nouveau/nvkm/engine/disp/head.h index 57030b3a4a75..7d55faf52fcb 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/head.h +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/head.h @@ -52,6 +52,14 @@ void nv50_head_rgpos(struct nvkm_head *, u16 *, u16 *); #define HEAD_DBG(h,f,a...) HEAD_MSG((h), debug, f, ##a) int nv04_head_new(struct nvkm_disp *, int id); + +int nv50_head_cnt(struct nvkm_disp *, unsigned long *); int nv50_head_new(struct nvkm_disp *, int id); + +int gf119_head_cnt(struct nvkm_disp *, unsigned long *); int gf119_head_new(struct nvkm_disp *, int id); +void gf119_head_rgclk(struct nvkm_head *, int); + +int gv100_head_cnt(struct nvkm_disp *, unsigned long *); +int gv100_head_new(struct nvkm_disp *, int id); #endif diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/headgf119.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/headgf119.c index 9fd7ae331308..e86298b35902 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/headgf119.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/headgf119.c @@ -39,7 +39,7 @@ gf119_head_vblank_get(struct nvkm_head *head) nvkm_mask(device, 0x6100c0 + hoff, 0x00000001, 0x00000001); } -static void +void gf119_head_rgclk(struct nvkm_head *head, int div) { struct nvkm_device *device = head->disp->engine.subdev.device; @@ -92,8 +92,13 @@ gf119_head = { int gf119_head_new(struct nvkm_disp *disp, int id) { - struct nvkm_device *device = disp->engine.subdev.device; - if (!(nvkm_rd32(device, 0x612004) & (0x00000001 << id))) - return 0; return nvkm_head_new_(&gf119_head, disp, id); } + +int +gf119_head_cnt(struct nvkm_disp *disp, unsigned long *pmask) +{ + struct nvkm_device *device = disp->engine.subdev.device; + *pmask = nvkm_rd32(device, 0x612004) & 0x0000000f; + return nvkm_rd32(device, 0x022448); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/headgv100.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/headgv100.c new file mode 100644 index 000000000000..1a061b42ae5c --- /dev/null +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/headgv100.c @@ -0,0 +1,105 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "head.h" + +static void +gv100_head_vblank_put(struct nvkm_head *head) +{ + struct nvkm_device *device = head->disp->engine.subdev.device; + nvkm_mask(device, 0x611d80 + (head->id * 4), 0x00000004, 0x00000000); +} + +static void +gv100_head_vblank_get(struct nvkm_head *head) +{ + struct nvkm_device *device = head->disp->engine.subdev.device; + nvkm_mask(device, 0x611d80 + (head->id * 4), 0x00000004, 0x00000004); +} + +static void +gv100_head_rgpos(struct nvkm_head *head, u16 *hline, u16 *vline) +{ + struct nvkm_device *device = head->disp->engine.subdev.device; + const u32 hoff = head->id * 0x800; + /* vline read locks hline. */ + *vline = nvkm_rd32(device, 0x616330 + hoff) & 0x0000ffff; + *hline = nvkm_rd32(device, 0x616334 + hoff) & 0x0000ffff; +} + +static void +gv100_head_state(struct nvkm_head *head, struct nvkm_head_state *state) +{ + struct nvkm_device *device = head->disp->engine.subdev.device; + const u32 hoff = (state == &head->arm) * 0x8000 + head->id * 0x400; + u32 data; + + data = nvkm_rd32(device, 0x682064 + hoff); + state->vtotal = (data & 0xffff0000) >> 16; + state->htotal = (data & 0x0000ffff); + data = nvkm_rd32(device, 0x682068 + hoff); + state->vsynce = (data & 0xffff0000) >> 16; + state->hsynce = (data & 0x0000ffff); + data = nvkm_rd32(device, 0x68206c + hoff); + state->vblanke = (data & 0xffff0000) >> 16; + state->hblanke = (data & 0x0000ffff); + data = nvkm_rd32(device, 0x682070 + hoff); + state->vblanks = (data & 0xffff0000) >> 16; + state->hblanks = (data & 0x0000ffff); + state->hz = nvkm_rd32(device, 0x68200c + hoff); + + data = nvkm_rd32(device, 0x682004 + hoff); + switch ((data & 0x000000f0) >> 4) { + case 5: state->or.depth = 30; break; + case 4: state->or.depth = 24; break; + case 1: state->or.depth = 18; break; + default: + state->or.depth = 18; + WARN_ON(1); + break; + } +} + +static const struct nvkm_head_func +gv100_head = { + .state = gv100_head_state, + .rgpos = gv100_head_rgpos, + .rgclk = gf119_head_rgclk, + .vblank_get = gv100_head_vblank_get, + .vblank_put = gv100_head_vblank_put, +}; + +int +gv100_head_new(struct nvkm_disp *disp, int id) +{ + struct nvkm_device *device = disp->engine.subdev.device; + if (!(nvkm_rd32(device, 0x610060) & (0x00000001 << id))) + return 0; + return nvkm_head_new_(&gv100_head, disp, id); +} + +int +gv100_head_cnt(struct nvkm_disp *disp, unsigned long *pmask) +{ + struct nvkm_device *device = disp->engine.subdev.device; + *pmask = nvkm_rd32(device, 0x610060) & 0x000000ff; + return nvkm_rd32(device, 0x610074) & 0x0000000f; +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/headnv50.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/headnv50.c index c80d06d5168f..e7d5c397cd29 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/headnv50.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/headnv50.c @@ -90,3 +90,10 @@ nv50_head_new(struct nvkm_disp *disp, int id) { return nvkm_head_new_(&nv50_head, disp, id); } + +int +nv50_head_cnt(struct nvkm_disp *disp, unsigned long *pmask) +{ + *pmask = 3; + return 2; +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/ior.h b/drivers/gpu/drm/nouveau/nvkm/engine/disp/ior.h index 4548c031b937..e0b4e0c5704e 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/ior.h +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/ior.h @@ -30,7 +30,7 @@ struct nvkm_ior { UNKNOWN } proto:3; unsigned link:2; - unsigned head:4; + unsigned head:8; } arm, asy; /* Armed DP state. */ @@ -106,7 +106,6 @@ nv50_sor_link(struct nvkm_ior *ior) return nv50_ior_base(ior) + ((ior->asy.link == 2) * 0x80); } -int nv50_sor_new_(const struct nvkm_ior_func *, struct nvkm_disp *, int id); void nv50_sor_state(struct nvkm_ior *, struct nvkm_ior_state *); void nv50_sor_power(struct nvkm_ior *, bool, bool, bool, bool, bool); void nv50_sor_clock(struct nvkm_ior *); @@ -122,7 +121,6 @@ void g94_sor_dp_watermark(struct nvkm_ior *, int, u8); void gt215_sor_dp_audio(struct nvkm_ior *, int, bool); -int gf119_sor_new_(const struct nvkm_ior_func *, struct nvkm_disp *, int id); void gf119_sor_state(struct nvkm_ior *, struct nvkm_ior_state *); void gf119_sor_clock(struct nvkm_ior *); int gf119_sor_dp_links(struct nvkm_ior *, struct nvkm_i2c_aux *); @@ -135,10 +133,15 @@ void gf119_sor_dp_watermark(struct nvkm_ior *, int, u8); void gm107_sor_dp_pattern(struct nvkm_ior *, int); +void gm200_sor_route_set(struct nvkm_outp *, struct nvkm_ior *); +int gm200_sor_route_get(struct nvkm_outp *, int *); +void gm200_sor_dp_drive(struct nvkm_ior *, int, int, int, int, int); + void g84_hdmi_ctrl(struct nvkm_ior *, int, bool, u8, u8, u8 *, u8 , u8 *, u8); void gt215_hdmi_ctrl(struct nvkm_ior *, int, bool, u8, u8, u8 *, u8 , u8 *, u8); void gf119_hdmi_ctrl(struct nvkm_ior *, int, bool, u8, u8, u8 *, u8 , u8 *, u8); void gk104_hdmi_ctrl(struct nvkm_ior *, int, bool, u8, u8, u8 *, u8 , u8 *, u8); +void gv100_hdmi_ctrl(struct nvkm_ior *, int, bool, u8, u8, u8 *, u8 , u8 *, u8); void gt215_hda_hpd(struct nvkm_ior *, int, bool); void gt215_hda_eld(struct nvkm_ior *, u8 *, u8); @@ -153,19 +156,34 @@ void gf119_hda_eld(struct nvkm_ior *, u8 *, u8); #define IOR_WARN(i,f,a...) IOR_MSG((i), warn, f, ##a) #define IOR_DBG(i,f,a...) IOR_MSG((i), debug, f, ##a) +int nv50_dac_cnt(struct nvkm_disp *, unsigned long *); int nv50_dac_new(struct nvkm_disp *, int); + +int gf119_dac_cnt(struct nvkm_disp *, unsigned long *); int gf119_dac_new(struct nvkm_disp *, int); +int nv50_pior_cnt(struct nvkm_disp *, unsigned long *); int nv50_pior_new(struct nvkm_disp *, int); +int nv50_sor_cnt(struct nvkm_disp *, unsigned long *); int nv50_sor_new(struct nvkm_disp *, int); + int g84_sor_new(struct nvkm_disp *, int); + +int g94_sor_cnt(struct nvkm_disp *, unsigned long *); int g94_sor_new(struct nvkm_disp *, int); + int mcp77_sor_new(struct nvkm_disp *, int); int gt215_sor_new(struct nvkm_disp *, int); int mcp89_sor_new(struct nvkm_disp *, int); + +int gf119_sor_cnt(struct nvkm_disp *, unsigned long *); int gf119_sor_new(struct nvkm_disp *, int); + int gk104_sor_new(struct nvkm_disp *, int); int gm107_sor_new(struct nvkm_disp *, int); int gm200_sor_new(struct nvkm_disp *, int); + +int gv100_sor_cnt(struct nvkm_disp *, unsigned long *); +int gv100_sor_new(struct nvkm_disp *, int); #endif diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/mcp77.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/mcp77.c index d7e0fbb12bf1..cfdce23ab83a 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/mcp77.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/mcp77.c @@ -26,18 +26,20 @@ static const struct nv50_disp_func mcp77_disp = { + .init = nv50_disp_init, + .fini = nv50_disp_fini, .intr = nv50_disp_intr, .uevent = &nv50_disp_chan_uevent, .super = nv50_disp_super, .root = &g94_disp_root_oclass, - .head.new = nv50_head_new, - .dac = { .nr = 3, .new = nv50_dac_new }, - .sor = { .nr = 4, .new = mcp77_sor_new }, - .pior = { .nr = 3, .new = nv50_pior_new }, + .head = { .cnt = nv50_head_cnt, .new = nv50_head_new }, + .dac = { .cnt = nv50_dac_cnt, .new = nv50_dac_new }, + .sor = { .cnt = g94_sor_cnt, .new = mcp77_sor_new }, + .pior = { .cnt = nv50_pior_cnt, .new = nv50_pior_new }, }; int mcp77_disp_new(struct nvkm_device *device, int index, struct nvkm_disp **pdisp) { - return nv50_disp_new_(&mcp77_disp, device, index, 2, pdisp); + return nv50_disp_new_(&mcp77_disp, device, index, pdisp); } diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/mcp89.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/mcp89.c index 7b75c57c12ed..85d9329cfa0e 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/mcp89.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/mcp89.c @@ -26,18 +26,20 @@ static const struct nv50_disp_func mcp89_disp = { + .init = nv50_disp_init, + .fini = nv50_disp_fini, .intr = nv50_disp_intr, .uevent = &nv50_disp_chan_uevent, .super = nv50_disp_super, .root = >215_disp_root_oclass, - .head.new = nv50_head_new, - .dac = { .nr = 3, .new = nv50_dac_new }, - .sor = { .nr = 4, .new = mcp89_sor_new }, - .pior = { .nr = 3, .new = nv50_pior_new }, + .head = { .cnt = nv50_head_cnt, .new = nv50_head_new }, + .dac = { .cnt = nv50_dac_cnt, .new = nv50_dac_new }, + .sor = { .cnt = g94_sor_cnt, .new = mcp89_sor_new }, + .pior = { .cnt = nv50_pior_cnt, .new = nv50_pior_new }, }; int mcp89_disp_new(struct nvkm_device *device, int index, struct nvkm_disp **pdisp) { - return nv50_disp_new_(&mcp89_disp, device, index, 2, pdisp); + return nv50_disp_new_(&mcp89_disp, device, index, pdisp); } diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/nv50.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/nv50.c index 0c570dbd3021..f89c7b977aa5 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/nv50.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/nv50.c @@ -24,11 +24,12 @@ #include "nv50.h" #include "head.h" #include "ior.h" +#include "channv50.h" #include "rootnv50.h" #include <core/client.h> #include <core/enum.h> -#include <core/gpuobj.h> +#include <core/ramht.h> #include <subdev/bios.h> #include <subdev/bios/disp.h> #include <subdev/bios/init.h> @@ -49,29 +50,115 @@ nv50_disp_intr_(struct nvkm_disp *base) disp->func->intr(disp); } +static void +nv50_disp_fini_(struct nvkm_disp *base) +{ + struct nv50_disp *disp = nv50_disp(base); + disp->func->fini(disp); +} + +static int +nv50_disp_init_(struct nvkm_disp *base) +{ + struct nv50_disp *disp = nv50_disp(base); + return disp->func->init(disp); +} + static void * nv50_disp_dtor_(struct nvkm_disp *base) { struct nv50_disp *disp = nv50_disp(base); + + nvkm_ramht_del(&disp->ramht); + nvkm_gpuobj_del(&disp->inst); + nvkm_event_fini(&disp->uevent); if (disp->wq) destroy_workqueue(disp->wq); + return disp; } +static int +nv50_disp_oneinit_(struct nvkm_disp *base) +{ + struct nv50_disp *disp = nv50_disp(base); + const struct nv50_disp_func *func = disp->func; + struct nvkm_subdev *subdev = &disp->base.engine.subdev; + struct nvkm_device *device = subdev->device; + int ret, i; + + if (func->wndw.cnt) { + disp->wndw.nr = func->wndw.cnt(&disp->base, &disp->wndw.mask); + nvkm_debug(subdev, "Window(s): %d (%08lx)\n", + disp->wndw.nr, disp->wndw.mask); + } + + disp->head.nr = func->head.cnt(&disp->base, &disp->head.mask); + nvkm_debug(subdev, " Head(s): %d (%02lx)\n", + disp->head.nr, disp->head.mask); + for_each_set_bit(i, &disp->head.mask, disp->head.nr) { + ret = func->head.new(&disp->base, i); + if (ret) + return ret; + } + + if (func->dac.cnt) { + disp->dac.nr = func->dac.cnt(&disp->base, &disp->dac.mask); + nvkm_debug(subdev, " DAC(s): %d (%02lx)\n", + disp->dac.nr, disp->dac.mask); + for_each_set_bit(i, &disp->dac.mask, disp->dac.nr) { + ret = func->dac.new(&disp->base, i); + if (ret) + return ret; + } + } + + if (func->pior.cnt) { + disp->pior.nr = func->pior.cnt(&disp->base, &disp->pior.mask); + nvkm_debug(subdev, " PIOR(s): %d (%02lx)\n", + disp->pior.nr, disp->pior.mask); + for_each_set_bit(i, &disp->pior.mask, disp->pior.nr) { + ret = func->pior.new(&disp->base, i); + if (ret) + return ret; + } + } + + disp->sor.nr = func->sor.cnt(&disp->base, &disp->sor.mask); + nvkm_debug(subdev, " SOR(s): %d (%02lx)\n", + disp->sor.nr, disp->sor.mask); + for_each_set_bit(i, &disp->sor.mask, disp->sor.nr) { + ret = func->sor.new(&disp->base, i); + if (ret) + return ret; + } + + ret = nvkm_gpuobj_new(device, 0x10000, 0x10000, false, NULL, + &disp->inst); + if (ret) + return ret; + + return nvkm_ramht_new(device, func->ramht_size ? func->ramht_size : + 0x1000, 0, disp->inst, &disp->ramht); +} + static const struct nvkm_disp_func nv50_disp_ = { .dtor = nv50_disp_dtor_, + .oneinit = nv50_disp_oneinit_, + .init = nv50_disp_init_, + .fini = nv50_disp_fini_, .intr = nv50_disp_intr_, .root = nv50_disp_root_, }; int nv50_disp_new_(const struct nv50_disp_func *func, struct nvkm_device *device, - int index, int heads, struct nvkm_disp **pdisp) + int index, struct nvkm_disp **pdisp) { struct nv50_disp *disp; - int ret, i; + int ret; if (!(disp = kzalloc(sizeof(*disp), GFP_KERNEL))) return -ENOMEM; @@ -85,33 +172,11 @@ nv50_disp_new_(const struct nv50_disp_func *func, struct nvkm_device *device, disp->wq = create_singlethread_workqueue("nvkm-disp"); if (!disp->wq) return -ENOMEM; - INIT_WORK(&disp->supervisor, func->super); - - for (i = 0; func->head.new && i < heads; i++) { - ret = func->head.new(&disp->base, i); - if (ret) - return ret; - } - - for (i = 0; func->dac.new && i < func->dac.nr; i++) { - ret = func->dac.new(&disp->base, i); - if (ret) - return ret; - } - - for (i = 0; func->pior.new && i < func->pior.nr; i++) { - ret = func->pior.new(&disp->base, i); - if (ret) - return ret; - } - for (i = 0; func->sor.new && i < func->sor.nr; i++) { - ret = func->sor.new(&disp->base, i); - if (ret) - return ret; - } + INIT_WORK(&disp->supervisor, func->super); - return nvkm_event_init(func->uevent, 1, 1 + (heads * 4), &disp->uevent); + return nvkm_event_init(func->uevent, 1, ARRAY_SIZE(disp->chan), + &disp->uevent); } static u32 @@ -613,20 +678,96 @@ nv50_disp_intr(struct nv50_disp *disp) } } +void +nv50_disp_fini(struct nv50_disp *disp) +{ + struct nvkm_device *device = disp->base.engine.subdev.device; + /* disable all interrupts */ + nvkm_wr32(device, 0x610024, 0x00000000); + nvkm_wr32(device, 0x610020, 0x00000000); +} + +int +nv50_disp_init(struct nv50_disp *disp) +{ + struct nvkm_device *device = disp->base.engine.subdev.device; + struct nvkm_head *head; + u32 tmp; + int i; + + /* The below segments of code copying values from one register to + * another appear to inform EVO of the display capabilities or + * something similar. NFI what the 0x614004 caps are for.. + */ + tmp = nvkm_rd32(device, 0x614004); + nvkm_wr32(device, 0x610184, tmp); + + /* ... CRTC caps */ + list_for_each_entry(head, &disp->base.head, head) { + tmp = nvkm_rd32(device, 0x616100 + (head->id * 0x800)); + nvkm_wr32(device, 0x610190 + (head->id * 0x10), tmp); + tmp = nvkm_rd32(device, 0x616104 + (head->id * 0x800)); + nvkm_wr32(device, 0x610194 + (head->id * 0x10), tmp); + tmp = nvkm_rd32(device, 0x616108 + (head->id * 0x800)); + nvkm_wr32(device, 0x610198 + (head->id * 0x10), tmp); + tmp = nvkm_rd32(device, 0x61610c + (head->id * 0x800)); + nvkm_wr32(device, 0x61019c + (head->id * 0x10), tmp); + } + + /* ... DAC caps */ + for (i = 0; i < disp->dac.nr; i++) { + tmp = nvkm_rd32(device, 0x61a000 + (i * 0x800)); + nvkm_wr32(device, 0x6101d0 + (i * 0x04), tmp); + } + + /* ... SOR caps */ + for (i = 0; i < disp->sor.nr; i++) { + tmp = nvkm_rd32(device, 0x61c000 + (i * 0x800)); + nvkm_wr32(device, 0x6101e0 + (i * 0x04), tmp); + } + + /* ... PIOR caps */ + for (i = 0; i < disp->pior.nr; i++) { + tmp = nvkm_rd32(device, 0x61e000 + (i * 0x800)); + nvkm_wr32(device, 0x6101f0 + (i * 0x04), tmp); + } + + /* steal display away from vbios, or something like that */ + if (nvkm_rd32(device, 0x610024) & 0x00000100) { + nvkm_wr32(device, 0x610024, 0x00000100); + nvkm_mask(device, 0x6194e8, 0x00000001, 0x00000000); + if (nvkm_msec(device, 2000, + if (!(nvkm_rd32(device, 0x6194e8) & 0x00000002)) + break; + ) < 0) + return -EBUSY; + } + + /* point at display engine memory area (hash table, objects) */ + nvkm_wr32(device, 0x610010, (disp->inst->addr >> 8) | 9); + + /* enable supervisor interrupts, disable everything else */ + nvkm_wr32(device, 0x61002c, 0x00000370); + nvkm_wr32(device, 0x610028, 0x00000000); + return 0; +} + static const struct nv50_disp_func nv50_disp = { + .init = nv50_disp_init, + .fini = nv50_disp_fini, .intr = nv50_disp_intr, .uevent = &nv50_disp_chan_uevent, .super = nv50_disp_super, .root = &nv50_disp_root_oclass, - .head.new = nv50_head_new, - .dac = { .nr = 3, .new = nv50_dac_new }, - .sor = { .nr = 2, .new = nv50_sor_new }, - .pior = { .nr = 3, .new = nv50_pior_new }, + .head = { .cnt = nv50_head_cnt, .new = nv50_head_new }, + .dac = { .cnt = nv50_dac_cnt, .new = nv50_dac_new }, + .sor = { .cnt = nv50_sor_cnt, .new = nv50_sor_new }, + .pior = { .cnt = nv50_pior_cnt, .new = nv50_pior_new }, }; int nv50_disp_new(struct nvkm_device *device, int index, struct nvkm_disp **pdisp) { - return nv50_disp_new_(&nv50_disp, device, index, 2, pdisp); + return nv50_disp_new_(&nv50_disp, device, index, pdisp); } diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/nv50.h b/drivers/gpu/drm/nouveau/nvkm/engine/disp/nv50.h index eb0b8acb1c5b..8580382ab248 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/nv50.h +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/nv50.h @@ -16,14 +16,26 @@ struct nv50_disp { struct nvkm_event uevent; struct { + unsigned long mask; + int nr; + } wndw, head, dac; + + struct { + unsigned long mask; + int nr; u32 lvdsconf; } sor; struct { + unsigned long mask; + int nr; u8 type[3]; } pior; - struct nv50_disp_chan *chan[21]; + struct nvkm_gpuobj *inst; + struct nvkm_ramht *ramht; + + struct nv50_disp_chan *chan[81]; }; void nv50_disp_super_1(struct nv50_disp *); @@ -34,11 +46,11 @@ void nv50_disp_super_2_2(struct nv50_disp *, struct nvkm_head *); void nv50_disp_super_3_0(struct nv50_disp *, struct nvkm_head *); int nv50_disp_new_(const struct nv50_disp_func *, struct nvkm_device *, - int index, int heads, struct nvkm_disp **); -int gf119_disp_new_(const struct nv50_disp_func *, struct nvkm_device *, - int index, struct nvkm_disp **); + int index, struct nvkm_disp **); struct nv50_disp_func { + int (*init)(struct nv50_disp *); + void (*fini)(struct nv50_disp *); void (*intr)(struct nv50_disp *); void (*intr_error)(struct nv50_disp *, int chid); @@ -48,28 +60,20 @@ struct nv50_disp_func { const struct nvkm_disp_oclass *root; struct { + int (*cnt)(struct nvkm_disp *, unsigned long *mask); int (*new)(struct nvkm_disp *, int id); - } head; + } wndw, head, dac, sor, pior; - struct { - int nr; - int (*new)(struct nvkm_disp *, int id); - } dac; - - struct { - int nr; - int (*new)(struct nvkm_disp *, int id); - } sor; - - struct { - int nr; - int (*new)(struct nvkm_disp *, int id); - } pior; + u16 ramht_size; }; +int nv50_disp_init(struct nv50_disp *); +void nv50_disp_fini(struct nv50_disp *); void nv50_disp_intr(struct nv50_disp *); void nv50_disp_super(struct work_struct *); +int gf119_disp_init(struct nv50_disp *); +void gf119_disp_fini(struct nv50_disp *); void gf119_disp_intr(struct nv50_disp *); void gf119_disp_super(struct work_struct *); void gf119_disp_intr_error(struct nv50_disp *, int); @@ -77,4 +81,12 @@ void gf119_disp_intr_error(struct nv50_disp *, int); void nv50_disp_dptmds_war_2(struct nv50_disp *, struct dcb_output *); void nv50_disp_dptmds_war_3(struct nv50_disp *, struct dcb_output *); void nv50_disp_update_sppll1(struct nv50_disp *); + +extern const struct nvkm_event_func nv50_disp_chan_uevent; +int nv50_disp_chan_uevent_ctor(struct nvkm_object *, void *, u32, + struct nvkm_notify *); +void nv50_disp_chan_uevent_send(struct nv50_disp *, int); + +extern const struct nvkm_event_func gf119_disp_chan_uevent; +extern const struct nvkm_event_func gv100_disp_chan_uevent; #endif diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/oimmgf119.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/oimmgf119.c index 1f9fd3403f07..1ae0bcfc89b9 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/oimmgf119.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/oimmgf119.c @@ -22,16 +22,11 @@ * Authors: Ben Skeggs */ #include "channv50.h" -#include "rootnv50.h" -#include <nvif/class.h> - -const struct nv50_disp_pioc_oclass -gf119_disp_oimm_oclass = { - .base.oclass = GF110_DISP_OVERLAY, - .base.minver = 0, - .base.maxver = 0, - .ctor = nv50_disp_oimm_new, - .func = &gf119_disp_pioc_func, - .chid = { 9, 9 }, -}; +int +gf119_disp_oimm_new(const struct nvkm_oclass *oclass, void *argv, u32 argc, + struct nv50_disp *disp, struct nvkm_object **pobject) +{ + return nv50_disp_oimm_new_(&gf119_disp_pioc_func, disp, 9, 9, + oclass, argv, argc, pobject); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/oimmgk104.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/oimmgk104.c deleted file mode 100644 index 0c09fe85e952..000000000000 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/oimmgk104.c +++ /dev/null @@ -1,37 +0,0 @@ -/* - * Copyright 2012 Red Hat Inc. - * - * Permission is hereby granted, free of charge, to any person obtaining a - * copy of this software and associated documentation files (the "Software"), - * to deal in the Software without restriction, including without limitation - * the rights to use, copy, modify, merge, publish, distribute, sublicense, - * and/or sell copies of the Software, and to permit persons to whom the - * Software is furnished to do so, subject to the following conditions: - * - * The above copyright notice and this permission notice shall be included in - * all copies or substantial portions of the Software. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL - * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR - * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, - * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR - * OTHER DEALINGS IN THE SOFTWARE. - * - * Authors: Ben Skeggs - */ -#include "channv50.h" -#include "rootnv50.h" - -#include <nvif/class.h> - -const struct nv50_disp_pioc_oclass -gk104_disp_oimm_oclass = { - .base.oclass = GK104_DISP_OVERLAY, - .base.minver = 0, - .base.maxver = 0, - .ctor = nv50_disp_oimm_new, - .func = &gf119_disp_pioc_func, - .chid = { 9, 9 }, -}; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/oimmgp102.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/oimmgp102.c index abf82365c671..30ffb1008505 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/oimmgp102.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/oimmgp102.c @@ -22,16 +22,11 @@ * Authors: Ben Skeggs <bskeggs@redhat.com> */ #include "channv50.h" -#include "rootnv50.h" -#include <nvif/class.h> - -const struct nv50_disp_pioc_oclass -gp102_disp_oimm_oclass = { - .base.oclass = GK104_DISP_OVERLAY, - .base.minver = 0, - .base.maxver = 0, - .ctor = nv50_disp_oimm_new, - .func = &gf119_disp_pioc_func, - .chid = { 9, 13 }, -}; +int +gp102_disp_oimm_new(const struct nvkm_oclass *oclass, void *argv, u32 argc, + struct nv50_disp *disp, struct nvkm_object **pobject) +{ + return nv50_disp_oimm_new_(&gf119_disp_pioc_func, disp, 9, 13, + oclass, argv, argc, pobject); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/oimmgt215.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/oimmgt215.c deleted file mode 100644 index 1281db28aebd..000000000000 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/oimmgt215.c +++ /dev/null @@ -1,37 +0,0 @@ -/* - * Copyright 2012 Red Hat Inc. - * - * Permission is hereby granted, free of charge, to any person obtaining a - * copy of this software and associated documentation files (the "Software"), - * to deal in the Software without restriction, including without limitation - * the rights to use, copy, modify, merge, publish, distribute, sublicense, - * and/or sell copies of the Software, and to permit persons to whom the - * Software is furnished to do so, subject to the following conditions: - * - * The above copyright notice and this permission notice shall be included in - * all copies or substantial portions of the Software. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL - * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR - * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, - * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR - * OTHER DEALINGS IN THE SOFTWARE. - * - * Authors: Ben Skeggs - */ -#include "channv50.h" -#include "rootnv50.h" - -#include <nvif/class.h> - -const struct nv50_disp_pioc_oclass -gt215_disp_oimm_oclass = { - .base.oclass = GT214_DISP_OVERLAY, - .base.minver = 0, - .base.maxver = 0, - .ctor = nv50_disp_oimm_new, - .func = &nv50_disp_pioc_func, - .chid = { 5, 5 }, -}; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/oimmnv50.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/oimmnv50.c index f3b0fa2c5924..0db99bfe9db9 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/oimmnv50.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/oimmnv50.c @@ -23,30 +23,26 @@ */ #include "channv50.h" #include "head.h" -#include "rootnv50.h" #include <core/client.h> -#include <nvif/class.h> #include <nvif/cl507b.h> #include <nvif/unpack.h> int -nv50_disp_oimm_new(const struct nv50_disp_chan_func *func, - const struct nv50_disp_chan_mthd *mthd, - struct nv50_disp_root *root, int ctrl, int user, - const struct nvkm_oclass *oclass, void *data, u32 size, - struct nvkm_object **pobject) +nv50_disp_oimm_new_(const struct nv50_disp_chan_func *func, + struct nv50_disp *disp, int ctrl, int user, + const struct nvkm_oclass *oclass, void *argv, u32 argc, + struct nvkm_object **pobject) { union { struct nv50_disp_overlay_v0 v0; - } *args = data; + } *args = argv; struct nvkm_object *parent = oclass->parent; - struct nv50_disp *disp = root->disp; int head, ret = -ENOSYS; - nvif_ioctl(parent, "create disp overlay size %d\n", size); - if (!(ret = nvif_unpack(ret, &data, &size, args->v0, 0, 0, false))) { + nvif_ioctl(parent, "create disp overlay size %d\n", argc); + if (!(ret = nvif_unpack(ret, &argv, &argc, args->v0, 0, 0, false))) { nvif_ioctl(parent, "create disp overlay vers %d head %d\n", args->v0.version, args->v0.head); if (!nvkm_head_find(&disp->base, args->v0.head)) @@ -55,16 +51,14 @@ nv50_disp_oimm_new(const struct nv50_disp_chan_func *func, } else return ret; - return nv50_disp_chan_new_(func, mthd, root, ctrl + head, user + head, + return nv50_disp_chan_new_(func, NULL, disp, ctrl + head, user + head, head, oclass, pobject); } -const struct nv50_disp_pioc_oclass -nv50_disp_oimm_oclass = { - .base.oclass = NV50_DISP_OVERLAY, - .base.minver = 0, - .base.maxver = 0, - .ctor = nv50_disp_oimm_new, - .func = &nv50_disp_pioc_func, - .chid = { 5, 5 }, -}; +int +nv50_disp_oimm_new(const struct nvkm_oclass *oclass, void *argv, u32 argc, + struct nv50_disp *disp, struct nvkm_object **pobject) +{ + return nv50_disp_oimm_new_(&nv50_disp_pioc_func, disp, 5, 5, + oclass, argv, argc, pobject); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/ovlyg84.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/ovlyg84.c index db6234eebc61..31b915d48699 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/ovlyg84.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/ovlyg84.c @@ -21,10 +21,7 @@ * * Authors: Ben Skeggs */ -#include "dmacnv50.h" -#include "rootnv50.h" - -#include <nvif/class.h> +#include "channv50.h" static const struct nv50_disp_mthd_list g84_disp_ovly_mthd_base = { @@ -54,8 +51,8 @@ g84_disp_ovly_mthd_base = { } }; -const struct nv50_disp_chan_mthd -g84_disp_ovly_chan_mthd = { +static const struct nv50_disp_chan_mthd +g84_disp_ovly_mthd = { .name = "Overlay", .addr = 0x000540, .prev = 0x000004, @@ -65,13 +62,10 @@ g84_disp_ovly_chan_mthd = { } }; -const struct nv50_disp_dmac_oclass -g84_disp_ovly_oclass = { - .base.oclass = G82_DISP_OVERLAY_CHANNEL_DMA, - .base.minver = 0, - .base.maxver = 0, - .ctor = nv50_disp_ovly_new, - .func = &nv50_disp_dmac_func, - .mthd = &g84_disp_ovly_chan_mthd, - .chid = 3, -}; +int +g84_disp_ovly_new(const struct nvkm_oclass *oclass, void *argv, u32 argc, + struct nv50_disp *disp, struct nvkm_object **pobject) +{ + return nv50_disp_ovly_new_(&nv50_disp_dmac_func, &g84_disp_ovly_mthd, + disp, 3, oclass, argv, argc, pobject); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/ovlygf119.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/ovlygf119.c index 5985879abd23..83fd534c44da 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/ovlygf119.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/ovlygf119.c @@ -21,10 +21,7 @@ * * Authors: Ben Skeggs */ -#include "dmacnv50.h" -#include "rootnv50.h" - -#include <nvif/class.h> +#include "channv50.h" static const struct nv50_disp_mthd_list gf119_disp_ovly_mthd_base = { @@ -79,7 +76,7 @@ gf119_disp_ovly_mthd_base = { }; static const struct nv50_disp_chan_mthd -gf119_disp_ovly_chan_mthd = { +gf119_disp_ovly_mthd = { .name = "Overlay", .addr = 0x001000, .prev = -0x020000, @@ -89,13 +86,10 @@ gf119_disp_ovly_chan_mthd = { } }; -const struct nv50_disp_dmac_oclass -gf119_disp_ovly_oclass = { - .base.oclass = GF110_DISP_OVERLAY_CONTROL_DMA, - .base.minver = 0, - .base.maxver = 0, - .ctor = nv50_disp_ovly_new, - .func = &gf119_disp_dmac_func, - .mthd = &gf119_disp_ovly_chan_mthd, - .chid = 5, -}; +int +gf119_disp_ovly_new(const struct nvkm_oclass *oclass, void *argv, u32 argc, + struct nv50_disp *disp, struct nvkm_object **pobject) +{ + return nv50_disp_ovly_new_(&gf119_disp_dmac_func, &gf119_disp_ovly_mthd, + disp, 5, oclass, argv, argc, pobject); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/ovlygk104.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/ovlygk104.c index 2f0220b39f34..a7acacbc92c1 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/ovlygk104.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/ovlygk104.c @@ -21,10 +21,7 @@ * * Authors: Ben Skeggs */ -#include "dmacnv50.h" -#include "rootnv50.h" - -#include <nvif/class.h> +#include "channv50.h" static const struct nv50_disp_mthd_list gk104_disp_ovly_mthd_base = { @@ -81,7 +78,7 @@ gk104_disp_ovly_mthd_base = { }; const struct nv50_disp_chan_mthd -gk104_disp_ovly_chan_mthd = { +gk104_disp_ovly_mthd = { .name = "Overlay", .addr = 0x001000, .prev = -0x020000, @@ -91,13 +88,10 @@ gk104_disp_ovly_chan_mthd = { } }; -const struct nv50_disp_dmac_oclass -gk104_disp_ovly_oclass = { - .base.oclass = GK104_DISP_OVERLAY_CONTROL_DMA, - .base.minver = 0, - .base.maxver = 0, - .ctor = nv50_disp_ovly_new, - .func = &gf119_disp_dmac_func, - .mthd = &gk104_disp_ovly_chan_mthd, - .chid = 5, -}; +int +gk104_disp_ovly_new(const struct nvkm_oclass *oclass, void *argv, u32 argc, + struct nv50_disp *disp, struct nvkm_object **pobject) +{ + return nv50_disp_ovly_new_(&gf119_disp_dmac_func, &gk104_disp_ovly_mthd, + disp, 5, oclass, argv, argc, pobject); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/ovlygp102.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/ovlygp102.c index 589bd2f12b41..e0eca6ea914c 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/ovlygp102.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/ovlygp102.c @@ -21,18 +21,12 @@ * * Authors: Ben Skeggs */ -#include "dmacnv50.h" -#include "rootnv50.h" +#include "channv50.h" -#include <nvif/class.h> - -const struct nv50_disp_dmac_oclass -gp102_disp_ovly_oclass = { - .base.oclass = GK104_DISP_OVERLAY_CONTROL_DMA, - .base.minver = 0, - .base.maxver = 0, - .ctor = nv50_disp_ovly_new, - .func = &gp102_disp_dmac_func, - .mthd = &gk104_disp_ovly_chan_mthd, - .chid = 5, -}; +int +gp102_disp_ovly_new(const struct nvkm_oclass *oclass, void *argv, u32 argc, + struct nv50_disp *disp, struct nvkm_object **pobject) +{ + return nv50_disp_ovly_new_(&gp102_disp_dmac_func, &gk104_disp_ovly_mthd, + disp, 5, oclass, argv, argc, pobject); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/ovlygt200.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/ovlygt200.c index f858053db83d..dc60cd00dc16 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/ovlygt200.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/ovlygt200.c @@ -21,10 +21,7 @@ * * Authors: Ben Skeggs */ -#include "dmacnv50.h" -#include "rootnv50.h" - -#include <nvif/class.h> +#include "channv50.h" static const struct nv50_disp_mthd_list gt200_disp_ovly_mthd_base = { @@ -58,7 +55,7 @@ gt200_disp_ovly_mthd_base = { }; static const struct nv50_disp_chan_mthd -gt200_disp_ovly_chan_mthd = { +gt200_disp_ovly_mthd = { .name = "Overlay", .addr = 0x000540, .prev = 0x000004, @@ -68,13 +65,10 @@ gt200_disp_ovly_chan_mthd = { } }; -const struct nv50_disp_dmac_oclass -gt200_disp_ovly_oclass = { - .base.oclass = GT200_DISP_OVERLAY_CHANNEL_DMA, - .base.minver = 0, - .base.maxver = 0, - .ctor = nv50_disp_ovly_new, - .func = &nv50_disp_dmac_func, - .mthd = >200_disp_ovly_chan_mthd, - .chid = 3, -}; +int +gt200_disp_ovly_new(const struct nvkm_oclass *oclass, void *argv, u32 argc, + struct nv50_disp *disp, struct nvkm_object **pobject) +{ + return nv50_disp_ovly_new_(&nv50_disp_dmac_func, >200_disp_ovly_mthd, + disp, 3, oclass, argv, argc, pobject); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/ovlygt215.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/ovlygt215.c deleted file mode 100644 index c947e1e16a37..000000000000 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/ovlygt215.c +++ /dev/null @@ -1,38 +0,0 @@ -/* - * Copyright 2015 Red Hat Inc. - * - * Permission is hereby granted, free of charge, to any person obtaining a - * copy of this software and associated documentation files (the "Software"), - * to deal in the Software without restriction, including without limitation - * the rights to use, copy, modify, merge, publish, distribute, sublicense, - * and/or sell copies of the Software, and to permit persons to whom the - * Software is furnished to do so, subject to the following conditions: - * - * The above copyright notice and this permission notice shall be included in - * all copies or substantial portions of the Software. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL - * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR - * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, - * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR - * OTHER DEALINGS IN THE SOFTWARE. - * - * Authors: Ben Skeggs <bskeggs@redhat.com> - */ -#include "dmacnv50.h" -#include "rootnv50.h" - -#include <nvif/class.h> - -const struct nv50_disp_dmac_oclass -gt215_disp_ovly_oclass = { - .base.oclass = GT214_DISP_OVERLAY_CHANNEL_DMA, - .base.minver = 0, - .base.maxver = 0, - .ctor = nv50_disp_ovly_new, - .func = &nv50_disp_dmac_func, - .mthd = &g84_disp_ovly_chan_mthd, - .chid = 3, -}; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/ovlynv50.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/ovlynv50.c index 9ebaaa6e9e33..6974c12c4518 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/ovlynv50.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/ovlynv50.c @@ -21,33 +21,30 @@ * * Authors: Ben Skeggs */ -#include "dmacnv50.h" +#include "channv50.h" #include "head.h" -#include "rootnv50.h" #include <core/client.h> -#include <nvif/class.h> #include <nvif/cl507e.h> #include <nvif/unpack.h> int -nv50_disp_ovly_new(const struct nv50_disp_dmac_func *func, - const struct nv50_disp_chan_mthd *mthd, - struct nv50_disp_root *root, int chid, - const struct nvkm_oclass *oclass, void *data, u32 size, - struct nvkm_object **pobject) +nv50_disp_ovly_new_(const struct nv50_disp_chan_func *func, + const struct nv50_disp_chan_mthd *mthd, + struct nv50_disp *disp, int chid, + const struct nvkm_oclass *oclass, void *argv, u32 argc, + struct nvkm_object **pobject) { union { struct nv50_disp_overlay_channel_dma_v0 v0; - } *args = data; + } *args = argv; struct nvkm_object *parent = oclass->parent; - struct nv50_disp *disp = root->disp; int head, ret = -ENOSYS; u64 push; - nvif_ioctl(parent, "create disp overlay channel dma size %d\n", size); - if (!(ret = nvif_unpack(ret, &data, &size, args->v0, 0, 0, false))) { + nvif_ioctl(parent, "create disp overlay channel dma size %d\n", argc); + if (!(ret = nvif_unpack(ret, &argv, &argc, args->v0, 0, 0, false))) { nvif_ioctl(parent, "create disp overlay channel dma vers %d " "pushbuf %016llx head %d\n", args->v0.version, args->v0.pushbuf, args->v0.head); @@ -58,7 +55,7 @@ nv50_disp_ovly_new(const struct nv50_disp_dmac_func *func, } else return ret; - return nv50_disp_dmac_new_(func, mthd, root, chid + head, + return nv50_disp_dmac_new_(func, mthd, disp, chid + head, head, push, oclass, pobject); } @@ -91,7 +88,7 @@ nv50_disp_ovly_mthd_base = { }; static const struct nv50_disp_chan_mthd -nv50_disp_ovly_chan_mthd = { +nv50_disp_ovly_mthd = { .name = "Overlay", .addr = 0x000540, .prev = 0x000004, @@ -101,13 +98,10 @@ nv50_disp_ovly_chan_mthd = { } }; -const struct nv50_disp_dmac_oclass -nv50_disp_ovly_oclass = { - .base.oclass = NV50_DISP_OVERLAY_CHANNEL_DMA, - .base.minver = 0, - .base.maxver = 0, - .ctor = nv50_disp_ovly_new, - .func = &nv50_disp_dmac_func, - .mthd = &nv50_disp_ovly_chan_mthd, - .chid = 3, -}; +int +nv50_disp_ovly_new(const struct nvkm_oclass *oclass, void *argv, u32 argc, + struct nv50_disp *disp, struct nvkm_object **pobject) +{ + return nv50_disp_ovly_new_(&nv50_disp_dmac_func, &nv50_disp_ovly_mthd, + disp, 3, oclass, argv, argc, pobject); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/piocgf119.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/piocgf119.c index 0abaa6431943..5296e7bee813 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/piocgf119.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/piocgf119.c @@ -29,7 +29,7 @@ static void gf119_disp_pioc_fini(struct nv50_disp_chan *chan) { - struct nv50_disp *disp = chan->root->disp; + struct nv50_disp *disp = chan->disp; struct nvkm_subdev *subdev = &disp->base.engine.subdev; struct nvkm_device *device = subdev->device; int ctrl = chan->chid.ctrl; @@ -43,24 +43,17 @@ gf119_disp_pioc_fini(struct nv50_disp_chan *chan) nvkm_error(subdev, "ch %d fini: %08x\n", user, nvkm_rd32(device, 0x610490 + (ctrl * 0x10))); } - - /* disable error reporting and completion notification */ - nvkm_mask(device, 0x610090, 0x00000001 << user, 0x00000000); - nvkm_mask(device, 0x6100a0, 0x00000001 << user, 0x00000000); } static int gf119_disp_pioc_init(struct nv50_disp_chan *chan) { - struct nv50_disp *disp = chan->root->disp; + struct nv50_disp *disp = chan->disp; struct nvkm_subdev *subdev = &disp->base.engine.subdev; struct nvkm_device *device = subdev->device; int ctrl = chan->chid.ctrl; int user = chan->chid.user; - /* enable error reporting */ - nvkm_mask(device, 0x6100a0, 0x00000001 << user, 0x00000001 << user); - /* activate channel */ nvkm_wr32(device, 0x610490 + (ctrl * 0x10), 0x00000001); if (nvkm_msec(device, 2000, @@ -80,4 +73,6 @@ const struct nv50_disp_chan_func gf119_disp_pioc_func = { .init = gf119_disp_pioc_init, .fini = gf119_disp_pioc_fini, + .intr = gf119_disp_chan_intr, + .user = nv50_disp_chan_user, }; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/piocnv50.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/piocnv50.c index 0211e0e8a35f..4faed6fce682 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/piocnv50.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/piocnv50.c @@ -29,7 +29,7 @@ static void nv50_disp_pioc_fini(struct nv50_disp_chan *chan) { - struct nv50_disp *disp = chan->root->disp; + struct nv50_disp *disp = chan->disp; struct nvkm_subdev *subdev = &disp->base.engine.subdev; struct nvkm_device *device = subdev->device; int ctrl = chan->chid.ctrl; @@ -48,7 +48,7 @@ nv50_disp_pioc_fini(struct nv50_disp_chan *chan) static int nv50_disp_pioc_init(struct nv50_disp_chan *chan) { - struct nv50_disp *disp = chan->root->disp; + struct nv50_disp *disp = chan->disp; struct nvkm_subdev *subdev = &disp->base.engine.subdev; struct nvkm_device *device = subdev->device; int ctrl = chan->chid.ctrl; @@ -82,4 +82,6 @@ const struct nv50_disp_chan_func nv50_disp_pioc_func = { .init = nv50_disp_pioc_init, .fini = nv50_disp_pioc_fini, + .intr = nv50_disp_chan_intr, + .user = nv50_disp_chan_user, }; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/piornv50.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/piornv50.c index 99b3b9050635..e997a207f546 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/piornv50.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/piornv50.c @@ -127,8 +127,13 @@ nv50_pior = { int nv50_pior_new(struct nvkm_disp *disp, int id) { - struct nvkm_device *device = disp->engine.subdev.device; - if (!(nvkm_rd32(device, 0x610184) & (0x10000000 << id))) - return 0; return nvkm_ior_new_(&nv50_pior, disp, PIOR, id); } + +int +nv50_pior_cnt(struct nvkm_disp *disp, unsigned long *pmask) +{ + struct nvkm_device *device = disp->engine.subdev.device; + *pmask = (nvkm_rd32(device, 0x610184) & 0x70000000) >> 28; + return 3; +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/priv.h b/drivers/gpu/drm/nouveau/nvkm/engine/disp/priv.h index 6c9bfff6d043..ef66c5f38ad5 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/priv.h +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/priv.h @@ -12,6 +12,9 @@ void nvkm_disp_vblank(struct nvkm_disp *, int head); struct nvkm_disp_func { void *(*dtor)(struct nvkm_disp *); + int (*oneinit)(struct nvkm_disp *); + int (*init)(struct nvkm_disp *); + void (*fini)(struct nvkm_disp *); void (*intr)(struct nvkm_disp *); const struct nvkm_disp_oclass *(*root)(struct nvkm_disp *); diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootg84.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootg84.c index 721e4f74d1fc..1ed371fd7ddf 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootg84.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootg84.c @@ -22,22 +22,19 @@ * Authors: Ben Skeggs */ #include "rootnv50.h" -#include "dmacnv50.h" +#include "channv50.h" #include <nvif/class.h> static const struct nv50_disp_root_func g84_disp_root = { - .init = nv50_disp_root_init, - .fini = nv50_disp_root_fini, - .dmac = { - &g84_disp_core_oclass, - &g84_disp_base_oclass, - &g84_disp_ovly_oclass, - }, - .pioc = { - &g84_disp_oimm_oclass, - &g84_disp_curs_oclass, + .user = { + {{0,0,G82_DISP_CURSOR }, nv50_disp_curs_new }, + {{0,0,G82_DISP_OVERLAY }, nv50_disp_oimm_new }, + {{0,0,G82_DISP_BASE_CHANNEL_DMA }, g84_disp_base_new }, + {{0,0,G82_DISP_CORE_CHANNEL_DMA }, g84_disp_core_new }, + {{0,0,G82_DISP_OVERLAY_CHANNEL_DMA}, g84_disp_ovly_new }, + {} }, }; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootg94.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootg94.c index 9493f6edf62b..ef579eb00238 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootg94.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootg94.c @@ -22,22 +22,19 @@ * Authors: Ben Skeggs */ #include "rootnv50.h" -#include "dmacnv50.h" +#include "channv50.h" #include <nvif/class.h> static const struct nv50_disp_root_func g94_disp_root = { - .init = nv50_disp_root_init, - .fini = nv50_disp_root_fini, - .dmac = { - &g94_disp_core_oclass, - >200_disp_base_oclass, - >200_disp_ovly_oclass, - }, - .pioc = { - &g84_disp_oimm_oclass, - &g84_disp_curs_oclass, + .user = { + {{0,0, G82_DISP_CURSOR }, nv50_disp_curs_new }, + {{0,0, G82_DISP_OVERLAY }, nv50_disp_oimm_new }, + {{0,0,GT200_DISP_BASE_CHANNEL_DMA }, g84_disp_base_new }, + {{0,0,GT206_DISP_CORE_CHANNEL_DMA }, g94_disp_core_new }, + {{0,0,GT200_DISP_OVERLAY_CHANNEL_DMA}, gt200_disp_ovly_new }, + {} }, }; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootgf119.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootgf119.c index 333c8424b413..fe011165dc02 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootgf119.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootgf119.c @@ -22,104 +22,19 @@ * Authors: Ben Skeggs */ #include "rootnv50.h" -#include "head.h" -#include "dmacnv50.h" - -#include <core/ramht.h> -#include <subdev/timer.h> +#include "channv50.h" #include <nvif/class.h> -void -gf119_disp_root_fini(struct nv50_disp_root *root) -{ - struct nvkm_device *device = root->disp->base.engine.subdev.device; - /* disable all interrupts */ - nvkm_wr32(device, 0x6100b0, 0x00000000); -} - -int -gf119_disp_root_init(struct nv50_disp_root *root) -{ - struct nv50_disp *disp = root->disp; - struct nvkm_head *head; - struct nvkm_device *device = disp->base.engine.subdev.device; - u32 tmp; - int i; - - /* The below segments of code copying values from one register to - * another appear to inform EVO of the display capabilities or - * something similar. - */ - - /* ... CRTC caps */ - list_for_each_entry(head, &disp->base.head, head) { - const u32 hoff = head->id * 0x800; - tmp = nvkm_rd32(device, 0x616104 + hoff); - nvkm_wr32(device, 0x6101b4 + hoff, tmp); - tmp = nvkm_rd32(device, 0x616108 + hoff); - nvkm_wr32(device, 0x6101b8 + hoff, tmp); - tmp = nvkm_rd32(device, 0x61610c + hoff); - nvkm_wr32(device, 0x6101bc + hoff, tmp); - } - - /* ... DAC caps */ - for (i = 0; i < disp->func->dac.nr; i++) { - tmp = nvkm_rd32(device, 0x61a000 + (i * 0x800)); - nvkm_wr32(device, 0x6101c0 + (i * 0x800), tmp); - } - - /* ... SOR caps */ - for (i = 0; i < disp->func->sor.nr; i++) { - tmp = nvkm_rd32(device, 0x61c000 + (i * 0x800)); - nvkm_wr32(device, 0x6301c4 + (i * 0x800), tmp); - } - - /* steal display away from vbios, or something like that */ - if (nvkm_rd32(device, 0x6100ac) & 0x00000100) { - nvkm_wr32(device, 0x6100ac, 0x00000100); - nvkm_mask(device, 0x6194e8, 0x00000001, 0x00000000); - if (nvkm_msec(device, 2000, - if (!(nvkm_rd32(device, 0x6194e8) & 0x00000002)) - break; - ) < 0) - return -EBUSY; - } - - /* point at display engine memory area (hash table, objects) */ - nvkm_wr32(device, 0x610010, (root->instmem->addr >> 8) | 9); - - /* enable supervisor interrupts, disable everything else */ - nvkm_wr32(device, 0x610090, 0x00000000); - nvkm_wr32(device, 0x6100a0, 0x00000000); - nvkm_wr32(device, 0x6100b0, 0x00000307); - - /* disable underflow reporting, preventing an intermittent issue - * on some gk104 boards where the production vbios left this - * setting enabled by default. - * - * ftp://download.nvidia.com/open-gpu-doc/gk104-disable-underflow-reporting/1/gk104-disable-underflow-reporting.txt - */ - list_for_each_entry(head, &disp->base.head, head) { - const u32 hoff = head->id * 0x800; - nvkm_mask(device, 0x616308 + hoff, 0x00000111, 0x00000010); - } - - return 0; -} - static const struct nv50_disp_root_func gf119_disp_root = { - .init = gf119_disp_root_init, - .fini = gf119_disp_root_fini, - .dmac = { - &gf119_disp_core_oclass, - &gf119_disp_base_oclass, - &gf119_disp_ovly_oclass, - }, - .pioc = { - &gf119_disp_oimm_oclass, - &gf119_disp_curs_oclass, + .user = { + {{0,0,GF110_DISP_CURSOR }, gf119_disp_curs_new }, + {{0,0,GF110_DISP_OVERLAY }, gf119_disp_oimm_new }, + {{0,0,GF110_DISP_BASE_CHANNEL_DMA }, gf119_disp_base_new }, + {{0,0,GF110_DISP_CORE_CHANNEL_DMA }, gf119_disp_core_new }, + {{0,0,GF110_DISP_OVERLAY_CONTROL_DMA}, gf119_disp_ovly_new }, + {} }, }; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootgk104.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootgk104.c index 0bfdb1d1c6ab..9e8ffd348b50 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootgk104.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootgk104.c @@ -22,22 +22,19 @@ * Authors: Ben Skeggs */ #include "rootnv50.h" -#include "dmacnv50.h" +#include "channv50.h" #include <nvif/class.h> static const struct nv50_disp_root_func gk104_disp_root = { - .init = gf119_disp_root_init, - .fini = gf119_disp_root_fini, - .dmac = { - &gk104_disp_core_oclass, - &gk104_disp_base_oclass, - &gk104_disp_ovly_oclass, - }, - .pioc = { - &gk104_disp_oimm_oclass, - &gk104_disp_curs_oclass, + .user = { + {{0,0,GK104_DISP_CURSOR }, gf119_disp_curs_new }, + {{0,0,GK104_DISP_OVERLAY }, gf119_disp_oimm_new }, + {{0,0,GK104_DISP_BASE_CHANNEL_DMA }, gf119_disp_base_new }, + {{0,0,GK104_DISP_CORE_CHANNEL_DMA }, gk104_disp_core_new }, + {{0,0,GK104_DISP_OVERLAY_CONTROL_DMA}, gk104_disp_ovly_new }, + {} }, }; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootgk110.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootgk110.c index 1e8dbed8a67c..dc85cc1c9490 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootgk110.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootgk110.c @@ -22,22 +22,19 @@ * Authors: Ben Skeggs */ #include "rootnv50.h" -#include "dmacnv50.h" +#include "channv50.h" #include <nvif/class.h> static const struct nv50_disp_root_func gk110_disp_root = { - .init = gf119_disp_root_init, - .fini = gf119_disp_root_fini, - .dmac = { - &gk110_disp_core_oclass, - &gk110_disp_base_oclass, - &gk104_disp_ovly_oclass, - }, - .pioc = { - &gk104_disp_oimm_oclass, - &gk104_disp_curs_oclass, + .user = { + {{0,0,GK104_DISP_CURSOR }, gf119_disp_curs_new }, + {{0,0,GK104_DISP_OVERLAY }, gf119_disp_oimm_new }, + {{0,0,GK110_DISP_BASE_CHANNEL_DMA }, gf119_disp_base_new }, + {{0,0,GK110_DISP_CORE_CHANNEL_DMA }, gk104_disp_core_new }, + {{0,0,GK104_DISP_OVERLAY_CONTROL_DMA}, gk104_disp_ovly_new }, + {} }, }; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootgm107.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootgm107.c index 44c55be69e99..e0181ca08840 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootgm107.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootgm107.c @@ -22,22 +22,19 @@ * Authors: Ben Skeggs */ #include "rootnv50.h" -#include "dmacnv50.h" +#include "channv50.h" #include <nvif/class.h> static const struct nv50_disp_root_func gm107_disp_root = { - .init = gf119_disp_root_init, - .fini = gf119_disp_root_fini, - .dmac = { - &gm107_disp_core_oclass, - &gk110_disp_base_oclass, - &gk104_disp_ovly_oclass, - }, - .pioc = { - &gk104_disp_oimm_oclass, - &gk104_disp_curs_oclass, + .user = { + {{0,0,GK104_DISP_CURSOR }, gf119_disp_curs_new }, + {{0,0,GK104_DISP_OVERLAY }, gf119_disp_oimm_new }, + {{0,0,GK110_DISP_BASE_CHANNEL_DMA }, gf119_disp_base_new }, + {{0,0,GM107_DISP_CORE_CHANNEL_DMA }, gk104_disp_core_new }, + {{0,0,GK104_DISP_OVERLAY_CONTROL_DMA}, gk104_disp_ovly_new }, + {} }, }; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootgm200.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootgm200.c index 38f5ee1dfc58..e5e590e19f62 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootgm200.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootgm200.c @@ -22,22 +22,19 @@ * Authors: Ben Skeggs */ #include "rootnv50.h" -#include "dmacnv50.h" +#include "channv50.h" #include <nvif/class.h> static const struct nv50_disp_root_func gm200_disp_root = { - .init = gf119_disp_root_init, - .fini = gf119_disp_root_fini, - .dmac = { - &gm200_disp_core_oclass, - &gk110_disp_base_oclass, - &gk104_disp_ovly_oclass, - }, - .pioc = { - &gk104_disp_oimm_oclass, - &gk104_disp_curs_oclass, + .user = { + {{0,0,GK104_DISP_CURSOR }, gf119_disp_curs_new }, + {{0,0,GK104_DISP_OVERLAY }, gf119_disp_oimm_new }, + {{0,0,GK110_DISP_BASE_CHANNEL_DMA }, gf119_disp_base_new }, + {{0,0,GM200_DISP_CORE_CHANNEL_DMA }, gk104_disp_core_new }, + {{0,0,GK104_DISP_OVERLAY_CONTROL_DMA}, gk104_disp_ovly_new }, + {} }, }; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootgp100.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootgp100.c index ac8fdd728ec6..762a1a922e05 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootgp100.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootgp100.c @@ -22,22 +22,19 @@ * Authors: Ben Skeggs <bskeggs@redhat.com> */ #include "rootnv50.h" -#include "dmacnv50.h" +#include "channv50.h" #include <nvif/class.h> static const struct nv50_disp_root_func gp100_disp_root = { - .init = gf119_disp_root_init, - .fini = gf119_disp_root_fini, - .dmac = { - &gp100_disp_core_oclass, - &gk110_disp_base_oclass, - &gk104_disp_ovly_oclass, - }, - .pioc = { - &gk104_disp_oimm_oclass, - &gk104_disp_curs_oclass, + .user = { + {{0,0,GK104_DISP_CURSOR }, gf119_disp_curs_new }, + {{0,0,GK104_DISP_OVERLAY }, gf119_disp_oimm_new }, + {{0,0,GK110_DISP_BASE_CHANNEL_DMA }, gf119_disp_base_new }, + {{0,0,GP100_DISP_CORE_CHANNEL_DMA }, gk104_disp_core_new }, + {{0,0,GK104_DISP_OVERLAY_CONTROL_DMA}, gk104_disp_ovly_new }, + {} }, }; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootgp102.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootgp102.c index 37122ca579ad..c7f00946c9af 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootgp102.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootgp102.c @@ -22,22 +22,19 @@ * Authors: Ben Skeggs <bskeggs@redhat.com> */ #include "rootnv50.h" -#include "dmacnv50.h" +#include "channv50.h" #include <nvif/class.h> static const struct nv50_disp_root_func gp102_disp_root = { - .init = gf119_disp_root_init, - .fini = gf119_disp_root_fini, - .dmac = { - &gp102_disp_core_oclass, - &gp102_disp_base_oclass, - &gp102_disp_ovly_oclass, - }, - .pioc = { - &gp102_disp_oimm_oclass, - &gp102_disp_curs_oclass, + .user = { + {{0,0,GK104_DISP_CURSOR }, gp102_disp_curs_new }, + {{0,0,GK104_DISP_OVERLAY }, gp102_disp_oimm_new }, + {{0,0,GK110_DISP_BASE_CHANNEL_DMA }, gp102_disp_base_new }, + {{0,0,GP102_DISP_CORE_CHANNEL_DMA }, gp102_disp_core_new }, + {{0,0,GK104_DISP_OVERLAY_CONTROL_DMA}, gp102_disp_ovly_new }, + {} }, }; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootgt200.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootgt200.c index 124a0c24f92c..a6963654087c 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootgt200.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootgt200.c @@ -22,22 +22,19 @@ * Authors: Ben Skeggs */ #include "rootnv50.h" -#include "dmacnv50.h" +#include "channv50.h" #include <nvif/class.h> static const struct nv50_disp_root_func gt200_disp_root = { - .init = nv50_disp_root_init, - .fini = nv50_disp_root_fini, - .dmac = { - >200_disp_core_oclass, - >200_disp_base_oclass, - >200_disp_ovly_oclass, - }, - .pioc = { - &g84_disp_oimm_oclass, - &g84_disp_curs_oclass, + .user = { + {{0,0, G82_DISP_CURSOR }, nv50_disp_curs_new }, + {{0,0, G82_DISP_OVERLAY }, nv50_disp_oimm_new }, + {{0,0,GT200_DISP_BASE_CHANNEL_DMA }, g84_disp_base_new }, + {{0,0,GT200_DISP_CORE_CHANNEL_DMA }, g84_disp_core_new }, + {{0,0,GT200_DISP_OVERLAY_CHANNEL_DMA}, gt200_disp_ovly_new }, + {} }, }; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootgt215.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootgt215.c index dff52f30668b..4fe0a3ae8891 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootgt215.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootgt215.c @@ -22,22 +22,19 @@ * Authors: Ben Skeggs */ #include "rootnv50.h" -#include "dmacnv50.h" +#include "channv50.h" #include <nvif/class.h> static const struct nv50_disp_root_func gt215_disp_root = { - .init = nv50_disp_root_init, - .fini = nv50_disp_root_fini, - .dmac = { - >215_disp_core_oclass, - >215_disp_base_oclass, - >215_disp_ovly_oclass, - }, - .pioc = { - >215_disp_oimm_oclass, - >215_disp_curs_oclass, + .user = { + {{0,0,GT214_DISP_CURSOR }, nv50_disp_curs_new }, + {{0,0,GT214_DISP_OVERLAY }, nv50_disp_oimm_new }, + {{0,0,GT214_DISP_BASE_CHANNEL_DMA }, g84_disp_base_new }, + {{0,0,GT214_DISP_CORE_CHANNEL_DMA }, g94_disp_core_new }, + {{0,0,GT214_DISP_OVERLAY_CHANNEL_DMA}, g84_disp_ovly_new }, + {} }, }; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootgv100.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootgv100.c new file mode 100644 index 000000000000..9c658d632d37 --- /dev/null +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootgv100.c @@ -0,0 +1,52 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "rootnv50.h" +#include "channv50.h" + +#include <nvif/class.h> + +static const struct nv50_disp_root_func +gv100_disp_root = { + .user = { + {{0,0,GV100_DISP_CURSOR }, gv100_disp_curs_new }, + {{0,0,GV100_DISP_WINDOW_IMM_CHANNEL_DMA}, gv100_disp_wimm_new }, + {{0,0,GV100_DISP_CORE_CHANNEL_DMA }, gv100_disp_core_new }, + {{0,0,GV100_DISP_WINDOW_CHANNEL_DMA }, gv100_disp_wndw_new }, + {} + }, +}; + +static int +gv100_disp_root_new(struct nvkm_disp *disp, const struct nvkm_oclass *oclass, + void *data, u32 size, struct nvkm_object **pobject) +{ + return nv50_disp_root_new_(&gv100_disp_root, disp, oclass, + data, size, pobject); +} + +const struct nvkm_disp_oclass +gv100_disp_root_oclass = { + .base.oclass = GV100_DISP, + .base.minver = -1, + .base.maxver = -1, + .ctor = gv100_disp_root_new, +}; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootnv50.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootnv50.c index 1208524aae14..3aa5a2879239 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootnv50.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootnv50.c @@ -22,14 +22,12 @@ * Authors: Ben Skeggs */ #include "rootnv50.h" -#include "dmacnv50.h" +#include "channv50.h" #include "dp.h" #include "head.h" #include "ior.h" #include <core/client.h> -#include <core/ramht.h> -#include <subdev/timer.h> #include <nvif/class.h> #include <nvif/cl5070.h> @@ -271,23 +269,12 @@ nv50_disp_root_mthd_(struct nvkm_object *object, u32 mthd, void *data, u32 size) } static int -nv50_disp_root_dmac_new_(const struct nvkm_oclass *oclass, - void *data, u32 size, struct nvkm_object **pobject) +nv50_disp_root_child_new_(const struct nvkm_oclass *oclass, + void *argv, u32 argc, struct nvkm_object **pobject) { - const struct nv50_disp_dmac_oclass *sclass = oclass->priv; - struct nv50_disp_root *root = nv50_disp_root(oclass->parent); - return sclass->ctor(sclass->func, sclass->mthd, root, sclass->chid, - oclass, data, size, pobject); -} - -static int -nv50_disp_root_pioc_new_(const struct nvkm_oclass *oclass, - void *data, u32 size, struct nvkm_object **pobject) -{ - const struct nv50_disp_pioc_oclass *sclass = oclass->priv; - struct nv50_disp_root *root = nv50_disp_root(oclass->parent); - return sclass->ctor(sclass->func, sclass->mthd, root, sclass->chid.ctrl, - sclass->chid.user, oclass, data, size, pobject); + struct nv50_disp *disp = nv50_disp_root(oclass->parent)->disp; + const struct nv50_disp_user *user = oclass->priv; + return user->ctor(oclass, argv, argc, disp, pobject); } static int @@ -296,68 +283,26 @@ nv50_disp_root_child_get_(struct nvkm_object *object, int index, { struct nv50_disp_root *root = nv50_disp_root(object); - if (index < ARRAY_SIZE(root->func->dmac)) { - sclass->base = root->func->dmac[index]->base; - sclass->priv = root->func->dmac[index]; - sclass->ctor = nv50_disp_root_dmac_new_; - return 0; - } - - index -= ARRAY_SIZE(root->func->dmac); - - if (index < ARRAY_SIZE(root->func->pioc)) { - sclass->base = root->func->pioc[index]->base; - sclass->priv = root->func->pioc[index]; - sclass->ctor = nv50_disp_root_pioc_new_; + if (root->func->user[index].ctor) { + sclass->base = root->func->user[index].base; + sclass->priv = root->func->user + index; + sclass->ctor = nv50_disp_root_child_new_; return 0; } return -EINVAL; } -static int -nv50_disp_root_fini_(struct nvkm_object *object, bool suspend) -{ - struct nv50_disp_root *root = nv50_disp_root(object); - root->func->fini(root); - return 0; -} - -static int -nv50_disp_root_init_(struct nvkm_object *object) -{ - struct nv50_disp_root *root = nv50_disp_root(object); - struct nvkm_ior *ior; - int ret; - - ret = root->func->init(root); - if (ret) - return ret; - - /* Set 'normal' (ie. when it's attached to a head) state for - * each output resource to 'fully enabled'. - */ - list_for_each_entry(ior, &root->disp->base.ior, head) { - ior->func->power(ior, true, true, true, true, true); - } - - return 0; -} - static void * nv50_disp_root_dtor_(struct nvkm_object *object) { struct nv50_disp_root *root = nv50_disp_root(object); - nvkm_ramht_del(&root->ramht); - nvkm_gpuobj_del(&root->instmem); return root; } static const struct nvkm_object_func nv50_disp_root_ = { .dtor = nv50_disp_root_dtor_, - .init = nv50_disp_root_init_, - .fini = nv50_disp_root_fini_, .mthd = nv50_disp_root_mthd_, .ntfy = nvkm_disp_ntfy, .sclass = nv50_disp_root_child_get_, @@ -370,8 +315,6 @@ nv50_disp_root_new_(const struct nv50_disp_root_func *func, { struct nv50_disp *disp = nv50_disp(base); struct nv50_disp_root *root; - struct nvkm_device *device = disp->base.engine.subdev.device; - int ret; if (!(root = kzalloc(sizeof(*root), GFP_KERNEL))) return -ENOMEM; @@ -380,102 +323,18 @@ nv50_disp_root_new_(const struct nv50_disp_root_func *func, nvkm_object_ctor(&nv50_disp_root_, oclass, &root->object); root->func = func; root->disp = disp; - - ret = nvkm_gpuobj_new(disp->base.engine.subdev.device, 0x10000, 0x10000, - false, NULL, &root->instmem); - if (ret) - return ret; - - return nvkm_ramht_new(device, 0x1000, 0, root->instmem, &root->ramht); -} - -void -nv50_disp_root_fini(struct nv50_disp_root *root) -{ - struct nvkm_device *device = root->disp->base.engine.subdev.device; - /* disable all interrupts */ - nvkm_wr32(device, 0x610024, 0x00000000); - nvkm_wr32(device, 0x610020, 0x00000000); -} - -int -nv50_disp_root_init(struct nv50_disp_root *root) -{ - struct nv50_disp *disp = root->disp; - struct nvkm_head *head; - struct nvkm_device *device = disp->base.engine.subdev.device; - u32 tmp; - int i; - - /* The below segments of code copying values from one register to - * another appear to inform EVO of the display capabilities or - * something similar. NFI what the 0x614004 caps are for.. - */ - tmp = nvkm_rd32(device, 0x614004); - nvkm_wr32(device, 0x610184, tmp); - - /* ... CRTC caps */ - list_for_each_entry(head, &disp->base.head, head) { - tmp = nvkm_rd32(device, 0x616100 + (head->id * 0x800)); - nvkm_wr32(device, 0x610190 + (head->id * 0x10), tmp); - tmp = nvkm_rd32(device, 0x616104 + (head->id * 0x800)); - nvkm_wr32(device, 0x610194 + (head->id * 0x10), tmp); - tmp = nvkm_rd32(device, 0x616108 + (head->id * 0x800)); - nvkm_wr32(device, 0x610198 + (head->id * 0x10), tmp); - tmp = nvkm_rd32(device, 0x61610c + (head->id * 0x800)); - nvkm_wr32(device, 0x61019c + (head->id * 0x10), tmp); - } - - /* ... DAC caps */ - for (i = 0; i < disp->func->dac.nr; i++) { - tmp = nvkm_rd32(device, 0x61a000 + (i * 0x800)); - nvkm_wr32(device, 0x6101d0 + (i * 0x04), tmp); - } - - /* ... SOR caps */ - for (i = 0; i < disp->func->sor.nr; i++) { - tmp = nvkm_rd32(device, 0x61c000 + (i * 0x800)); - nvkm_wr32(device, 0x6101e0 + (i * 0x04), tmp); - } - - /* ... PIOR caps */ - for (i = 0; i < disp->func->pior.nr; i++) { - tmp = nvkm_rd32(device, 0x61e000 + (i * 0x800)); - nvkm_wr32(device, 0x6101f0 + (i * 0x04), tmp); - } - - /* steal display away from vbios, or something like that */ - if (nvkm_rd32(device, 0x610024) & 0x00000100) { - nvkm_wr32(device, 0x610024, 0x00000100); - nvkm_mask(device, 0x6194e8, 0x00000001, 0x00000000); - if (nvkm_msec(device, 2000, - if (!(nvkm_rd32(device, 0x6194e8) & 0x00000002)) - break; - ) < 0) - return -EBUSY; - } - - /* point at display engine memory area (hash table, objects) */ - nvkm_wr32(device, 0x610010, (root->instmem->addr >> 8) | 9); - - /* enable supervisor interrupts, disable everything else */ - nvkm_wr32(device, 0x61002c, 0x00000370); - nvkm_wr32(device, 0x610028, 0x00000000); return 0; } static const struct nv50_disp_root_func nv50_disp_root = { - .init = nv50_disp_root_init, - .fini = nv50_disp_root_fini, - .dmac = { - &nv50_disp_core_oclass, - &nv50_disp_base_oclass, - &nv50_disp_ovly_oclass, - }, - .pioc = { - &nv50_disp_oimm_oclass, - &nv50_disp_curs_oclass, + .user = { + {{0,0,NV50_DISP_CURSOR }, nv50_disp_curs_new }, + {{0,0,NV50_DISP_OVERLAY }, nv50_disp_oimm_new }, + {{0,0,NV50_DISP_BASE_CHANNEL_DMA }, nv50_disp_base_new }, + {{0,0,NV50_DISP_CORE_CHANNEL_DMA }, nv50_disp_core_new }, + {{0,0,NV50_DISP_OVERLAY_CHANNEL_DMA}, nv50_disp_ovly_new }, + {} }, }; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootnv50.h b/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootnv50.h index 4818fa69ae6c..6ca4f9184b51 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootnv50.h +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootnv50.h @@ -2,34 +2,27 @@ #ifndef __NV50_DISP_ROOT_H__ #define __NV50_DISP_ROOT_H__ #define nv50_disp_root(p) container_of((p), struct nv50_disp_root, object) +#include <core/object.h> #include "nv50.h" -#include "channv50.h" -#include "dmacnv50.h" struct nv50_disp_root { const struct nv50_disp_root_func *func; struct nv50_disp *disp; struct nvkm_object object; - - struct nvkm_gpuobj *instmem; - struct nvkm_ramht *ramht; }; struct nv50_disp_root_func { - int (*init)(struct nv50_disp_root *); - void (*fini)(struct nv50_disp_root *); - const struct nv50_disp_dmac_oclass *dmac[3]; - const struct nv50_disp_pioc_oclass *pioc[2]; + int blah; + struct nv50_disp_user { + struct nvkm_sclass base; + int (*ctor)(const struct nvkm_oclass *, void *argv, u32 argc, + struct nv50_disp *, struct nvkm_object **); + } user[]; }; int nv50_disp_root_new_(const struct nv50_disp_root_func *, struct nvkm_disp *, const struct nvkm_oclass *, void *data, u32 size, struct nvkm_object **); -int nv50_disp_root_init(struct nv50_disp_root *); -void nv50_disp_root_fini(struct nv50_disp_root *); - -int gf119_disp_root_init(struct nv50_disp_root *); -void gf119_disp_root_fini(struct nv50_disp_root *); extern const struct nvkm_disp_oclass nv50_disp_root_oclass; extern const struct nvkm_disp_oclass g84_disp_root_oclass; @@ -43,4 +36,5 @@ extern const struct nvkm_disp_oclass gm107_disp_root_oclass; extern const struct nvkm_disp_oclass gm200_disp_root_oclass; extern const struct nvkm_disp_oclass gp100_disp_root_oclass; extern const struct nvkm_disp_oclass gp102_disp_root_oclass; +extern const struct nvkm_disp_oclass gv100_disp_root_oclass; #endif diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorg84.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorg84.c index f40b909b4ca2..ec3a7db08118 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorg84.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorg84.c @@ -34,5 +34,5 @@ g84_sor = { int g84_sor_new(struct nvkm_disp *disp, int id) { - return nv50_sor_new_(&g84_sor, disp, id); + return nvkm_ior_new_(&g84_sor, disp, SOR, id); } diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorg94.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorg94.c index 49aeafde0031..4d59d02525d9 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorg94.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorg94.c @@ -279,5 +279,13 @@ g94_sor = { int g94_sor_new(struct nvkm_disp *disp, int id) { - return nv50_sor_new_(&g94_sor, disp, id); + return nvkm_ior_new_(&g94_sor, disp, SOR, id); +} + +int +g94_sor_cnt(struct nvkm_disp *disp, unsigned long *pmask) +{ + struct nvkm_device *device = disp->engine.subdev.device; + *pmask = (nvkm_rd32(device, 0x610184) & 0x0f000000) >> 24; + return 4; } diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorgf119.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorgf119.c index 700fc754f28a..e6e6dfbb1283 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorgf119.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorgf119.c @@ -152,15 +152,6 @@ gf119_sor_state(struct nvkm_ior *sor, struct nvkm_ior_state *state) state->head = ctrl & 0x0000000f; } -int -gf119_sor_new_(const struct nvkm_ior_func *func, struct nvkm_disp *disp, int id) -{ - struct nvkm_device *device = disp->engine.subdev.device; - if (!(nvkm_rd32(device, 0x612004) & (0x00000100 << id))) - return 0; - return nvkm_ior_new_(func, disp, SOR, id); -} - static const struct nvkm_ior_func gf119_sor = { .state = gf119_sor_state, @@ -189,5 +180,13 @@ gf119_sor = { int gf119_sor_new(struct nvkm_disp *disp, int id) { - return gf119_sor_new_(&gf119_sor, disp, id); + return nvkm_ior_new_(&gf119_sor, disp, SOR, id); +} + +int +gf119_sor_cnt(struct nvkm_disp *disp, unsigned long *pmask) +{ + struct nvkm_device *device = disp->engine.subdev.device; + *pmask = (nvkm_rd32(device, 0x612004) & 0x0000ff00) >> 8; + return 8; } diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorgk104.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorgk104.c index a1547bdf490b..b94090edaebf 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorgk104.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorgk104.c @@ -49,5 +49,5 @@ gk104_sor = { int gk104_sor_new(struct nvkm_disp *disp, int id) { - return gf119_sor_new_(&gk104_sor, disp, id); + return nvkm_ior_new_(&gk104_sor, disp, SOR, id); } diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorgm107.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorgm107.c index 60230957d82b..e6965dec09c9 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorgm107.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorgm107.c @@ -63,5 +63,5 @@ gm107_sor = { int gm107_sor_new(struct nvkm_disp *disp, int id) { - return gf119_sor_new_(&gm107_sor, disp, id); + return nvkm_ior_new_(&gm107_sor, disp, SOR, id); } diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorgm200.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorgm200.c index f9b8107aa2a2..d892bdf04034 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorgm200.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorgm200.c @@ -23,7 +23,7 @@ */ #include "ior.h" -static void +void gm200_sor_dp_drive(struct nvkm_ior *sor, int ln, int pc, int dc, int pe, int pu) { struct nvkm_device *device = sor->disp->engine.subdev.device; @@ -45,7 +45,7 @@ gm200_sor_dp_drive(struct nvkm_ior *sor, int ln, int pc, int dc, int pe, int pu) nvkm_wr32(device, 0x61c13c + loff, data[3] | (pc << shift)); } -static void +void gm200_sor_route_set(struct nvkm_outp *outp, struct nvkm_ior *ior) { struct nvkm_device *device = outp->disp->engine.subdev.device; @@ -62,7 +62,7 @@ gm200_sor_route_set(struct nvkm_outp *outp, struct nvkm_ior *ior) nvkm_mask(device, 0x612388 + moff, 0x0000001f, link << 4 | sor); } -static int +int gm200_sor_route_get(struct nvkm_outp *outp, int *link) { struct nvkm_device *device = outp->disp->engine.subdev.device; @@ -120,5 +120,5 @@ gm200_sor = { int gm200_sor_new(struct nvkm_disp *disp, int id) { - return gf119_sor_new_(&gm200_sor, disp, id); + return nvkm_ior_new_(&gm200_sor, disp, SOR, id); } diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorgt215.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorgt215.c index da228b54b43e..54d134d4ca1d 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorgt215.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorgt215.c @@ -65,5 +65,5 @@ gt215_sor = { int gt215_sor_new(struct nvkm_disp *disp, int id) { - return nv50_sor_new_(>215_sor, disp, id); + return nvkm_ior_new_(>215_sor, disp, SOR, id); } diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorgv100.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorgv100.c new file mode 100644 index 000000000000..040db8a338de --- /dev/null +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorgv100.c @@ -0,0 +1,120 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "ior.h" + +#include <subdev/timer.h> + +static void +gv100_sor_dp_watermark(struct nvkm_ior *sor, int head, u8 watermark) +{ + struct nvkm_device *device = sor->disp->engine.subdev.device; + const u32 hoff = head * 0x800; + nvkm_mask(device, 0x616550 + hoff, 0x0c00003f, 0x08000000 | watermark); +} + +static void +gv100_sor_dp_audio_sym(struct nvkm_ior *sor, int head, u16 h, u32 v) +{ + struct nvkm_device *device = sor->disp->engine.subdev.device; + const u32 hoff = head * 0x800; + nvkm_mask(device, 0x616568 + hoff, 0x0000ffff, h); + nvkm_mask(device, 0x61656c + hoff, 0x00ffffff, v); +} + +static void +gv100_sor_dp_audio(struct nvkm_ior *sor, int head, bool enable) +{ + struct nvkm_device *device = sor->disp->engine.subdev.device; + const u32 hoff = 0x800 * head; + const u32 data = 0x80000000 | (0x00000001 * enable); + const u32 mask = 0x8000000d; + nvkm_mask(device, 0x616560 + hoff, mask, data); + nvkm_msec(device, 2000, + if (!(nvkm_rd32(device, 0x616560 + hoff) & 0x80000000)) + break; + ); +} + +static void +gv100_sor_state(struct nvkm_ior *sor, struct nvkm_ior_state *state) +{ + struct nvkm_device *device = sor->disp->engine.subdev.device; + const u32 coff = (state == &sor->arm) * 0x8000 + sor->id * 0x20; + u32 ctrl = nvkm_rd32(device, 0x680300 + coff); + + state->proto_evo = (ctrl & 0x00000f00) >> 8; + switch (state->proto_evo) { + case 0: state->proto = LVDS; state->link = 1; break; + case 1: state->proto = TMDS; state->link = 1; break; + case 2: state->proto = TMDS; state->link = 2; break; + case 5: state->proto = TMDS; state->link = 3; break; + case 8: state->proto = DP; state->link = 1; break; + case 9: state->proto = DP; state->link = 2; break; + default: + state->proto = UNKNOWN; + break; + } + + state->head = ctrl & 0x000000ff; +} + +static const struct nvkm_ior_func +gv100_sor = { + .route = { + .get = gm200_sor_route_get, + .set = gm200_sor_route_set, + }, + .state = gv100_sor_state, + .power = nv50_sor_power, + .clock = gf119_sor_clock, + .hdmi = { + .ctrl = gv100_hdmi_ctrl, + }, + .dp = { + .lanes = { 0, 1, 2, 3 }, + .links = gf119_sor_dp_links, + .power = g94_sor_dp_power, + .pattern = gm107_sor_dp_pattern, + .drive = gm200_sor_dp_drive, + .audio = gv100_sor_dp_audio, + .audio_sym = gv100_sor_dp_audio_sym, + .watermark = gv100_sor_dp_watermark, + }, + .hda = { + .hpd = gf119_hda_hpd, + .eld = gf119_hda_eld, + }, +}; + +int +gv100_sor_new(struct nvkm_disp *disp, int id) +{ + return nvkm_ior_new_(&gv100_sor, disp, SOR, id); +} + +int +gv100_sor_cnt(struct nvkm_disp *disp, unsigned long *pmask) +{ + struct nvkm_device *device = disp->engine.subdev.device; + *pmask = (nvkm_rd32(device, 0x610060) & 0x0000ff00) >> 8; + return (nvkm_rd32(device, 0x610074) & 0x00000f00) >> 8; +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/sormcp77.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/sormcp77.c index c0179ccb956d..8a70dd25b13a 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/sormcp77.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/sormcp77.c @@ -44,5 +44,5 @@ mcp77_sor = { int mcp77_sor_new(struct nvkm_disp *disp, int id) { - return nv50_sor_new_(&mcp77_sor, disp, id); + return nvkm_ior_new_(&mcp77_sor, disp, SOR, id); } diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/sormcp89.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/sormcp89.c index 9bb01cd96697..eac9c5be9166 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/sormcp89.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/sormcp89.c @@ -49,5 +49,5 @@ mcp89_sor = { int mcp89_sor_new(struct nvkm_disp *disp, int id) { - return nv50_sor_new_(&mcp89_sor, disp, id); + return nvkm_ior_new_(&mcp89_sor, disp, SOR, id); } diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/sornv50.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/sornv50.c index f3ebd0c22e7d..b4729f8798af 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/sornv50.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/sornv50.c @@ -84,15 +84,6 @@ nv50_sor_state(struct nvkm_ior *sor, struct nvkm_ior_state *state) state->head = ctrl & 0x00000003; } -int -nv50_sor_new_(const struct nvkm_ior_func *func, struct nvkm_disp *disp, int id) -{ - struct nvkm_device *device = disp->engine.subdev.device; - if (!(nvkm_rd32(device, 0x610184) & (0x01000000 << id))) - return 0; - return nvkm_ior_new_(func, disp, SOR, id); -} - static const struct nvkm_ior_func nv50_sor = { .state = nv50_sor_state, @@ -103,5 +94,13 @@ nv50_sor = { int nv50_sor_new(struct nvkm_disp *disp, int id) { - return nv50_sor_new_(&nv50_sor, disp, id); + return nvkm_ior_new_(&nv50_sor, disp, SOR, id); +} + +int +nv50_sor_cnt(struct nvkm_disp *disp, unsigned long *pmask) +{ + struct nvkm_device *device = disp->engine.subdev.device; + *pmask = (nvkm_rd32(device, 0x610184) & 0x03000000) >> 24; + return 2; } diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/wimmgv100.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/wimmgv100.c new file mode 100644 index 000000000000..89d783368b4f --- /dev/null +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/wimmgv100.c @@ -0,0 +1,82 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "channv50.h" + +#include <core/client.h> + +#include <nvif/clc37b.h> +#include <nvif/unpack.h> + +static void +gv100_disp_wimm_intr(struct nv50_disp_chan *chan, bool en) +{ + struct nvkm_device *device = chan->disp->base.engine.subdev.device; + const u32 mask = 0x00000001 << chan->head; + const u32 data = en ? mask : 0; + nvkm_mask(device, 0x611da8, mask, data); +} + +const struct nv50_disp_chan_func +gv100_disp_wimm = { + .init = gv100_disp_dmac_init, + .fini = gv100_disp_dmac_fini, + .intr = gv100_disp_wimm_intr, + .user = gv100_disp_chan_user, +}; + +static int +gv100_disp_wimm_new_(const struct nv50_disp_chan_func *func, + const struct nv50_disp_chan_mthd *mthd, + struct nv50_disp *disp, int chid, + const struct nvkm_oclass *oclass, void *argv, u32 argc, + struct nvkm_object **pobject) +{ + union { + struct nvc37b_window_imm_channel_dma_v0 v0; + } *args = argv; + struct nvkm_object *parent = oclass->parent; + int wndw, ret = -ENOSYS; + u64 push; + + nvif_ioctl(parent, "create window imm channel dma size %d\n", argc); + if (!(ret = nvif_unpack(ret, &argv, &argc, args->v0, 0, 0, false))) { + nvif_ioctl(parent, "create window imm channel dma vers %d " + "pushbuf %016llx index %d\n", + args->v0.version, args->v0.pushbuf, args->v0.index); + if (!(disp->wndw.mask & BIT(args->v0.index))) + return -EINVAL; + push = args->v0.pushbuf; + wndw = args->v0.index; + } else + return ret; + + return nv50_disp_dmac_new_(func, mthd, disp, chid + wndw, + wndw, push, oclass, pobject); +} + +int +gv100_disp_wimm_new(const struct nvkm_oclass *oclass, void *argv, u32 argc, + struct nv50_disp *disp, struct nvkm_object **pobject) +{ + return gv100_disp_wimm_new_(&gv100_disp_wimm, NULL, disp, 33, + oclass, argv, argc, pobject); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/wndwgv100.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/wndwgv100.c new file mode 100644 index 000000000000..98911805aabf --- /dev/null +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/wndwgv100.c @@ -0,0 +1,184 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "channv50.h" + +#include <core/client.h> + +#include <nvif/clc37e.h> +#include <nvif/unpack.h> + +static const struct nv50_disp_mthd_list +gv100_disp_wndw_mthd_base = { + .mthd = 0x0000, + .addr = 0x000000, + .data = { + { 0x0200, 0x690200 }, + { 0x020c, 0x69020c }, + { 0x0210, 0x690210 }, + { 0x0214, 0x690214 }, + { 0x0218, 0x690218 }, + { 0x021c, 0x69021c }, + { 0x0220, 0x690220 }, + { 0x0224, 0x690224 }, + { 0x0228, 0x690228 }, + { 0x022c, 0x69022c }, + { 0x0230, 0x690230 }, + { 0x0234, 0x690234 }, + { 0x0238, 0x690238 }, + { 0x0240, 0x690240 }, + { 0x0244, 0x690244 }, + { 0x0248, 0x690248 }, + { 0x024c, 0x69024c }, + { 0x0250, 0x690250 }, + { 0x0254, 0x690254 }, + { 0x0260, 0x690260 }, + { 0x0264, 0x690264 }, + { 0x0268, 0x690268 }, + { 0x026c, 0x69026c }, + { 0x0270, 0x690270 }, + { 0x0274, 0x690274 }, + { 0x0280, 0x690280 }, + { 0x0284, 0x690284 }, + { 0x0288, 0x690288 }, + { 0x028c, 0x69028c }, + { 0x0290, 0x690290 }, + { 0x0298, 0x690298 }, + { 0x029c, 0x69029c }, + { 0x02a0, 0x6902a0 }, + { 0x02a4, 0x6902a4 }, + { 0x02a8, 0x6902a8 }, + { 0x02ac, 0x6902ac }, + { 0x02b0, 0x6902b0 }, + { 0x02b4, 0x6902b4 }, + { 0x02b8, 0x6902b8 }, + { 0x02bc, 0x6902bc }, + { 0x02c0, 0x6902c0 }, + { 0x02c4, 0x6902c4 }, + { 0x02c8, 0x6902c8 }, + { 0x02cc, 0x6902cc }, + { 0x02d0, 0x6902d0 }, + { 0x02d4, 0x6902d4 }, + { 0x02d8, 0x6902d8 }, + { 0x02dc, 0x6902dc }, + { 0x02e0, 0x6902e0 }, + { 0x02e4, 0x6902e4 }, + { 0x02e8, 0x6902e8 }, + { 0x02ec, 0x6902ec }, + { 0x02f0, 0x6902f0 }, + { 0x02f4, 0x6902f4 }, + { 0x02f8, 0x6902f8 }, + { 0x02fc, 0x6902fc }, + { 0x0300, 0x690300 }, + { 0x0304, 0x690304 }, + { 0x0308, 0x690308 }, + { 0x0310, 0x690310 }, + { 0x0314, 0x690314 }, + { 0x0318, 0x690318 }, + { 0x031c, 0x69031c }, + { 0x0320, 0x690320 }, + { 0x0324, 0x690324 }, + { 0x0328, 0x690328 }, + { 0x032c, 0x69032c }, + { 0x033c, 0x69033c }, + { 0x0340, 0x690340 }, + { 0x0344, 0x690344 }, + { 0x0348, 0x690348 }, + { 0x034c, 0x69034c }, + { 0x0350, 0x690350 }, + { 0x0354, 0x690354 }, + { 0x0358, 0x690358 }, + { 0x0364, 0x690364 }, + { 0x0368, 0x690368 }, + { 0x036c, 0x69036c }, + { 0x0370, 0x690370 }, + { 0x0374, 0x690374 }, + { 0x0380, 0x690380 }, + {} + } +}; + +const struct nv50_disp_chan_mthd +gv100_disp_wndw_mthd = { + .name = "Base", + .addr = 0x001000, + .prev = 0x000800, + .data = { + { "Global", 1, &gv100_disp_wndw_mthd_base }, + {} + } +}; + +static void +gv100_disp_wndw_intr(struct nv50_disp_chan *chan, bool en) +{ + struct nvkm_device *device = chan->disp->base.engine.subdev.device; + const u32 mask = 0x00000001 << chan->head; + const u32 data = en ? mask : 0; + nvkm_mask(device, 0x611da4, mask, data); +} + +const struct nv50_disp_chan_func +gv100_disp_wndw = { + .init = gv100_disp_dmac_init, + .fini = gv100_disp_dmac_fini, + .intr = gv100_disp_wndw_intr, + .user = gv100_disp_chan_user, + .bind = gv100_disp_dmac_bind, +}; + +static int +gv100_disp_wndw_new_(const struct nv50_disp_chan_func *func, + const struct nv50_disp_chan_mthd *mthd, + struct nv50_disp *disp, int chid, + const struct nvkm_oclass *oclass, void *argv, u32 argc, + struct nvkm_object **pobject) +{ + union { + struct nvc37e_window_channel_dma_v0 v0; + } *args = argv; + struct nvkm_object *parent = oclass->parent; + int wndw, ret = -ENOSYS; + u64 push; + + nvif_ioctl(parent, "create window channel dma size %d\n", argc); + if (!(ret = nvif_unpack(ret, &argv, &argc, args->v0, 0, 0, false))) { + nvif_ioctl(parent, "create window channel dma vers %d " + "pushbuf %016llx index %d\n", + args->v0.version, args->v0.pushbuf, args->v0.index); + if (!(disp->wndw.mask & BIT(args->v0.index))) + return -EINVAL; + push = args->v0.pushbuf; + wndw = args->v0.index; + } else + return ret; + + return nv50_disp_dmac_new_(func, mthd, disp, chid + wndw, + wndw, push, oclass, pobject); +} + +int +gv100_disp_wndw_new(const struct nvkm_oclass *oclass, void *argv, u32 argc, + struct nv50_disp *disp, struct nvkm_object **pobject) +{ + return gv100_disp_wndw_new_(&gv100_disp_wndw, &gv100_disp_wndw_mthd, + disp, 1, oclass, argv, argc, pobject); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/dma/Kbuild b/drivers/gpu/drm/nouveau/nvkm/engine/dma/Kbuild index c4a2ce9b0d71..e96d1f57f9f9 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/dma/Kbuild +++ b/drivers/gpu/drm/nouveau/nvkm/engine/dma/Kbuild @@ -3,9 +3,11 @@ nvkm-y += nvkm/engine/dma/nv04.o nvkm-y += nvkm/engine/dma/nv50.o nvkm-y += nvkm/engine/dma/gf100.o nvkm-y += nvkm/engine/dma/gf119.o +nvkm-y += nvkm/engine/dma/gv100.o nvkm-y += nvkm/engine/dma/user.o nvkm-y += nvkm/engine/dma/usernv04.o nvkm-y += nvkm/engine/dma/usernv50.o nvkm-y += nvkm/engine/dma/usergf100.o nvkm-y += nvkm/engine/dma/usergf119.o +nvkm-y += nvkm/engine/dma/usergv100.o diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/oimmg84.c b/drivers/gpu/drm/nouveau/nvkm/engine/dma/gv100.c index 5ad5d0f5db05..c65a4c2ea93d 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/oimmg84.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/dma/gv100.c @@ -1,5 +1,5 @@ /* - * Copyright 2012 Red Hat Inc. + * Copyright 2018 Red Hat Inc. * * Permission is hereby granted, free of charge, to any person obtaining a * copy of this software and associated documentation files (the "Software"), @@ -18,20 +18,17 @@ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR * OTHER DEALINGS IN THE SOFTWARE. - * - * Authors: Ben Skeggs */ -#include "channv50.h" -#include "rootnv50.h" - -#include <nvif/class.h> +#include "priv.h" +#include "user.h" -const struct nv50_disp_pioc_oclass -g84_disp_oimm_oclass = { - .base.oclass = G82_DISP_OVERLAY, - .base.minver = 0, - .base.maxver = 0, - .ctor = nv50_disp_oimm_new, - .func = &nv50_disp_pioc_func, - .chid = { 5, 5 }, +static const struct nvkm_dma_func +gv100_dma = { + .class_new = gv100_dmaobj_new, }; + +int +gv100_dma_new(struct nvkm_device *device, int index, struct nvkm_dma **pdma) +{ + return nvkm_dma_new_(&gv100_dma, device, index, pdma); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/dma/user.h b/drivers/gpu/drm/nouveau/nvkm/engine/dma/user.h index 4bbac8a21c71..9fe01fd75474 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/dma/user.h +++ b/drivers/gpu/drm/nouveau/nvkm/engine/dma/user.h @@ -16,4 +16,6 @@ int gf100_dmaobj_new(struct nvkm_dma *, const struct nvkm_oclass *, void *, u32, struct nvkm_dmaobj **); int gf119_dmaobj_new(struct nvkm_dma *, const struct nvkm_oclass *, void *, u32, struct nvkm_dmaobj **); +int gv100_dmaobj_new(struct nvkm_dma *, const struct nvkm_oclass *, void *, u32, + struct nvkm_dmaobj **); #endif diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/dma/usergv100.c b/drivers/gpu/drm/nouveau/nvkm/engine/dma/usergv100.c new file mode 100644 index 000000000000..39eba9fc82be --- /dev/null +++ b/drivers/gpu/drm/nouveau/nvkm/engine/dma/usergv100.c @@ -0,0 +1,119 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#define gv100_dmaobj(p) container_of((p), struct gv100_dmaobj, base) +#include "user.h" + +#include <core/client.h> +#include <core/gpuobj.h> +#include <subdev/fb.h> + +#include <nvif/cl0002.h> +#include <nvif/unpack.h> + +struct gv100_dmaobj { + struct nvkm_dmaobj base; + u32 flags0; +}; + +static int +gv100_dmaobj_bind(struct nvkm_dmaobj *base, struct nvkm_gpuobj *parent, + int align, struct nvkm_gpuobj **pgpuobj) +{ + struct gv100_dmaobj *dmaobj = gv100_dmaobj(base); + struct nvkm_device *device = dmaobj->base.dma->engine.subdev.device; + u64 start = dmaobj->base.start >> 8; + u64 limit = dmaobj->base.limit >> 8; + int ret; + + ret = nvkm_gpuobj_new(device, 24, align, false, parent, pgpuobj); + if (ret == 0) { + nvkm_kmap(*pgpuobj); + nvkm_wo32(*pgpuobj, 0x00, dmaobj->flags0); + nvkm_wo32(*pgpuobj, 0x04, lower_32_bits(start)); + nvkm_wo32(*pgpuobj, 0x08, upper_32_bits(start)); + nvkm_wo32(*pgpuobj, 0x0c, lower_32_bits(limit)); + nvkm_wo32(*pgpuobj, 0x10, upper_32_bits(limit)); + nvkm_done(*pgpuobj); + } + + return ret; +} + +static const struct nvkm_dmaobj_func +gv100_dmaobj_func = { + .bind = gv100_dmaobj_bind, +}; + +int +gv100_dmaobj_new(struct nvkm_dma *dma, const struct nvkm_oclass *oclass, + void *data, u32 size, struct nvkm_dmaobj **pdmaobj) +{ + union { + struct gf119_dma_v0 v0; + } *args; + struct nvkm_object *parent = oclass->parent; + struct gv100_dmaobj *dmaobj; + u32 kind, page; + int ret; + + if (!(dmaobj = kzalloc(sizeof(*dmaobj), GFP_KERNEL))) + return -ENOMEM; + *pdmaobj = &dmaobj->base; + + ret = nvkm_dmaobj_ctor(&gv100_dmaobj_func, dma, oclass, + &data, &size, &dmaobj->base); + if (ret) + return ret; + + ret = -ENOSYS; + args = data; + + nvif_ioctl(parent, "create gv100 dma size %d\n", size); + if (!(ret = nvif_unpack(ret, &data, &size, args->v0, 0, 0, false))) { + nvif_ioctl(parent, + "create gv100 dma vers %d page %d kind %02x\n", + args->v0.version, args->v0.page, args->v0.kind); + kind = args->v0.kind != 0; + page = args->v0.page != 0; + } else + if (size == 0) { + kind = 0; + page = GF119_DMA_V0_PAGE_SP; + } else + return ret; + + if (kind) + dmaobj->flags0 |= 0x00100000; + if (page) + dmaobj->flags0 |= 0x00000040; + dmaobj->flags0 |= 0x00000004; /* rw */ + + switch (dmaobj->base.target) { + case NV_MEM_TARGET_VRAM : dmaobj->flags0 |= 0x00000001; break; + case NV_MEM_TARGET_PCI : dmaobj->flags0 |= 0x00000002; break; + case NV_MEM_TARGET_PCI_NOSNOOP: dmaobj->flags0 |= 0x00000003; break; + default: + return -EINVAL; + } + + return 0; +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/Kbuild b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/Kbuild index 64e51838edf8..f00408577a6a 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/Kbuild +++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/Kbuild @@ -15,6 +15,7 @@ nvkm-y += nvkm/engine/fifo/gm200.o nvkm-y += nvkm/engine/fifo/gm20b.o nvkm-y += nvkm/engine/fifo/gp100.o nvkm-y += nvkm/engine/fifo/gp10b.o +nvkm-y += nvkm/engine/fifo/gv100.o nvkm-y += nvkm/engine/fifo/chan.o nvkm-y += nvkm/engine/fifo/channv50.o @@ -31,6 +32,6 @@ nvkm-y += nvkm/engine/fifo/gpfifonv50.o nvkm-y += nvkm/engine/fifo/gpfifog84.o nvkm-y += nvkm/engine/fifo/gpfifogf100.o nvkm-y += nvkm/engine/fifo/gpfifogk104.o -nvkm-y += nvkm/engine/fifo/gpfifogk110.o -nvkm-y += nvkm/engine/fifo/gpfifogm200.o -nvkm-y += nvkm/engine/fifo/gpfifogp100.o +nvkm-y += nvkm/engine/fifo/gpfifogv100.o + +nvkm-y += nvkm/engine/fifo/usergv100.o diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/base.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/base.c index 64f6b7654a08..c773caf21f6b 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/base.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/base.c @@ -30,6 +30,7 @@ #include <subdev/mc.h> #include <nvif/event.h> +#include <nvif/cl0080.h> #include <nvif/unpack.h> void @@ -56,6 +57,12 @@ nvkm_fifo_start(struct nvkm_fifo *fifo, unsigned long *flags) } void +nvkm_fifo_fault(struct nvkm_fifo *fifo, struct nvkm_fault_data *info) +{ + return fifo->func->fault(fifo, info); +} + +void nvkm_fifo_chan_put(struct nvkm_fifo *fifo, unsigned long flags, struct nvkm_fifo_chan **pchan) { @@ -209,6 +216,20 @@ nvkm_fifo_uevent(struct nvkm_fifo *fifo) } static int +nvkm_fifo_class_new_(struct nvkm_device *device, + const struct nvkm_oclass *oclass, void *data, u32 size, + struct nvkm_object **pobject) +{ + struct nvkm_fifo *fifo = nvkm_fifo(oclass->engine); + return fifo->func->class_new(fifo, oclass, data, size, pobject); +} + +static const struct nvkm_device_oclass +nvkm_fifo_class_ = { + .ctor = nvkm_fifo_class_new_, +}; + +static int nvkm_fifo_class_new(struct nvkm_device *device, const struct nvkm_oclass *oclass, void *data, u32 size, struct nvkm_object **pobject) @@ -232,13 +253,9 @@ nvkm_fifo_class_get(struct nvkm_oclass *oclass, int index, int c = 0; if (fifo->func->class_get) { - int ret = fifo->func->class_get(fifo, index, &sclass); - if (ret == 0) { - oclass->base = sclass->base; - oclass->engn = sclass; - *class = &nvkm_fifo_class; - return 0; - } + int ret = fifo->func->class_get(fifo, index, oclass); + if (ret == 0) + *class = &nvkm_fifo_class_; return ret; } @@ -271,6 +288,20 @@ nvkm_fifo_fini(struct nvkm_engine *engine, bool suspend) } static int +nvkm_fifo_info(struct nvkm_engine *engine, u64 mthd, u64 *data) +{ + struct nvkm_fifo *fifo = nvkm_fifo(engine); + switch (mthd) { + case NV_DEVICE_FIFO_CHANNELS: *data = fifo->nr; return 0; + default: + if (fifo->func->info) + return fifo->func->info(fifo, mthd, data); + break; + } + return -ENOSYS; +} + +static int nvkm_fifo_oneinit(struct nvkm_engine *engine) { struct nvkm_fifo *fifo = nvkm_fifo(engine); @@ -311,6 +342,7 @@ nvkm_fifo = { .dtor = nvkm_fifo_dtor, .preinit = nvkm_fifo_preinit, .oneinit = nvkm_fifo_oneinit, + .info = nvkm_fifo_info, .init = nvkm_fifo_init, .fini = nvkm_fifo_fini, .intr = nvkm_fifo_intr, diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/cgrp.h b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/cgrp.h new file mode 100644 index 000000000000..d0ac60b06720 --- /dev/null +++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/cgrp.h @@ -0,0 +1,11 @@ +#ifndef __NVKM_FIFO_CGRP_H__ +#define __NVKM_FIFO_CGRP_H__ +#include "priv.h" + +struct nvkm_fifo_cgrp { + int id; + struct list_head head; + struct list_head chan; + int chan_nr; +}; +#endif diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/changk104.h b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/changk104.h index 1208e3d9dbe2..8e28ba6b2307 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/changk104.h +++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/changk104.h @@ -10,6 +10,7 @@ struct gk104_fifo_chan { struct gk104_fifo *fifo; int runl; + struct nvkm_fifo_cgrp *cgrp; struct list_head head; bool killed; @@ -19,11 +20,20 @@ struct gk104_fifo_chan { } engn[NVKM_SUBDEV_NR]; }; -int gk104_fifo_gpfifo_new(struct nvkm_fifo *, const struct nvkm_oclass *, +extern const struct nvkm_fifo_chan_func gk104_fifo_gpfifo_func; + +int gk104_fifo_gpfifo_new(struct gk104_fifo *, const struct nvkm_oclass *, void *data, u32 size, struct nvkm_object **); +void *gk104_fifo_gpfifo_dtor(struct nvkm_fifo_chan *); +void gk104_fifo_gpfifo_init(struct nvkm_fifo_chan *); +void gk104_fifo_gpfifo_fini(struct nvkm_fifo_chan *); +int gk104_fifo_gpfifo_engine_ctor(struct nvkm_fifo_chan *, struct nvkm_engine *, + struct nvkm_object *); +void gk104_fifo_gpfifo_engine_dtor(struct nvkm_fifo_chan *, + struct nvkm_engine *); +int gk104_fifo_gpfifo_kick(struct gk104_fifo_chan *); +int gk104_fifo_gpfifo_kick_locked(struct gk104_fifo_chan *); -extern const struct nvkm_fifo_chan_oclass gk104_fifo_gpfifo_oclass; -extern const struct nvkm_fifo_chan_oclass gk110_fifo_gpfifo_oclass; -extern const struct nvkm_fifo_chan_oclass gm200_fifo_gpfifo_oclass; -extern const struct nvkm_fifo_chan_oclass gp100_fifo_gpfifo_oclass; +int gv100_fifo_gpfifo_new(struct gk104_fifo *, const struct nvkm_oclass *, + void *data, u32 size, struct nvkm_object **); #endif diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk104.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk104.c index 84bd703dd897..a99046414a18 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk104.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk104.c @@ -22,16 +22,19 @@ * Authors: Ben Skeggs */ #include "gk104.h" +#include "cgrp.h" #include "changk104.h" #include <core/client.h> #include <core/gpuobj.h> #include <subdev/bar.h> +#include <subdev/fault.h> #include <subdev/timer.h> #include <subdev/top.h> #include <engine/sw.h> #include <nvif/class.h> +#include <nvif/cl0080.h> struct gk104_fifo_engine_status { bool busy; @@ -93,15 +96,39 @@ gk104_fifo_engine_status(struct gk104_fifo *fifo, int engn, } static int +gk104_fifo_class_new(struct nvkm_fifo *base, const struct nvkm_oclass *oclass, + void *argv, u32 argc, struct nvkm_object **pobject) +{ + struct gk104_fifo *fifo = gk104_fifo(base); + if (oclass->engn == &fifo->func->chan) { + const struct gk104_fifo_chan_user *user = oclass->engn; + return user->ctor(fifo, oclass, argv, argc, pobject); + } else + if (oclass->engn == &fifo->func->user) { + const struct gk104_fifo_user_user *user = oclass->engn; + return user->ctor(oclass, argv, argc, pobject); + } + WARN_ON(1); + return -EINVAL; +} + +static int gk104_fifo_class_get(struct nvkm_fifo *base, int index, - const struct nvkm_fifo_chan_oclass **psclass) + struct nvkm_oclass *oclass) { struct gk104_fifo *fifo = gk104_fifo(base); int c = 0; - while ((*psclass = fifo->func->chan[c])) { - if (c++ == index) - return 0; + if (fifo->func->user.ctor && c++ == index) { + oclass->base = fifo->func->user.user; + oclass->engn = &fifo->func->user; + return 0; + } + + if (fifo->func->chan.ctor && c++ == index) { + oclass->base = fifo->func->chan.user; + oclass->engn = &fifo->func->chan; + return 0; } return c; @@ -124,10 +151,12 @@ gk104_fifo_uevent_init(struct nvkm_fifo *fifo) void gk104_fifo_runlist_commit(struct gk104_fifo *fifo, int runl) { + const struct gk104_fifo_runlist_func *func = fifo->func->runlist; struct gk104_fifo_chan *chan; struct nvkm_subdev *subdev = &fifo->base.engine.subdev; struct nvkm_device *device = subdev->device; struct nvkm_memory *mem; + struct nvkm_fifo_cgrp *cgrp; int nr = 0; int target; @@ -137,9 +166,14 @@ gk104_fifo_runlist_commit(struct gk104_fifo *fifo, int runl) nvkm_kmap(mem); list_for_each_entry(chan, &fifo->runlist[runl].chan, head) { - nvkm_wo32(mem, (nr * 8) + 0, chan->base.chid); - nvkm_wo32(mem, (nr * 8) + 4, 0x00000000); - nr++; + func->chan(chan, mem, nr++ * func->size); + } + + list_for_each_entry(cgrp, &fifo->runlist[runl].cgrp, head) { + func->cgrp(cgrp, mem, nr++ * func->size); + list_for_each_entry(chan, &cgrp->chan, head) { + func->chan(chan, mem, nr++ * func->size); + } } nvkm_done(mem); @@ -155,10 +189,10 @@ gk104_fifo_runlist_commit(struct gk104_fifo *fifo, int runl) (target << 28)); nvkm_wr32(device, 0x002274, (runl << 20) | nr); - if (wait_event_timeout(fifo->runlist[runl].wait, - !(nvkm_rd32(device, 0x002284 + (runl * 0x08)) - & 0x00100000), - msecs_to_jiffies(2000)) == 0) + if (nvkm_msec(device, 2000, + if (!(nvkm_rd32(device, 0x002284 + (runl * 0x08)) & 0x00100000)) + break; + ) < 0) nvkm_error(subdev, "runlist %d update timeout\n", runl); unlock: mutex_unlock(&subdev->mutex); @@ -167,19 +201,45 @@ unlock: void gk104_fifo_runlist_remove(struct gk104_fifo *fifo, struct gk104_fifo_chan *chan) { + struct nvkm_fifo_cgrp *cgrp = chan->cgrp; mutex_lock(&fifo->base.engine.subdev.mutex); - list_del_init(&chan->head); + if (!list_empty(&chan->head)) { + list_del_init(&chan->head); + if (cgrp && !--cgrp->chan_nr) + list_del_init(&cgrp->head); + } mutex_unlock(&fifo->base.engine.subdev.mutex); } void gk104_fifo_runlist_insert(struct gk104_fifo *fifo, struct gk104_fifo_chan *chan) { + struct nvkm_fifo_cgrp *cgrp = chan->cgrp; mutex_lock(&fifo->base.engine.subdev.mutex); - list_add_tail(&chan->head, &fifo->runlist[chan->runl].chan); + if (cgrp) { + if (!cgrp->chan_nr++) + list_add_tail(&cgrp->head, &fifo->runlist[chan->runl].cgrp); + list_add_tail(&chan->head, &cgrp->chan); + } else { + list_add_tail(&chan->head, &fifo->runlist[chan->runl].chan); + } mutex_unlock(&fifo->base.engine.subdev.mutex); } +void +gk104_fifo_runlist_chan(struct gk104_fifo_chan *chan, + struct nvkm_memory *memory, u32 offset) +{ + nvkm_wo32(memory, offset + 0, chan->base.chid); + nvkm_wo32(memory, offset + 4, 0x00000000); +} + +const struct gk104_fifo_runlist_func +gk104_fifo_runlist = { + .size = 8, + .chan = gk104_fifo_runlist_chan, +}; + static void gk104_fifo_recover_work(struct work_struct *w) { @@ -235,6 +295,32 @@ gk104_fifo_recover_runl(struct gk104_fifo *fifo, int runl) schedule_work(&fifo->recover.work); } +static struct gk104_fifo_chan * +gk104_fifo_recover_chid(struct gk104_fifo *fifo, int runl, int chid) +{ + struct gk104_fifo_chan *chan; + struct nvkm_fifo_cgrp *cgrp; + + list_for_each_entry(chan, &fifo->runlist[runl].chan, head) { + if (chan->base.chid == chid) { + list_del_init(&chan->head); + return chan; + } + } + + list_for_each_entry(cgrp, &fifo->runlist[runl].cgrp, head) { + if (cgrp->id == chid) { + chan = list_first_entry(&cgrp->chan, typeof(*chan), head); + list_del_init(&chan->head); + if (!--cgrp->chan_nr) + list_del_init(&cgrp->head); + return chan; + } + } + + return NULL; +} + static void gk104_fifo_recover_chan(struct nvkm_fifo *base, int chid) { @@ -252,13 +338,10 @@ gk104_fifo_recover_chan(struct nvkm_fifo *base, int chid) return; /* Lookup SW state for channel, and mark it as dead. */ - list_for_each_entry(chan, &fifo->runlist[runl].chan, head) { - if (chan->base.chid == chid) { - list_del_init(&chan->head); - chan->killed = true; - nvkm_fifo_kevent(&fifo->base, chid); - break; - } + chan = gk104_fifo_recover_chid(fifo, runl, chid); + if (chan) { + chan->killed = true; + nvkm_fifo_kevent(&fifo->base, chid); } /* Disable channel. */ @@ -347,6 +430,90 @@ gk104_fifo_recover_engn(struct gk104_fifo *fifo, int engn) schedule_work(&fifo->recover.work); } +static void +gk104_fifo_fault(struct nvkm_fifo *base, struct nvkm_fault_data *info) +{ + struct gk104_fifo *fifo = gk104_fifo(base); + struct nvkm_subdev *subdev = &fifo->base.engine.subdev; + struct nvkm_device *device = subdev->device; + const struct nvkm_enum *er, *ee, *ec, *ea; + struct nvkm_engine *engine = NULL; + struct nvkm_fifo_chan *chan; + unsigned long flags; + char ct[8] = "HUB/", en[16] = ""; + int engn; + + er = nvkm_enum_find(fifo->func->fault.reason, info->reason); + ee = nvkm_enum_find(fifo->func->fault.engine, info->engine); + if (info->hub) { + ec = nvkm_enum_find(fifo->func->fault.hubclient, info->client); + } else { + ec = nvkm_enum_find(fifo->func->fault.gpcclient, info->client); + snprintf(ct, sizeof(ct), "GPC%d/", info->gpc); + } + ea = nvkm_enum_find(fifo->func->fault.access, info->access); + + if (ee && ee->data2) { + switch (ee->data2) { + case NVKM_SUBDEV_BAR: + nvkm_mask(device, 0x001704, 0x00000000, 0x00000000); + break; + case NVKM_SUBDEV_INSTMEM: + nvkm_mask(device, 0x001714, 0x00000000, 0x00000000); + break; + case NVKM_ENGINE_IFB: + nvkm_mask(device, 0x001718, 0x00000000, 0x00000000); + break; + default: + engine = nvkm_device_engine(device, ee->data2); + break; + } + } + + if (ee == NULL) { + enum nvkm_devidx engidx = nvkm_top_fault(device, info->engine); + if (engidx < NVKM_SUBDEV_NR) { + const char *src = nvkm_subdev_name[engidx]; + char *dst = en; + do { + *dst++ = toupper(*src++); + } while(*src); + engine = nvkm_device_engine(device, engidx); + } + } else { + snprintf(en, sizeof(en), "%s", ee->name); + } + + spin_lock_irqsave(&fifo->base.lock, flags); + chan = nvkm_fifo_chan_inst_locked(&fifo->base, info->inst); + + nvkm_error(subdev, + "fault %02x [%s] at %016llx engine %02x [%s] client %02x " + "[%s%s] reason %02x [%s] on channel %d [%010llx %s]\n", + info->access, ea ? ea->name : "", info->addr, + info->engine, ee ? ee->name : en, + info->client, ct, ec ? ec->name : "", + info->reason, er ? er->name : "", chan ? chan->chid : -1, + info->inst, chan ? chan->object.client->name : "unknown"); + + /* Kill the channel that caused the fault. */ + if (chan) + gk104_fifo_recover_chan(&fifo->base, chan->chid); + + /* Channel recovery will probably have already done this for the + * correct engine(s), but just in case we can't find the channel + * information... + */ + for (engn = 0; engn < fifo->engine_nr && engine; engn++) { + if (fifo->engine[engn].engine == engine) { + gk104_fifo_recover_engn(fifo, engn); + break; + } + } + + spin_unlock_irqrestore(&fifo->base.lock, flags); +} + static const struct nvkm_enum gk104_fifo_bind_reason[] = { { 0x01, "BIND_NOT_UNBOUND" }, @@ -456,88 +623,21 @@ gk104_fifo_intr_fault(struct gk104_fifo *fifo, int unit) u32 inst = nvkm_rd32(device, 0x002800 + (unit * 0x10)); u32 valo = nvkm_rd32(device, 0x002804 + (unit * 0x10)); u32 vahi = nvkm_rd32(device, 0x002808 + (unit * 0x10)); - u32 stat = nvkm_rd32(device, 0x00280c + (unit * 0x10)); - u32 gpc = (stat & 0x1f000000) >> 24; - u32 client = (stat & 0x00001f00) >> 8; - u32 write = (stat & 0x00000080); - u32 hub = (stat & 0x00000040); - u32 reason = (stat & 0x0000000f); - const struct nvkm_enum *er, *eu, *ec; - struct nvkm_engine *engine = NULL; - struct nvkm_fifo_chan *chan; - unsigned long flags; - char gpcid[8] = "", en[16] = ""; - int engn; - - er = nvkm_enum_find(fifo->func->fault.reason, reason); - eu = nvkm_enum_find(fifo->func->fault.engine, unit); - if (hub) { - ec = nvkm_enum_find(fifo->func->fault.hubclient, client); - } else { - ec = nvkm_enum_find(fifo->func->fault.gpcclient, client); - snprintf(gpcid, sizeof(gpcid), "GPC%d/", gpc); - } - - if (eu && eu->data2) { - switch (eu->data2) { - case NVKM_SUBDEV_BAR: - nvkm_mask(device, 0x001704, 0x00000000, 0x00000000); - break; - case NVKM_SUBDEV_INSTMEM: - nvkm_mask(device, 0x001714, 0x00000000, 0x00000000); - break; - case NVKM_ENGINE_IFB: - nvkm_mask(device, 0x001718, 0x00000000, 0x00000000); - break; - default: - engine = nvkm_device_engine(device, eu->data2); - break; - } - } - - if (eu == NULL) { - enum nvkm_devidx engidx = nvkm_top_fault(device, unit); - if (engidx < NVKM_SUBDEV_NR) { - const char *src = nvkm_subdev_name[engidx]; - char *dst = en; - do { - *dst++ = toupper(*src++); - } while(*src); - engine = nvkm_device_engine(device, engidx); - } - } else { - snprintf(en, sizeof(en), "%s", eu->name); - } - - spin_lock_irqsave(&fifo->base.lock, flags); - chan = nvkm_fifo_chan_inst_locked(&fifo->base, (u64)inst << 12); - - nvkm_error(subdev, - "%s fault at %010llx engine %02x [%s] client %02x [%s%s] " - "reason %02x [%s] on channel %d [%010llx %s]\n", - write ? "write" : "read", (u64)vahi << 32 | valo, - unit, en, client, gpcid, ec ? ec->name : "", - reason, er ? er->name : "", chan ? chan->chid : -1, - (u64)inst << 12, - chan ? chan->object.client->name : "unknown"); - - - /* Kill the channel that caused the fault. */ - if (chan) - gk104_fifo_recover_chan(&fifo->base, chan->chid); - - /* Channel recovery will probably have already done this for the - * correct engine(s), but just in case we can't find the channel - * information... - */ - for (engn = 0; engn < fifo->engine_nr && engine; engn++) { - if (fifo->engine[engn].engine == engine) { - gk104_fifo_recover_engn(fifo, engn); - break; - } - } - - spin_unlock_irqrestore(&fifo->base.lock, flags); + u32 type = nvkm_rd32(device, 0x00280c + (unit * 0x10)); + struct nvkm_fault_data info; + + info.inst = (u64)inst << 12; + info.addr = ((u64)vahi << 32) | valo; + info.time = 0; + info.engine = unit; + info.valid = 1; + info.gpc = (type & 0x1f000000) >> 24; + info.client = (type & 0x00001f00) >> 8; + info.access = (type & 0x00000080) >> 7; + info.hub = (type & 0x00000040) >> 6; + info.reason = (type & 0x000000ff); + + nvkm_fifo_fault(&fifo->base, &info); } static const struct nvkm_bitfield gk104_fifo_pbdma_intr_0[] = { @@ -766,6 +866,34 @@ gk104_fifo_fini(struct nvkm_fifo *base) } static int +gk104_fifo_info(struct nvkm_fifo *base, u64 mthd, u64 *data) +{ + struct gk104_fifo *fifo = gk104_fifo(base); + switch (mthd) { + case NV_DEVICE_FIFO_RUNLISTS: + *data = (1ULL << fifo->runlist_nr) - 1; + return 0; + case NV_DEVICE_FIFO_RUNLIST_ENGINES(0)... + NV_DEVICE_FIFO_RUNLIST_ENGINES(63): { + int runl = mthd - NV_DEVICE_FIFO_RUNLIST_ENGINES(0), engn; + if (runl < fifo->runlist_nr) { + unsigned long engm = fifo->runlist[runl].engm; + struct nvkm_engine *engine; + *data = 0; + for_each_set_bit(engn, &engm, fifo->engine_nr) { + if ((engine = fifo->engine[engn].engine)) + *data |= BIT_ULL(engine->subdev.index); + } + return 0; + } + } + return -EINVAL; + default: + return -EINVAL; + } +} + +static int gk104_fifo_oneinit(struct nvkm_fifo *base) { struct gk104_fifo *fifo = gk104_fifo(base); @@ -813,19 +941,18 @@ gk104_fifo_oneinit(struct nvkm_fifo *base) kfree(map); for (i = 0; i < fifo->runlist_nr; i++) { - ret = nvkm_memory_new(device, NVKM_MEM_TARGET_INST, - 0x8000, 0x1000, false, - &fifo->runlist[i].mem[0]); - if (ret) - return ret; - - ret = nvkm_memory_new(device, NVKM_MEM_TARGET_INST, - 0x8000, 0x1000, false, - &fifo->runlist[i].mem[1]); - if (ret) - return ret; + for (j = 0; j < ARRAY_SIZE(fifo->runlist[i].mem); j++) { + ret = nvkm_memory_new(device, NVKM_MEM_TARGET_INST, + fifo->base.nr * 2/* TSG+chan */ * + fifo->func->runlist->size, + 0x1000, false, + &fifo->runlist[i].mem[j]); + if (ret) + return ret; + } init_waitqueue_head(&fifo->runlist[i].wait); + INIT_LIST_HEAD(&fifo->runlist[i].cgrp); INIT_LIST_HEAD(&fifo->runlist[i].chan); } @@ -868,6 +995,9 @@ gk104_fifo_init(struct nvkm_fifo *base) nvkm_wr32(device, 0x002254, 0x10000000 | fifo->user.bar->addr >> 12); + if (fifo->func->init_pbdma_timeout) + fifo->func->init_pbdma_timeout(fifo); + nvkm_wr32(device, 0x002100, 0xffffffff); nvkm_wr32(device, 0x002140, 0x7fffffff); } @@ -894,13 +1024,16 @@ static const struct nvkm_fifo_func gk104_fifo_ = { .dtor = gk104_fifo_dtor, .oneinit = gk104_fifo_oneinit, + .info = gk104_fifo_info, .init = gk104_fifo_init, .fini = gk104_fifo_fini, .intr = gk104_fifo_intr, + .fault = gk104_fifo_fault, .uevent_init = gk104_fifo_uevent_init, .uevent_fini = gk104_fifo_uevent_fini, .recover_chan = gk104_fifo_recover_chan, .class_get = gk104_fifo_class_get, + .class_new = gk104_fifo_class_new, }; int @@ -919,6 +1052,13 @@ gk104_fifo_new_(const struct gk104_fifo_func *func, struct nvkm_device *device, } const struct nvkm_enum +gk104_fifo_fault_access[] = { + { 0x0, "READ" }, + { 0x1, "WRITE" }, + {} +}; + +const struct nvkm_enum gk104_fifo_fault_engine[] = { { 0x00, "GR", NULL, NVKM_ENGINE_GR }, { 0x01, "DISPLAY" }, @@ -1035,14 +1175,13 @@ gk104_fifo_fault_gpcclient[] = { static const struct gk104_fifo_func gk104_fifo = { + .fault.access = gk104_fifo_fault_access, .fault.engine = gk104_fifo_fault_engine, .fault.reason = gk104_fifo_fault_reason, .fault.hubclient = gk104_fifo_fault_hubclient, .fault.gpcclient = gk104_fifo_fault_gpcclient, - .chan = { - &gk104_fifo_gpfifo_oclass, - NULL - }, + .runlist = &gk104_fifo_runlist, + .chan = {{0,0,KEPLER_CHANNEL_GPFIFO_A}, gk104_fifo_gpfifo_new }, }; int diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk104.h b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk104.h index 1579785cf941..d295b81e18d6 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk104.h +++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk104.h @@ -3,6 +3,7 @@ #define __GK104_FIFO_H__ #define gk104_fifo(p) container_of((p), struct gk104_fifo, base) #include "priv.h" +struct nvkm_fifo_cgrp; #include <core/enum.h> #include <subdev/mmu.h> @@ -31,6 +32,7 @@ struct gk104_fifo { struct nvkm_memory *mem[2]; int next; wait_queue_head_t wait; + struct list_head cgrp; struct list_head chan; u32 engm; } runlist[16]; @@ -43,14 +45,36 @@ struct gk104_fifo { }; struct gk104_fifo_func { + void (*init_pbdma_timeout)(struct gk104_fifo *); + struct { + const struct nvkm_enum *access; const struct nvkm_enum *engine; const struct nvkm_enum *reason; const struct nvkm_enum *hubclient; const struct nvkm_enum *gpcclient; } fault; - const struct nvkm_fifo_chan_oclass *chan[]; + const struct gk104_fifo_runlist_func { + u8 size; + void (*cgrp)(struct nvkm_fifo_cgrp *, + struct nvkm_memory *, u32 offset); + void (*chan)(struct gk104_fifo_chan *, + struct nvkm_memory *, u32 offset); + } *runlist; + + struct gk104_fifo_user_user { + struct nvkm_sclass user; + int (*ctor)(const struct nvkm_oclass *, void *, u32, + struct nvkm_object **); + } user; + + struct gk104_fifo_chan_user { + struct nvkm_sclass user; + int (*ctor)(struct gk104_fifo *, const struct nvkm_oclass *, + void *, u32, struct nvkm_object **); + } chan; + bool cgrp_force; }; int gk104_fifo_new_(const struct gk104_fifo_func *, struct nvkm_device *, @@ -59,30 +83,23 @@ void gk104_fifo_runlist_insert(struct gk104_fifo *, struct gk104_fifo_chan *); void gk104_fifo_runlist_remove(struct gk104_fifo *, struct gk104_fifo_chan *); void gk104_fifo_runlist_commit(struct gk104_fifo *, int runl); -static inline u64 -gk104_fifo_engine_subdev(int engine) -{ - switch (engine) { - case 0: return (1ULL << NVKM_ENGINE_GR) | - (1ULL << NVKM_ENGINE_SW) | - (1ULL << NVKM_ENGINE_CE2); - case 1: return (1ULL << NVKM_ENGINE_MSPDEC); - case 2: return (1ULL << NVKM_ENGINE_MSPPP); - case 3: return (1ULL << NVKM_ENGINE_MSVLD); - case 4: return (1ULL << NVKM_ENGINE_CE0); - case 5: return (1ULL << NVKM_ENGINE_CE1); - case 6: return (1ULL << NVKM_ENGINE_MSENC); - default: - WARN_ON(1); - return 0; - } -} - +extern const struct nvkm_enum gk104_fifo_fault_access[]; extern const struct nvkm_enum gk104_fifo_fault_engine[]; extern const struct nvkm_enum gk104_fifo_fault_reason[]; extern const struct nvkm_enum gk104_fifo_fault_hubclient[]; extern const struct nvkm_enum gk104_fifo_fault_gpcclient[]; +extern const struct gk104_fifo_runlist_func gk104_fifo_runlist; +void gk104_fifo_runlist_chan(struct gk104_fifo_chan *, + struct nvkm_memory *, u32); + +extern const struct gk104_fifo_runlist_func gk110_fifo_runlist; +void gk110_fifo_runlist_cgrp(struct nvkm_fifo_cgrp *, + struct nvkm_memory *, u32); + +void gk208_fifo_init_pbdma_timeout(struct gk104_fifo *); extern const struct nvkm_enum gm107_fifo_fault_engine[]; +extern const struct gk104_fifo_runlist_func gm107_fifo_runlist; + extern const struct nvkm_enum gp100_fifo_fault_engine[]; #endif diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk110.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk110.c index b2f8ab7bf847..ac7655a130fb 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk110.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk110.c @@ -22,18 +22,38 @@ * Authors: Ben Skeggs */ #include "gk104.h" +#include "cgrp.h" #include "changk104.h" +#include <core/memory.h> + +#include <nvif/class.h> + +void +gk110_fifo_runlist_cgrp(struct nvkm_fifo_cgrp *cgrp, + struct nvkm_memory *memory, u32 offset) +{ + nvkm_wo32(memory, offset + 0, (cgrp->chan_nr << 26) | (128 << 18) | + (3 << 14) | 0x00002000 | cgrp->id); + nvkm_wo32(memory, offset + 4, 0x00000000); +} + +const struct gk104_fifo_runlist_func +gk110_fifo_runlist = { + .size = 8, + .cgrp = gk110_fifo_runlist_cgrp, + .chan = gk104_fifo_runlist_chan, +}; + static const struct gk104_fifo_func gk110_fifo = { + .fault.access = gk104_fifo_fault_access, .fault.engine = gk104_fifo_fault_engine, .fault.reason = gk104_fifo_fault_reason, .fault.hubclient = gk104_fifo_fault_hubclient, .fault.gpcclient = gk104_fifo_fault_gpcclient, - .chan = { - &gk110_fifo_gpfifo_oclass, - NULL - }, + .runlist = &gk110_fifo_runlist, + .chan = {{0,0,KEPLER_CHANNEL_GPFIFO_B}, gk104_fifo_gpfifo_new }, }; int diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk208.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk208.c index 160617d376e4..5ea7e452cc66 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk208.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk208.c @@ -24,16 +24,28 @@ #include "gk104.h" #include "changk104.h" +#include <nvif/class.h> + +void +gk208_fifo_init_pbdma_timeout(struct gk104_fifo *fifo) +{ + struct nvkm_device *device = fifo->base.engine.subdev.device; + int i; + + for (i = 0; i < fifo->pbdma_nr; i++) + nvkm_wr32(device, 0x04012c + (i * 0x2000), 0x0000ffff); +} + static const struct gk104_fifo_func gk208_fifo = { + .init_pbdma_timeout = gk208_fifo_init_pbdma_timeout, + .fault.access = gk104_fifo_fault_access, .fault.engine = gk104_fifo_fault_engine, .fault.reason = gk104_fifo_fault_reason, .fault.hubclient = gk104_fifo_fault_hubclient, .fault.gpcclient = gk104_fifo_fault_gpcclient, - .chan = { - &gk104_fifo_gpfifo_oclass, - NULL - }, + .runlist = &gk110_fifo_runlist, + .chan = {{0,0,KEPLER_CHANNEL_GPFIFO_A}, gk104_fifo_gpfifo_new }, }; int diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk20a.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk20a.c index be9f5c16ed7d..535a0eb67a5f 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk20a.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk20a.c @@ -22,16 +22,18 @@ #include "gk104.h" #include "changk104.h" +#include <nvif/class.h> + static const struct gk104_fifo_func gk20a_fifo = { + .init_pbdma_timeout = gk208_fifo_init_pbdma_timeout, + .fault.access = gk104_fifo_fault_access, .fault.engine = gk104_fifo_fault_engine, .fault.reason = gk104_fifo_fault_reason, .fault.hubclient = gk104_fifo_fault_hubclient, .fault.gpcclient = gk104_fifo_fault_gpcclient, - .chan = { - &gk104_fifo_gpfifo_oclass, - NULL - }, + .runlist = &gk110_fifo_runlist, + .chan = {{0,0,KEPLER_CHANNEL_GPFIFO_A}, gk104_fifo_gpfifo_new }, }; int diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gm107.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gm107.c index 29c080683b32..79ae19b1db67 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gm107.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gm107.c @@ -24,6 +24,25 @@ #include "gk104.h" #include "changk104.h" +#include <core/gpuobj.h> + +#include <nvif/class.h> + +static void +gm107_fifo_runlist_chan(struct gk104_fifo_chan *chan, + struct nvkm_memory *memory, u32 offset) +{ + nvkm_wo32(memory, offset + 0, chan->base.chid); + nvkm_wo32(memory, offset + 4, chan->base.inst->addr >> 12); +} + +const struct gk104_fifo_runlist_func +gm107_fifo_runlist = { + .size = 8, + .cgrp = gk110_fifo_runlist_cgrp, + .chan = gm107_fifo_runlist_chan, +}; + const struct nvkm_enum gm107_fifo_fault_engine[] = { { 0x01, "DISPLAY" }, @@ -49,14 +68,14 @@ gm107_fifo_fault_engine[] = { static const struct gk104_fifo_func gm107_fifo = { + .init_pbdma_timeout = gk208_fifo_init_pbdma_timeout, + .fault.access = gk104_fifo_fault_access, .fault.engine = gm107_fifo_fault_engine, .fault.reason = gk104_fifo_fault_reason, .fault.hubclient = gk104_fifo_fault_hubclient, .fault.gpcclient = gk104_fifo_fault_gpcclient, - .chan = { - &gk110_fifo_gpfifo_oclass, - NULL - }, + .runlist = &gm107_fifo_runlist, + .chan = {{0,0,KEPLER_CHANNEL_GPFIFO_B}, gk104_fifo_gpfifo_new }, }; int diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gm200.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gm200.c index b069f785c5d8..49565faa854d 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gm200.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gm200.c @@ -24,16 +24,18 @@ #include "gk104.h" #include "changk104.h" +#include <nvif/class.h> + static const struct gk104_fifo_func gm200_fifo = { + .init_pbdma_timeout = gk208_fifo_init_pbdma_timeout, + .fault.access = gk104_fifo_fault_access, .fault.engine = gm107_fifo_fault_engine, .fault.reason = gk104_fifo_fault_reason, .fault.hubclient = gk104_fifo_fault_hubclient, .fault.gpcclient = gk104_fifo_fault_gpcclient, - .chan = { - &gm200_fifo_gpfifo_oclass, - NULL - }, + .runlist = &gm107_fifo_runlist, + .chan = {{0,0,MAXWELL_CHANNEL_GPFIFO_A}, gk104_fifo_gpfifo_new }, }; int diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gm20b.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gm20b.c index 2ed87c2e8299..46736513bd11 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gm20b.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gm20b.c @@ -22,16 +22,18 @@ #include "gk104.h" #include "changk104.h" +#include <nvif/class.h> + static const struct gk104_fifo_func gm20b_fifo = { + .init_pbdma_timeout = gk208_fifo_init_pbdma_timeout, + .fault.access = gk104_fifo_fault_access, .fault.engine = gm107_fifo_fault_engine, .fault.reason = gk104_fifo_fault_reason, .fault.hubclient = gk104_fifo_fault_hubclient, .fault.gpcclient = gk104_fifo_fault_gpcclient, - .chan = { - &gm200_fifo_gpfifo_oclass, - NULL - }, + .runlist = &gm107_fifo_runlist, + .chan = {{0,0,MAXWELL_CHANNEL_GPFIFO_A}, gk104_fifo_gpfifo_new }, }; int diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gp100.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gp100.c index 41f16cf5a918..e2f8f9087d7c 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gp100.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gp100.c @@ -24,6 +24,8 @@ #include "gk104.h" #include "changk104.h" +#include <nvif/class.h> + const struct nvkm_enum gp100_fifo_fault_engine[] = { { 0x01, "DISPLAY" }, @@ -50,14 +52,15 @@ gp100_fifo_fault_engine[] = { static const struct gk104_fifo_func gp100_fifo = { + .init_pbdma_timeout = gk208_fifo_init_pbdma_timeout, + .fault.access = gk104_fifo_fault_access, .fault.engine = gp100_fifo_fault_engine, .fault.reason = gk104_fifo_fault_reason, .fault.hubclient = gk104_fifo_fault_hubclient, .fault.gpcclient = gk104_fifo_fault_gpcclient, - .chan = { - &gp100_fifo_gpfifo_oclass, - NULL - }, + .runlist = &gm107_fifo_runlist, + .chan = {{0,0,PASCAL_CHANNEL_GPFIFO_A}, gk104_fifo_gpfifo_new }, + .cgrp_force = true, }; int diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gp10b.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gp10b.c index 4af96c3e69ff..7733bf7c6545 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gp10b.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gp10b.c @@ -22,16 +22,19 @@ #include "gk104.h" #include "changk104.h" +#include <nvif/class.h> + static const struct gk104_fifo_func gp10b_fifo = { + .init_pbdma_timeout = gk208_fifo_init_pbdma_timeout, + .fault.access = gk104_fifo_fault_access, .fault.engine = gp100_fifo_fault_engine, .fault.reason = gk104_fifo_fault_reason, .fault.hubclient = gk104_fifo_fault_hubclient, .fault.gpcclient = gk104_fifo_fault_gpcclient, - .chan = { - &gp100_fifo_gpfifo_oclass, - NULL - }, + .runlist = &gm107_fifo_runlist, + .chan = {{0,0,PASCAL_CHANNEL_GPFIFO_A}, gk104_fifo_gpfifo_new }, + .cgrp_force = true, }; int diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gpfifogk104.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gpfifogk104.c index 80c87521bebe..118b37aea318 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gpfifogk104.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gpfifogk104.c @@ -22,6 +22,7 @@ * Authors: Ben Skeggs */ #include "changk104.h" +#include "cgrp.h" #include <core/client.h> #include <core/gpuobj.h> @@ -33,27 +34,40 @@ #include <nvif/cla06f.h> #include <nvif/unpack.h> -static int -gk104_fifo_gpfifo_kick(struct gk104_fifo_chan *chan) +int +gk104_fifo_gpfifo_kick_locked(struct gk104_fifo_chan *chan) { struct gk104_fifo *fifo = chan->fifo; struct nvkm_subdev *subdev = &fifo->base.engine.subdev; struct nvkm_device *device = subdev->device; struct nvkm_client *client = chan->base.object.client; + struct nvkm_fifo_cgrp *cgrp = chan->cgrp; int ret = 0; - mutex_lock(&subdev->mutex); - nvkm_wr32(device, 0x002634, chan->base.chid); + if (cgrp) + nvkm_wr32(device, 0x002634, cgrp->id | 0x01000000); + else + nvkm_wr32(device, 0x002634, chan->base.chid); if (nvkm_msec(device, 2000, if (!(nvkm_rd32(device, 0x002634) & 0x00100000)) break; ) < 0) { - nvkm_error(subdev, "channel %d [%s] kick timeout\n", - chan->base.chid, client->name); + nvkm_error(subdev, "%s %d [%s] kick timeout\n", + cgrp ? "tsg" : "channel", + cgrp ? cgrp->id : chan->base.chid, client->name); nvkm_fifo_recover_chan(&fifo->base, chan->base.chid); ret = -ETIMEDOUT; } - mutex_unlock(&subdev->mutex); + return ret; +} + +int +gk104_fifo_gpfifo_kick(struct gk104_fifo_chan *chan) +{ + int ret; + mutex_lock(&chan->base.fifo->engine.subdev.mutex); + ret = gk104_fifo_gpfifo_kick_locked(chan); + mutex_unlock(&chan->base.fifo->engine.subdev.mutex); return ret; } @@ -62,9 +76,8 @@ gk104_fifo_gpfifo_engine_addr(struct nvkm_engine *engine) { switch (engine->subdev.index) { case NVKM_ENGINE_SW : - case NVKM_ENGINE_CE0 : - case NVKM_ENGINE_CE1 : - case NVKM_ENGINE_CE2 : return 0x0000; + case NVKM_ENGINE_CE0...NVKM_ENGINE_CE_LAST: + return 0; case NVKM_ENGINE_GR : return 0x0210; case NVKM_ENGINE_SEC : return 0x0220; case NVKM_ENGINE_MSPDEC: return 0x0250; @@ -133,7 +146,7 @@ gk104_fifo_gpfifo_engine_init(struct nvkm_fifo_chan *base, return 0; } -static void +void gk104_fifo_gpfifo_engine_dtor(struct nvkm_fifo_chan *base, struct nvkm_engine *engine) { @@ -142,7 +155,7 @@ gk104_fifo_gpfifo_engine_dtor(struct nvkm_fifo_chan *base, nvkm_gpuobj_del(&chan->engn[engine->subdev.index].inst); } -static int +int gk104_fifo_gpfifo_engine_ctor(struct nvkm_fifo_chan *base, struct nvkm_engine *engine, struct nvkm_object *object) @@ -167,7 +180,7 @@ gk104_fifo_gpfifo_engine_ctor(struct nvkm_fifo_chan *base, chan->engn[engn].vma, NULL, 0); } -static void +void gk104_fifo_gpfifo_fini(struct nvkm_fifo_chan *base) { struct gk104_fifo_chan *chan = gk104_fifo_chan(base); @@ -185,7 +198,7 @@ gk104_fifo_gpfifo_fini(struct nvkm_fifo_chan *base) nvkm_wr32(device, 0x800000 + coff, 0x00000000); } -static void +void gk104_fifo_gpfifo_init(struct nvkm_fifo_chan *base) { struct gk104_fifo_chan *chan = gk104_fifo_chan(base); @@ -205,13 +218,15 @@ gk104_fifo_gpfifo_init(struct nvkm_fifo_chan *base) } } -static void * +void * gk104_fifo_gpfifo_dtor(struct nvkm_fifo_chan *base) { - return gk104_fifo_chan(base); + struct gk104_fifo_chan *chan = gk104_fifo_chan(base); + kfree(chan->cgrp); + return chan; } -static const struct nvkm_fifo_chan_func +const struct nvkm_fifo_chan_func gk104_fifo_gpfifo_func = { .dtor = gk104_fifo_gpfifo_dtor, .init = gk104_fifo_gpfifo_init, @@ -223,62 +238,30 @@ gk104_fifo_gpfifo_func = { .engine_fini = gk104_fifo_gpfifo_engine_fini, }; -struct gk104_fifo_chan_func { - u32 engine; - u64 subdev; -}; - static int -gk104_fifo_gpfifo_new_(const struct gk104_fifo_chan_func *func, - struct gk104_fifo *fifo, u32 *engmask, u16 *chid, +gk104_fifo_gpfifo_new_(struct gk104_fifo *fifo, u64 *runlists, u16 *chid, u64 vmm, u64 ioffset, u64 ilength, const struct nvkm_oclass *oclass, struct nvkm_object **pobject) { struct gk104_fifo_chan *chan; - int runlist = -1, ret = -ENOSYS, i, j; - u32 engines = 0, present = 0; + int runlist = ffs(*runlists) -1, ret, i; + unsigned long engm; u64 subdevs = 0; u64 usermem; - if (!vmm) + if (!vmm || runlist < 0 || runlist >= fifo->runlist_nr) return -EINVAL; + *runlists = BIT_ULL(runlist); - /* Determine which downstream engines are present */ - for (i = 0; i < fifo->engine_nr; i++) { - struct nvkm_engine *engine = fifo->engine[i].engine; - if (engine) { - u64 submask = BIT_ULL(engine->subdev.index); - for (j = 0; func[j].subdev; j++) { - if (func[j].subdev & submask) { - present |= func[j].engine; - break; - } - } - - if (!func[j].subdev) - continue; - - if (runlist < 0 && (*engmask & present)) - runlist = fifo->engine[i].runl; - if (runlist == fifo->engine[i].runl) { - engines |= func[j].engine; - subdevs |= func[j].subdev; - } - } - } - - /* Just an engine mask query? All done here! */ - if (!*engmask) { - *engmask = present; - return nvkm_object_new(oclass, NULL, 0, pobject); + engm = fifo->runlist[runlist].engm; + for_each_set_bit(i, &engm, fifo->engine_nr) { + if (fifo->engine[i].engine) + subdevs |= BIT_ULL(fifo->engine[i].engine->subdev.index); } - /* No runlist? No supported engines. */ - *engmask = present; - if (runlist < 0) - return -ENODEV; - *engmask = engines; + if (subdevs & BIT_ULL(NVKM_ENGINE_GR)) + subdevs |= BIT_ULL(NVKM_ENGINE_SW); /* Allocate the channel. */ if (!(chan = kzalloc(sizeof(*chan), GFP_KERNEL))) @@ -297,6 +280,18 @@ gk104_fifo_gpfifo_new_(const struct gk104_fifo_chan_func *func, *chid = chan->base.chid; + /* Hack to support GPUs where even individual channels should be + * part of a channel group. + */ + if (fifo->func->cgrp_force) { + if (!(chan->cgrp = kmalloc(sizeof(*chan->cgrp), GFP_KERNEL))) + return -ENOMEM; + chan->cgrp->id = chan->base.chid; + INIT_LIST_HEAD(&chan->cgrp->head); + INIT_LIST_HEAD(&chan->cgrp->chan); + chan->cgrp->chan_nr = 0; + } + /* Clear channel control registers. */ usermem = chan->base.chid * 0x200; ilength = order_base_2(ilength / 8); @@ -328,45 +323,25 @@ gk104_fifo_gpfifo_new_(const struct gk104_fifo_chan_func *func, return 0; } -static const struct gk104_fifo_chan_func -gk104_fifo_gpfifo[] = { - { NVA06F_V0_ENGINE_SW | NVA06F_V0_ENGINE_GR, - BIT_ULL(NVKM_ENGINE_SW) | BIT_ULL(NVKM_ENGINE_GR) - }, - { NVA06F_V0_ENGINE_SEC , BIT_ULL(NVKM_ENGINE_SEC ) }, - { NVA06F_V0_ENGINE_MSVLD , BIT_ULL(NVKM_ENGINE_MSVLD ) }, - { NVA06F_V0_ENGINE_MSPDEC, BIT_ULL(NVKM_ENGINE_MSPDEC) }, - { NVA06F_V0_ENGINE_MSPPP , BIT_ULL(NVKM_ENGINE_MSPPP ) }, - { NVA06F_V0_ENGINE_MSENC , BIT_ULL(NVKM_ENGINE_MSENC ) }, - { NVA06F_V0_ENGINE_VIC , BIT_ULL(NVKM_ENGINE_VIC ) }, - { NVA06F_V0_ENGINE_NVDEC , BIT_ULL(NVKM_ENGINE_NVDEC ) }, - { NVA06F_V0_ENGINE_NVENC0, BIT_ULL(NVKM_ENGINE_NVENC0) }, - { NVA06F_V0_ENGINE_NVENC1, BIT_ULL(NVKM_ENGINE_NVENC1) }, - { NVA06F_V0_ENGINE_CE0 , BIT_ULL(NVKM_ENGINE_CE0 ) }, - { NVA06F_V0_ENGINE_CE1 , BIT_ULL(NVKM_ENGINE_CE1 ) }, - { NVA06F_V0_ENGINE_CE2 , BIT_ULL(NVKM_ENGINE_CE2 ) }, - {} -}; - int -gk104_fifo_gpfifo_new(struct nvkm_fifo *base, const struct nvkm_oclass *oclass, +gk104_fifo_gpfifo_new(struct gk104_fifo *fifo, const struct nvkm_oclass *oclass, void *data, u32 size, struct nvkm_object **pobject) { struct nvkm_object *parent = oclass->parent; union { struct kepler_channel_gpfifo_a_v0 v0; } *args = data; - struct gk104_fifo *fifo = gk104_fifo(base); int ret = -ENOSYS; nvif_ioctl(parent, "create channel gpfifo size %d\n", size); if (!(ret = nvif_unpack(ret, &data, &size, args->v0, 0, 0, false))) { nvif_ioctl(parent, "create channel gpfifo vers %d vmm %llx " - "ioffset %016llx ilength %08x engine %08x\n", + "ioffset %016llx ilength %08x " + "runlist %016llx\n", args->v0.version, args->v0.vmm, args->v0.ioffset, - args->v0.ilength, args->v0.engines); - return gk104_fifo_gpfifo_new_(gk104_fifo_gpfifo, fifo, - &args->v0.engines, + args->v0.ilength, args->v0.runlist); + return gk104_fifo_gpfifo_new_(fifo, + &args->v0.runlist, &args->v0.chid, args->v0.vmm, args->v0.ioffset, @@ -376,11 +351,3 @@ gk104_fifo_gpfifo_new(struct nvkm_fifo *base, const struct nvkm_oclass *oclass, return ret; } - -const struct nvkm_fifo_chan_oclass -gk104_fifo_gpfifo_oclass = { - .base.oclass = KEPLER_CHANNEL_GPFIFO_A, - .base.minver = 0, - .base.maxver = 0, - .ctor = gk104_fifo_gpfifo_new, -}; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gpfifogv100.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gpfifogv100.c new file mode 100644 index 000000000000..9598853ced56 --- /dev/null +++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gpfifogv100.c @@ -0,0 +1,225 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "changk104.h" +#include "cgrp.h" + +#include <core/client.h> +#include <core/gpuobj.h> + +#include <nvif/cla06f.h> +#include <nvif/unpack.h> + +static int +gv100_fifo_gpfifo_engine_valid(struct gk104_fifo_chan *chan, bool ce, bool valid) +{ + struct nvkm_subdev *subdev = &chan->base.fifo->engine.subdev; + struct nvkm_device *device = subdev->device; + const u32 mask = ce ? 0x00020000 : 0x00010000; + const u32 data = valid ? mask : 0x00000000; + int ret; + + /* Block runlist to prevent the channel from being rescheduled. */ + mutex_lock(&subdev->mutex); + nvkm_mask(device, 0x002630, BIT(chan->runl), BIT(chan->runl)); + + /* Preempt the channel. */ + ret = gk104_fifo_gpfifo_kick_locked(chan); + if (ret == 0) { + /* Update engine context validity. */ + nvkm_kmap(chan->base.inst); + nvkm_mo32(chan->base.inst, 0x0ac, mask, data); + nvkm_done(chan->base.inst); + } + + /* Resume runlist. */ + nvkm_mask(device, 0x002630, BIT(chan->runl), 0); + mutex_unlock(&subdev->mutex); + return ret; +} + +static int +gv100_fifo_gpfifo_engine_fini(struct nvkm_fifo_chan *base, + struct nvkm_engine *engine, bool suspend) +{ + struct gk104_fifo_chan *chan = gk104_fifo_chan(base); + struct nvkm_gpuobj *inst = chan->base.inst; + int ret; + + if (engine->subdev.index >= NVKM_ENGINE_CE0 && + engine->subdev.index <= NVKM_ENGINE_CE_LAST) + return gk104_fifo_gpfifo_kick(chan); + + ret = gv100_fifo_gpfifo_engine_valid(chan, false, false); + if (ret && suspend) + return ret; + + nvkm_kmap(inst); + nvkm_wo32(inst, 0x0210, 0x00000000); + nvkm_wo32(inst, 0x0214, 0x00000000); + nvkm_done(inst); + return ret; +} + +static int +gv100_fifo_gpfifo_engine_init(struct nvkm_fifo_chan *base, + struct nvkm_engine *engine) +{ + struct gk104_fifo_chan *chan = gk104_fifo_chan(base); + struct nvkm_gpuobj *inst = chan->base.inst; + u64 addr; + + if (engine->subdev.index >= NVKM_ENGINE_CE0 && + engine->subdev.index <= NVKM_ENGINE_CE_LAST) + return 0; + + addr = chan->engn[engine->subdev.index].vma->addr; + nvkm_kmap(inst); + nvkm_wo32(inst, 0x210, lower_32_bits(addr) | 0x00000004); + nvkm_wo32(inst, 0x214, upper_32_bits(addr)); + nvkm_done(inst); + + return gv100_fifo_gpfifo_engine_valid(chan, false, true); +} + +const struct nvkm_fifo_chan_func +gv100_fifo_gpfifo_func = { + .dtor = gk104_fifo_gpfifo_dtor, + .init = gk104_fifo_gpfifo_init, + .fini = gk104_fifo_gpfifo_fini, + .ntfy = gf100_fifo_chan_ntfy, + .engine_ctor = gk104_fifo_gpfifo_engine_ctor, + .engine_dtor = gk104_fifo_gpfifo_engine_dtor, + .engine_init = gv100_fifo_gpfifo_engine_init, + .engine_fini = gv100_fifo_gpfifo_engine_fini, +}; + +static int +gv100_fifo_gpfifo_new_(struct gk104_fifo *fifo, u64 *runlists, u16 *chid, + u64 vmm, u64 ioffset, u64 ilength, + const struct nvkm_oclass *oclass, + struct nvkm_object **pobject) +{ + struct gk104_fifo_chan *chan; + int runlist = ffs(*runlists) -1, ret, i; + unsigned long engm; + u64 subdevs = 0; + u64 usermem; + + if (!vmm || runlist < 0 || runlist >= fifo->runlist_nr) + return -EINVAL; + *runlists = BIT_ULL(runlist); + + engm = fifo->runlist[runlist].engm; + for_each_set_bit(i, &engm, fifo->engine_nr) { + if (fifo->engine[i].engine) + subdevs |= BIT_ULL(fifo->engine[i].engine->subdev.index); + } + + /* Allocate the channel. */ + if (!(chan = kzalloc(sizeof(*chan), GFP_KERNEL))) + return -ENOMEM; + *pobject = &chan->base.object; + chan->fifo = fifo; + chan->runl = runlist; + INIT_LIST_HEAD(&chan->head); + + ret = nvkm_fifo_chan_ctor(&gv100_fifo_gpfifo_func, &fifo->base, + 0x1000, 0x1000, true, vmm, 0, subdevs, + 1, fifo->user.bar->addr, 0x200, + oclass, &chan->base); + if (ret) + return ret; + + *chid = chan->base.chid; + + /* Hack to support GPUs where even individual channels should be + * part of a channel group. + */ + if (fifo->func->cgrp_force) { + if (!(chan->cgrp = kmalloc(sizeof(*chan->cgrp), GFP_KERNEL))) + return -ENOMEM; + chan->cgrp->id = chan->base.chid; + INIT_LIST_HEAD(&chan->cgrp->head); + INIT_LIST_HEAD(&chan->cgrp->chan); + chan->cgrp->chan_nr = 0; + } + + /* Clear channel control registers. */ + usermem = chan->base.chid * 0x200; + ilength = order_base_2(ilength / 8); + + nvkm_kmap(fifo->user.mem); + for (i = 0; i < 0x200; i += 4) + nvkm_wo32(fifo->user.mem, usermem + i, 0x00000000); + nvkm_done(fifo->user.mem); + usermem = nvkm_memory_addr(fifo->user.mem) + usermem; + + /* RAMFC */ + nvkm_kmap(chan->base.inst); + nvkm_wo32(chan->base.inst, 0x008, lower_32_bits(usermem)); + nvkm_wo32(chan->base.inst, 0x00c, upper_32_bits(usermem)); + nvkm_wo32(chan->base.inst, 0x010, 0x0000face); + nvkm_wo32(chan->base.inst, 0x030, 0x7ffff902); + nvkm_wo32(chan->base.inst, 0x048, lower_32_bits(ioffset)); + nvkm_wo32(chan->base.inst, 0x04c, upper_32_bits(ioffset) | + (ilength << 16)); + nvkm_wo32(chan->base.inst, 0x084, 0x20400000); + nvkm_wo32(chan->base.inst, 0x094, 0x30000001); + nvkm_wo32(chan->base.inst, 0x0e4, 0x00000020); + nvkm_wo32(chan->base.inst, 0x0e8, chan->base.chid); + nvkm_wo32(chan->base.inst, 0x0f4, 0x00001100); + nvkm_wo32(chan->base.inst, 0x0f8, 0x10003080); + nvkm_mo32(chan->base.inst, 0x218, 0x00000000, 0x00000000); + nvkm_wo32(chan->base.inst, 0x220, 0x020a1000); + nvkm_wo32(chan->base.inst, 0x224, 0x00000000); + nvkm_done(chan->base.inst); + return gv100_fifo_gpfifo_engine_valid(chan, true, true); +} + +int +gv100_fifo_gpfifo_new(struct gk104_fifo *fifo, const struct nvkm_oclass *oclass, + void *data, u32 size, struct nvkm_object **pobject) +{ + struct nvkm_object *parent = oclass->parent; + union { + struct kepler_channel_gpfifo_a_v0 v0; + } *args = data; + int ret = -ENOSYS; + + nvif_ioctl(parent, "create channel gpfifo size %d\n", size); + if (!(ret = nvif_unpack(ret, &data, &size, args->v0, 0, 0, false))) { + nvif_ioctl(parent, "create channel gpfifo vers %d vmm %llx " + "ioffset %016llx ilength %08x " + "runlist %016llx\n", + args->v0.version, args->v0.vmm, args->v0.ioffset, + args->v0.ilength, args->v0.runlist); + return gv100_fifo_gpfifo_new_(fifo, + &args->v0.runlist, + &args->v0.chid, + args->v0.vmm, + args->v0.ioffset, + args->v0.ilength, + oclass, pobject); + } + + return ret; +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gv100.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gv100.c new file mode 100644 index 000000000000..4e1d159c0ae7 --- /dev/null +++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gv100.c @@ -0,0 +1,306 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "gk104.h" +#include "cgrp.h" +#include "changk104.h" +#include "user.h" + +#include <core/gpuobj.h> + +#include <nvif/class.h> + +static void +gv100_fifo_runlist_chan(struct gk104_fifo_chan *chan, + struct nvkm_memory *memory, u32 offset) +{ + struct nvkm_memory *usermem = chan->fifo->user.mem; + const u64 user = nvkm_memory_addr(usermem) + (chan->base.chid * 0x200); + const u64 inst = chan->base.inst->addr; + + nvkm_wo32(memory, offset + 0x0, lower_32_bits(user)); + nvkm_wo32(memory, offset + 0x4, upper_32_bits(user)); + nvkm_wo32(memory, offset + 0x8, lower_32_bits(inst) | chan->base.chid); + nvkm_wo32(memory, offset + 0xc, upper_32_bits(inst)); +} + +static void +gv100_fifo_runlist_cgrp(struct nvkm_fifo_cgrp *cgrp, + struct nvkm_memory *memory, u32 offset) +{ + nvkm_wo32(memory, offset + 0x0, (128 << 24) | (3 << 16) | 0x00000001); + nvkm_wo32(memory, offset + 0x4, cgrp->chan_nr); + nvkm_wo32(memory, offset + 0x8, cgrp->id); + nvkm_wo32(memory, offset + 0xc, 0x00000000); +} + +const struct gk104_fifo_runlist_func +gv100_fifo_runlist = { + .size = 16, + .cgrp = gv100_fifo_runlist_cgrp, + .chan = gv100_fifo_runlist_chan, +}; + +static const struct nvkm_enum +gv100_fifo_fault_gpcclient[] = { + { 0x00, "T1_0" }, + { 0x01, "T1_1" }, + { 0x02, "T1_2" }, + { 0x03, "T1_3" }, + { 0x04, "T1_4" }, + { 0x05, "T1_5" }, + { 0x06, "T1_6" }, + { 0x07, "T1_7" }, + { 0x08, "PE_0" }, + { 0x09, "PE_1" }, + { 0x0a, "PE_2" }, + { 0x0b, "PE_3" }, + { 0x0c, "PE_4" }, + { 0x0d, "PE_5" }, + { 0x0e, "PE_6" }, + { 0x0f, "PE_7" }, + { 0x10, "RAST" }, + { 0x11, "GCC" }, + { 0x12, "GPCCS" }, + { 0x13, "PROP_0" }, + { 0x14, "PROP_1" }, + { 0x15, "PROP_2" }, + { 0x16, "PROP_3" }, + { 0x17, "GPM" }, + { 0x18, "LTP_UTLB_0" }, + { 0x19, "LTP_UTLB_1" }, + { 0x1a, "LTP_UTLB_2" }, + { 0x1b, "LTP_UTLB_3" }, + { 0x1c, "LTP_UTLB_4" }, + { 0x1d, "LTP_UTLB_5" }, + { 0x1e, "LTP_UTLB_6" }, + { 0x1f, "LTP_UTLB_7" }, + { 0x20, "RGG_UTLB" }, + { 0x21, "T1_8" }, + { 0x22, "T1_9" }, + { 0x23, "T1_10" }, + { 0x24, "T1_11" }, + { 0x25, "T1_12" }, + { 0x26, "T1_13" }, + { 0x27, "T1_14" }, + { 0x28, "T1_15" }, + { 0x29, "TPCCS_0" }, + { 0x2a, "TPCCS_1" }, + { 0x2b, "TPCCS_2" }, + { 0x2c, "TPCCS_3" }, + { 0x2d, "TPCCS_4" }, + { 0x2e, "TPCCS_5" }, + { 0x2f, "TPCCS_6" }, + { 0x30, "TPCCS_7" }, + { 0x31, "PE_8" }, + { 0x32, "PE_9" }, + { 0x33, "TPCCS_8" }, + { 0x34, "TPCCS_9" }, + { 0x35, "T1_16" }, + { 0x36, "T1_17" }, + { 0x37, "T1_18" }, + { 0x38, "T1_19" }, + { 0x39, "PE_10" }, + { 0x3a, "PE_11" }, + { 0x3b, "TPCCS_10" }, + { 0x3c, "TPCCS_11" }, + { 0x3d, "T1_20" }, + { 0x3e, "T1_21" }, + { 0x3f, "T1_22" }, + { 0x40, "T1_23" }, + { 0x41, "PE_12" }, + { 0x42, "PE_13" }, + { 0x43, "TPCCS_12" }, + { 0x44, "TPCCS_13" }, + { 0x45, "T1_24" }, + { 0x46, "T1_25" }, + { 0x47, "T1_26" }, + { 0x48, "T1_27" }, + { 0x49, "PE_14" }, + { 0x4a, "PE_15" }, + { 0x4b, "TPCCS_14" }, + { 0x4c, "TPCCS_15" }, + { 0x4d, "T1_28" }, + { 0x4e, "T1_29" }, + { 0x4f, "T1_30" }, + { 0x50, "T1_31" }, + { 0x51, "PE_16" }, + { 0x52, "PE_17" }, + { 0x53, "TPCCS_16" }, + { 0x54, "TPCCS_17" }, + { 0x55, "T1_32" }, + { 0x56, "T1_33" }, + { 0x57, "T1_34" }, + { 0x58, "T1_35" }, + { 0x59, "PE_18" }, + { 0x5a, "PE_19" }, + { 0x5b, "TPCCS_18" }, + { 0x5c, "TPCCS_19" }, + { 0x5d, "T1_36" }, + { 0x5e, "T1_37" }, + { 0x5f, "T1_38" }, + { 0x60, "T1_39" }, + {} +}; + +static const struct nvkm_enum +gv100_fifo_fault_hubclient[] = { + { 0x00, "VIP" }, + { 0x01, "CE0" }, + { 0x02, "CE1" }, + { 0x03, "DNISO" }, + { 0x04, "FE" }, + { 0x05, "FECS" }, + { 0x06, "HOST" }, + { 0x07, "HOST_CPU" }, + { 0x08, "HOST_CPU_NB" }, + { 0x09, "ISO" }, + { 0x0a, "MMU" }, + { 0x0b, "NVDEC" }, + { 0x0d, "NVENC1" }, + { 0x0e, "NISO" }, + { 0x0f, "P2P" }, + { 0x10, "PD" }, + { 0x11, "PERF" }, + { 0x12, "PMU" }, + { 0x13, "RASTERTWOD" }, + { 0x14, "SCC" }, + { 0x15, "SCC_NB" }, + { 0x16, "SEC" }, + { 0x17, "SSYNC" }, + { 0x18, "CE2" }, + { 0x19, "XV" }, + { 0x1a, "MMU_NB" }, + { 0x1b, "NVENC0" }, + { 0x1c, "DFALCON" }, + { 0x1d, "SKED" }, + { 0x1e, "AFALCON" }, + { 0x1f, "DONT_CARE" }, + { 0x20, "HSCE0" }, + { 0x21, "HSCE1" }, + { 0x22, "HSCE2" }, + { 0x23, "HSCE3" }, + { 0x24, "HSCE4" }, + { 0x25, "HSCE5" }, + { 0x26, "HSCE6" }, + { 0x27, "HSCE7" }, + { 0x28, "HSCE8" }, + { 0x29, "HSCE9" }, + { 0x2a, "HSHUB" }, + { 0x2b, "PTP_X0" }, + { 0x2c, "PTP_X1" }, + { 0x2d, "PTP_X2" }, + { 0x2e, "PTP_X3" }, + { 0x2f, "PTP_X4" }, + { 0x30, "PTP_X5" }, + { 0x31, "PTP_X6" }, + { 0x32, "PTP_X7" }, + { 0x33, "NVENC2" }, + { 0x34, "VPR_SCRUBBER0" }, + { 0x35, "VPR_SCRUBBER1" }, + { 0x36, "DWBIF" }, + { 0x37, "FBFALCON" }, + { 0x38, "CE_SHIM" }, + { 0x39, "GSP" }, + {} +}; + +static const struct nvkm_enum +gv100_fifo_fault_reason[] = { + { 0x00, "PDE" }, + { 0x01, "PDE_SIZE" }, + { 0x02, "PTE" }, + { 0x03, "VA_LIMIT_VIOLATION" }, + { 0x04, "UNBOUND_INST_BLOCK" }, + { 0x05, "PRIV_VIOLATION" }, + { 0x06, "RO_VIOLATION" }, + { 0x07, "WO_VIOLATION" }, + { 0x08, "PITCH_MASK_VIOLATION" }, + { 0x09, "WORK_CREATION" }, + { 0x0a, "UNSUPPORTED_APERTURE" }, + { 0x0b, "COMPRESSION_FAILURE" }, + { 0x0c, "UNSUPPORTED_KIND" }, + { 0x0d, "REGION_VIOLATION" }, + { 0x0e, "POISONED" }, + { 0x0f, "ATOMIC_VIOLATION" }, + {} +}; + +static const struct nvkm_enum +gv100_fifo_fault_engine[] = { + { 0x01, "DISPLAY" }, + { 0x03, "PTP" }, + { 0x04, "BAR1", NULL, NVKM_SUBDEV_BAR }, + { 0x05, "BAR2", NULL, NVKM_SUBDEV_INSTMEM }, + { 0x06, "PWR_PMU" }, + { 0x08, "IFB", NULL, NVKM_ENGINE_IFB }, + { 0x09, "PERF" }, + { 0x1f, "PHYSICAL" }, + { 0x20, "HOST0" }, + { 0x21, "HOST1" }, + { 0x22, "HOST2" }, + { 0x23, "HOST3" }, + { 0x24, "HOST4" }, + { 0x25, "HOST5" }, + { 0x26, "HOST6" }, + { 0x27, "HOST7" }, + { 0x28, "HOST8" }, + { 0x29, "HOST9" }, + { 0x2a, "HOST10" }, + { 0x2b, "HOST11" }, + { 0x2c, "HOST12" }, + { 0x2d, "HOST13" }, + {} +}; + +static const struct nvkm_enum +gv100_fifo_fault_access[] = { + { 0x0, "VIRT_READ" }, + { 0x1, "VIRT_WRITE" }, + { 0x2, "VIRT_ATOMIC" }, + { 0x3, "VIRT_PREFETCH" }, + { 0x4, "VIRT_ATOMIC_WEAK" }, + { 0x8, "PHYS_READ" }, + { 0x9, "PHYS_WRITE" }, + { 0xa, "PHYS_ATOMIC" }, + { 0xb, "PHYS_PREFETCH" }, + {} +}; + +static const struct gk104_fifo_func +gv100_fifo = { + .init_pbdma_timeout = gk208_fifo_init_pbdma_timeout, + .fault.access = gv100_fifo_fault_access, + .fault.engine = gv100_fifo_fault_engine, + .fault.reason = gv100_fifo_fault_reason, + .fault.hubclient = gv100_fifo_fault_hubclient, + .fault.gpcclient = gv100_fifo_fault_gpcclient, + .runlist = &gv100_fifo_runlist, + .user = {{-1,-1,VOLTA_USERMODE_A }, gv100_fifo_user_new }, + .chan = {{ 0, 0,VOLTA_CHANNEL_GPFIFO_A}, gv100_fifo_gpfifo_new }, + .cgrp_force = true, +}; + +int +gv100_fifo_new(struct nvkm_device *device, int index, struct nvkm_fifo **pfifo) +{ + return gk104_fifo_new_(&gv100_fifo, device, index, 4096, pfifo); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/priv.h b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/priv.h index ae76b1aaccd4..d5acbba293f4 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/priv.h +++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/priv.h @@ -18,16 +18,19 @@ struct nvkm_fifo_chan_oclass; struct nvkm_fifo_func { void *(*dtor)(struct nvkm_fifo *); int (*oneinit)(struct nvkm_fifo *); + int (*info)(struct nvkm_fifo *, u64 mthd, u64 *data); void (*init)(struct nvkm_fifo *); void (*fini)(struct nvkm_fifo *); void (*intr)(struct nvkm_fifo *); + void (*fault)(struct nvkm_fifo *, struct nvkm_fault_data *); void (*pause)(struct nvkm_fifo *, unsigned long *); void (*start)(struct nvkm_fifo *, unsigned long *); void (*uevent_init)(struct nvkm_fifo *); void (*uevent_fini)(struct nvkm_fifo *); void (*recover_chan)(struct nvkm_fifo *, int chid); - int (*class_get)(struct nvkm_fifo *, int index, - const struct nvkm_fifo_chan_oclass **); + int (*class_get)(struct nvkm_fifo *, int index, struct nvkm_oclass *); + int (*class_new)(struct nvkm_fifo *, const struct nvkm_oclass *, + void *, u32, struct nvkm_object **); const struct nvkm_fifo_chan_oclass *chan[]; }; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/user.h b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/user.h new file mode 100644 index 000000000000..ed840921ebe8 --- /dev/null +++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/user.h @@ -0,0 +1,6 @@ +#ifndef __NVKM_FIFO_USER_H__ +#define __NVKM_FIFO_USER_H__ +#include "priv.h" +int gv100_fifo_user_new(const struct nvkm_oclass *, void *, u32, + struct nvkm_object **); +#endif diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/usergv100.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/usergv100.c new file mode 100644 index 000000000000..3dc3b8b312de --- /dev/null +++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/usergv100.c @@ -0,0 +1,45 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "user.h" + +static int +gv100_fifo_user_map(struct nvkm_object *object, void *argv, u32 argc, + enum nvkm_object_map *type, u64 *addr, u64 *size) +{ + struct nvkm_device *device = object->engine->subdev.device; + *addr = 0x810000 + device->func->resource_addr(device, 0); + *size = 0x010000; + *type = NVKM_OBJECT_MAP_IO; + return 0; +} + +static const struct nvkm_object_func +gv100_fifo_user = { + .map = gv100_fifo_user_map, +}; + +int +gv100_fifo_user_new(const struct nvkm_oclass *oclass, void *argv, u32 argc, + struct nvkm_object **pobject) +{ + return nvkm_object_new_(&gv100_fifo_user, oclass, argv, argc, pobject); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/Kbuild b/drivers/gpu/drm/nouveau/nvkm/engine/gr/Kbuild index 8a22558b7b52..93e3733f54e2 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/Kbuild +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/Kbuild @@ -33,8 +33,10 @@ nvkm-y += nvkm/engine/gr/gm200.o nvkm-y += nvkm/engine/gr/gm20b.o nvkm-y += nvkm/engine/gr/gp100.o nvkm-y += nvkm/engine/gr/gp102.o +nvkm-y += nvkm/engine/gr/gp104.o nvkm-y += nvkm/engine/gr/gp107.o nvkm-y += nvkm/engine/gr/gp10b.o +nvkm-y += nvkm/engine/gr/gv100.o nvkm-y += nvkm/engine/gr/ctxnv40.o nvkm-y += nvkm/engine/gr/ctxnv50.o @@ -54,4 +56,6 @@ nvkm-y += nvkm/engine/gr/ctxgm200.o nvkm-y += nvkm/engine/gr/ctxgm20b.o nvkm-y += nvkm/engine/gr/ctxgp100.o nvkm-y += nvkm/engine/gr/ctxgp102.o +nvkm-y += nvkm/engine/gr/ctxgp104.o nvkm-y += nvkm/engine/gr/ctxgp107.o +nvkm-y += nvkm/engine/gr/ctxgv100.o diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf100.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf100.c index 881015080d83..e813a3f8ea93 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf100.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf100.c @@ -850,12 +850,17 @@ gf100_grctx_init_gcc_0[] = { }; const struct gf100_gr_pack -gf100_grctx_pack_gpc[] = { +gf100_grctx_pack_gpc_0[] = { { gf100_grctx_init_gpc_unk_0 }, { gf100_grctx_init_prop_0 }, { gf100_grctx_init_gpc_unk_1 }, { gf100_grctx_init_setup_0 }, { gf100_grctx_init_zcull_0 }, + {} +}; + +const struct gf100_gr_pack +gf100_grctx_pack_gpc_1[] = { { gf100_grctx_init_crstr_0 }, { gf100_grctx_init_gpm_0 }, { gf100_grctx_init_gcc_0 }, @@ -1025,6 +1030,13 @@ gf100_grctx_mmio_item(struct gf100_grctx *info, u32 addr, u32 data, } void +gf100_grctx_generate_r419cb8(struct gf100_gr *gr) +{ + struct nvkm_device *device = gr->base.engine.subdev.device; + nvkm_mask(device, 0x419cb8, 0x00007c00, 0x00000000); +} + +void gf100_grctx_generate_bundle(struct gf100_grctx *info) { const struct gf100_grctx_func *grctx = info->gr->func->grctx; @@ -1080,89 +1092,38 @@ gf100_grctx_generate_unkn(struct gf100_gr *gr) } void -gf100_grctx_generate_tpcid(struct gf100_gr *gr) -{ - struct nvkm_device *device = gr->base.engine.subdev.device; - int gpc, tpc, id; - - for (tpc = 0, id = 0; tpc < 4; tpc++) { - for (gpc = 0; gpc < gr->gpc_nr; gpc++) { - if (tpc < gr->tpc_nr[gpc]) { - nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x698), id); - nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x4e8), id); - nvkm_wr32(device, GPC_UNIT(gpc, 0x0c10 + tpc * 4), id); - nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x088), id); - id++; - } - - nvkm_wr32(device, GPC_UNIT(gpc, 0x0c08), gr->tpc_nr[gpc]); - nvkm_wr32(device, GPC_UNIT(gpc, 0x0c8c), gr->tpc_nr[gpc]); - } - } -} - -void -gf100_grctx_generate_r406028(struct gf100_gr *gr) -{ - struct nvkm_device *device = gr->base.engine.subdev.device; - u32 tmp[GPC_MAX / 8] = {}, i = 0; - for (i = 0; i < gr->gpc_nr; i++) - tmp[i / 8] |= gr->tpc_nr[i] << ((i % 8) * 4); - for (i = 0; i < 4; i++) { - nvkm_wr32(device, 0x406028 + (i * 4), tmp[i]); - nvkm_wr32(device, 0x405870 + (i * 4), tmp[i]); - } -} - -void gf100_grctx_generate_r4060a8(struct gf100_gr *gr) { struct nvkm_device *device = gr->base.engine.subdev.device; - u8 tpcnr[GPC_MAX], data[TPC_MAX]; - int gpc, tpc, i; - - memcpy(tpcnr, gr->tpc_nr, sizeof(gr->tpc_nr)); - memset(data, 0x1f, sizeof(data)); - - gpc = -1; - for (tpc = 0; tpc < gr->tpc_total; tpc++) { - do { - gpc = (gpc + 1) % gr->gpc_nr; - } while (!tpcnr[gpc]); - tpcnr[gpc]--; - data[tpc] = gpc; + const u8 gpcmax = nvkm_rd32(device, 0x022430); + const u8 tpcmax = nvkm_rd32(device, 0x022434) * gpcmax; + int i, j, sm = 0; + u32 data; + + for (i = 0; i < DIV_ROUND_UP(tpcmax, 4); i++) { + for (data = 0, j = 0; j < 4; j++) { + if (sm < gr->sm_nr) + data |= gr->sm[sm++].gpc << (j * 8); + else + data |= 0x1f << (j * 8); + } + nvkm_wr32(device, 0x4060a8 + (i * 4), data); } - - for (i = 0; i < 4; i++) - nvkm_wr32(device, 0x4060a8 + (i * 4), ((u32 *)data)[i]); } void -gf100_grctx_generate_r418bb8(struct gf100_gr *gr) +gf100_grctx_generate_rop_mapping(struct gf100_gr *gr) { struct nvkm_device *device = gr->base.engine.subdev.device; u32 data[6] = {}, data2[2] = {}; - u8 tpcnr[GPC_MAX]; u8 shift, ntpcv; - int gpc, tpc, i; - - /* calculate first set of magics */ - memcpy(tpcnr, gr->tpc_nr, sizeof(gr->tpc_nr)); - - gpc = -1; - for (tpc = 0; tpc < gr->tpc_total; tpc++) { - do { - gpc = (gpc + 1) % gr->gpc_nr; - } while (!tpcnr[gpc]); - tpcnr[gpc]--; - - data[tpc / 6] |= gpc << ((tpc % 6) * 5); - } + int i; - for (; tpc < 32; tpc++) - data[tpc / 6] |= 7 << ((tpc % 6) * 5); + /* Pack tile map into register format. */ + for (i = 0; i < 32; i++) + data[i / 6] |= (gr->tile[i] & 0x07) << ((i % 6) * 5); - /* and the second... */ + /* Magic. */ shift = 0; ntpcv = gr->tpc_total; while (!(ntpcv & (1 << 4))) { @@ -1197,40 +1158,214 @@ gf100_grctx_generate_r418bb8(struct gf100_gr *gr) } void -gf100_grctx_generate_r406800(struct gf100_gr *gr) +gf100_grctx_generate_max_ways_evict(struct gf100_gr *gr) { struct nvkm_device *device = gr->base.engine.subdev.device; - u64 tpc_mask = 0, tpc_set = 0; - u8 tpcnr[GPC_MAX]; - int gpc, tpc; - int i, a, b; - - memcpy(tpcnr, gr->tpc_nr, sizeof(gr->tpc_nr)); - for (gpc = 0; gpc < gr->gpc_nr; gpc++) - tpc_mask |= ((1ULL << gr->tpc_nr[gpc]) - 1) << (gpc * 8); - - for (i = 0, gpc = -1, b = -1; i < 32; i++) { - a = (i * (gr->tpc_total - 1)) / 32; - if (a != b) { - b = a; - do { - gpc = (gpc + 1) % gr->gpc_nr; - } while (!tpcnr[gpc]); - tpc = gr->tpc_nr[gpc] - tpcnr[gpc]--; - - tpc_set |= 1ULL << ((gpc * 8) + tpc); + u32 fbps = nvkm_rd32(device, 0x121c74); + if (fbps == 1) + nvkm_mask(device, 0x17e91c, 0x001f0000, 0x00090000); +} + +static const u32 +gf100_grctx_alpha_beta_map[17][32] = { + [1] = { + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + }, + [2] = { + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + }, + //XXX: 3 + [4] = { + 1, 1, 1, 1, 1, 1, 1, 1, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 3, 3, 3, 3, 3, 3, 3, 3, + }, + //XXX: 5 + //XXX: 6 + [7] = { + 1, 1, 1, 1, + 2, 2, 2, 2, 2, 2, + 3, 3, 3, 3, 3, 3, + 4, 4, 4, 4, 4, 4, + 5, 5, 5, 5, 5, 5, + 6, 6, 6, 6, + }, + [8] = { + 1, 1, 1, + 2, 2, 2, 2, 2, + 3, 3, 3, 3, 3, + 4, 4, 4, 4, 4, 4, + 5, 5, 5, 5, 5, + 6, 6, 6, 6, 6, + 7, 7, 7, + }, + //XXX: 9 + //XXX: 10 + [11] = { + 1, 1, + 2, 2, 2, 2, + 3, 3, 3, + 4, 4, 4, 4, + 5, 5, 5, + 6, 6, 6, + 7, 7, 7, 7, + 8, 8, 8, + 9, 9, 9, 9, + 10, 10, + }, + //XXX: 12 + //XXX: 13 + [14] = { + 1, 1, + 2, 2, + 3, 3, 3, + 4, 4, 4, + 5, 5, + 6, 6, 6, + 7, 7, + 8, 8, 8, + 9, 9, + 10, 10, 10, + 11, 11, 11, + 12, 12, + 13, 13, + }, + [15] = { + 1, 1, + 2, 2, + 3, 3, + 4, 4, 4, + 5, 5, + 6, 6, 6, + 7, 7, + 8, 8, + 9, 9, 9, + 10, 10, + 11, 11, 11, + 12, 12, + 13, 13, + 14, 14, + }, + [16] = { + 1, 1, + 2, 2, + 3, 3, + 4, 4, + 5, 5, + 6, 6, 6, + 7, 7, + 8, 8, + 9, 9, + 10, 10, 10, + 11, 11, + 12, 12, + 13, 13, + 14, 14, + 15, 15, + }, +}; + +void +gf100_grctx_generate_alpha_beta_tables(struct gf100_gr *gr) +{ + struct nvkm_subdev *subdev = &gr->base.engine.subdev; + struct nvkm_device *device = subdev->device; + int i, gpc; + + for (i = 0; i < 32; i++) { + u32 atarget = gf100_grctx_alpha_beta_map[gr->tpc_total][i]; + u32 abits[GPC_MAX] = {}, amask = 0, bmask = 0; + + if (!atarget) { + nvkm_warn(subdev, "missing alpha/beta mapping table\n"); + atarget = max_t(u32, gr->tpc_total * i / 32, 1); + } + + while (atarget) { + for (gpc = 0; atarget && gpc < gr->gpc_nr; gpc++) { + if (abits[gpc] < gr->tpc_nr[gpc]) { + abits[gpc]++; + atarget--; + } + } } - nvkm_wr32(device, 0x406800 + (i * 0x20), lower_32_bits(tpc_set)); - nvkm_wr32(device, 0x406c00 + (i * 0x20), lower_32_bits(tpc_set ^ tpc_mask)); - if (gr->gpc_nr > 4) { - nvkm_wr32(device, 0x406804 + (i * 0x20), upper_32_bits(tpc_set)); - nvkm_wr32(device, 0x406c04 + (i * 0x20), upper_32_bits(tpc_set ^ tpc_mask)); + for (gpc = 0; gpc < gr->gpc_nr; gpc++) { + u32 bbits = gr->tpc_nr[gpc] - abits[gpc]; + amask |= ((1 << abits[gpc]) - 1) << (gpc * 8); + bmask |= ((1 << bbits) - 1) << abits[gpc] << (gpc * 8); } + + nvkm_wr32(device, 0x406800 + (i * 0x20), amask); + nvkm_wr32(device, 0x406c00 + (i * 0x20), bmask); } } void +gf100_grctx_generate_tpc_nr(struct gf100_gr *gr, int gpc) +{ + struct nvkm_device *device = gr->base.engine.subdev.device; + nvkm_wr32(device, GPC_UNIT(gpc, 0x0c08), gr->tpc_nr[gpc]); + nvkm_wr32(device, GPC_UNIT(gpc, 0x0c8c), gr->tpc_nr[gpc]); +} + +void +gf100_grctx_generate_sm_id(struct gf100_gr *gr, int gpc, int tpc, int sm) +{ + struct nvkm_device *device = gr->base.engine.subdev.device; + nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x698), sm); + nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x4e8), sm); + nvkm_wr32(device, GPC_UNIT(gpc, 0x0c10 + tpc * 4), sm); + nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x088), sm); +} + +void +gf100_grctx_generate_floorsweep(struct gf100_gr *gr) +{ + struct nvkm_device *device = gr->base.engine.subdev.device; + const struct gf100_grctx_func *func = gr->func->grctx; + int gpc, sm, i, j; + u32 data; + + for (sm = 0; sm < gr->sm_nr; sm++) { + func->sm_id(gr, gr->sm[sm].gpc, gr->sm[sm].tpc, sm); + if (func->tpc_nr) + func->tpc_nr(gr, gr->sm[sm].gpc); + } + + for (gpc = 0, i = 0; i < 4; i++) { + for (data = 0, j = 0; j < 8 && gpc < gr->gpc_nr; j++, gpc++) + data |= gr->tpc_nr[gpc] << (j * 4); + nvkm_wr32(device, 0x406028 + (i * 4), data); + nvkm_wr32(device, 0x405870 + (i * 4), data); + } + + if (func->r4060a8) + func->r4060a8(gr); + + func->rop_mapping(gr); + + if (func->alpha_beta_tables) + func->alpha_beta_tables(gr); + if (func->max_ways_evict) + func->max_ways_evict(gr); + if (func->dist_skip_table) + func->dist_skip_table(gr); + if (func->r406500) + func->r406500(gr); + if (func->gpc_tpc_nr) + func->gpc_tpc_nr(gr); + if (func->r419f78) + func->r419f78(gr); + if (func->tpc_mask) + func->tpc_mask(gr); + if (func->smid_config) + func->smid_config(gr); +} + +void gf100_grctx_generate_main(struct gf100_gr *gr, struct gf100_grctx *info) { struct nvkm_device *device = gr->base.engine.subdev.device; @@ -1239,29 +1374,63 @@ gf100_grctx_generate_main(struct gf100_gr *gr, struct gf100_grctx *info) nvkm_mc_unk260(device, 0); - gf100_gr_mmio(gr, grctx->hub); - gf100_gr_mmio(gr, grctx->gpc); - gf100_gr_mmio(gr, grctx->zcull); - gf100_gr_mmio(gr, grctx->tpc); - gf100_gr_mmio(gr, grctx->ppc); + if (!gr->fuc_sw_ctx) { + gf100_gr_mmio(gr, grctx->hub); + gf100_gr_mmio(gr, grctx->gpc_0); + gf100_gr_mmio(gr, grctx->zcull); + gf100_gr_mmio(gr, grctx->gpc_1); + gf100_gr_mmio(gr, grctx->tpc); + gf100_gr_mmio(gr, grctx->ppc); + } else { + gf100_gr_mmio(gr, gr->fuc_sw_ctx); + } + + gf100_gr_wait_idle(gr); idle_timeout = nvkm_mask(device, 0x404154, 0xffffffff, 0x00000000); - grctx->bundle(info); grctx->pagepool(info); + grctx->bundle(info); grctx->attrib(info); + if (grctx->patch_ltc) + grctx->patch_ltc(info); grctx->unkn(gr); - gf100_grctx_generate_tpcid(gr); - gf100_grctx_generate_r406028(gr); - gf100_grctx_generate_r4060a8(gr); - gf100_grctx_generate_r418bb8(gr); - gf100_grctx_generate_r406800(gr); + gf100_grctx_generate_floorsweep(gr); + + gf100_gr_wait_idle(gr); + + if (grctx->r400088) grctx->r400088(gr, false); + if (gr->fuc_bundle) + gf100_gr_icmd(gr, gr->fuc_bundle); + else + gf100_gr_icmd(gr, grctx->icmd); + if (grctx->sw_veid_bundle_init) + gf100_gr_icmd(gr, grctx->sw_veid_bundle_init); + if (grctx->r400088) grctx->r400088(gr, true); - gf100_gr_icmd(gr, grctx->icmd); nvkm_wr32(device, 0x404154, idle_timeout); - gf100_gr_mthd(gr, grctx->mthd); + + if (gr->fuc_method) + gf100_gr_mthd(gr, gr->fuc_method); + else + gf100_gr_mthd(gr, grctx->mthd); nvkm_mc_unk260(device, 1); + + if (grctx->r419cb8) + grctx->r419cb8(gr); + if (grctx->r418800) + grctx->r418800(gr); + if (grctx->r419eb0) + grctx->r419eb0(gr); + if (grctx->r419e00) + grctx->r419e00(gr); + if (grctx->r418e94) + grctx->r418e94(gr); + if (grctx->r419a3c) + grctx->r419a3c(gr); + if (grctx->r408840) + grctx->r408840(gr); } #define CB_RESERVED 0x80000 @@ -1280,6 +1449,32 @@ gf100_grctx_generate(struct gf100_gr *gr) int ret, i; u64 addr; + /* NV_PGRAPH_FE_PWR_MODE_FORCE_ON. */ + nvkm_wr32(device, 0x404170, 0x00000012); + nvkm_msec(device, 2000, + if (!(nvkm_rd32(device, 0x404170) & 0x00000010)) + break; + ); + + if (grctx->unkn88c) + grctx->unkn88c(gr, true); + + /* Reset FECS. */ + nvkm_wr32(device, 0x409614, 0x00000070); + nvkm_usec(device, 10, NVKM_DELAY); + nvkm_mask(device, 0x409614, 0x00000700, 0x00000700); + nvkm_usec(device, 10, NVKM_DELAY); + nvkm_rd32(device, 0x409614); + + if (grctx->unkn88c) + grctx->unkn88c(gr, false); + + /* NV_PGRAPH_FE_PWR_MODE_AUTO. */ + nvkm_wr32(device, 0x404170, 0x00000010); + + /* Init SCC RAM. */ + nvkm_wr32(device, 0x40802c, 0x00000001); + /* Allocate memory to for a "channel", which we'll use to generate * the default context values. */ @@ -1392,7 +1587,8 @@ gf100_grctx = { .main = gf100_grctx_generate_main, .unkn = gf100_grctx_generate_unkn, .hub = gf100_grctx_pack_hub, - .gpc = gf100_grctx_pack_gpc, + .gpc_0 = gf100_grctx_pack_gpc_0, + .gpc_1 = gf100_grctx_pack_gpc_1, .zcull = gf100_grctx_pack_zcull, .tpc = gf100_grctx_pack_tpc, .icmd = gf100_grctx_pack_icmd, @@ -1404,4 +1600,11 @@ gf100_grctx = { .attrib = gf100_grctx_generate_attrib, .attrib_nr_max = 0x324, .attrib_nr = 0x218, + .sm_id = gf100_grctx_generate_sm_id, + .tpc_nr = gf100_grctx_generate_tpc_nr, + .r4060a8 = gf100_grctx_generate_r4060a8, + .rop_mapping = gf100_grctx_generate_rop_mapping, + .alpha_beta_tables = gf100_grctx_generate_alpha_beta_tables, + .max_ways_evict = gf100_grctx_generate_max_ways_evict, + .r419cb8 = gf100_grctx_generate_r419cb8, }; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf100.h b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf100.h index 5199e5aa0cb7..33e932bd73b1 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf100.h +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf100.h @@ -21,19 +21,22 @@ void gf100_grctx_mmio_item(struct gf100_grctx *, u32 addr, u32 data, int s, int) #define mmio_wr32(a,b,c) mmio_refn((a), (b), (c), 0, -1) struct gf100_grctx_func { + void (*unkn88c)(struct gf100_gr *, bool on); /* main context generation function */ void (*main)(struct gf100_gr *, struct gf100_grctx *); /* context-specific modify-on-first-load list generation function */ void (*unkn)(struct gf100_gr *); /* mmio context data */ const struct gf100_gr_pack *hub; - const struct gf100_gr_pack *gpc; + const struct gf100_gr_pack *gpc_0; + const struct gf100_gr_pack *gpc_1; const struct gf100_gr_pack *zcull; const struct gf100_gr_pack *tpc; const struct gf100_gr_pack *ppc; /* indirect context data, generated with icmds/mthds */ const struct gf100_gr_pack *icmd; const struct gf100_gr_pack *mthd; + const struct gf100_gr_pack *sw_veid_bundle_init; /* bundle circular buffer */ void (*bundle)(struct gf100_grctx *); u32 bundle_size; @@ -48,6 +51,31 @@ struct gf100_grctx_func { u32 attrib_nr; u32 alpha_nr_max; u32 alpha_nr; + u32 gfxp_nr; + /* other patch buffer stuff */ + void (*patch_ltc)(struct gf100_grctx *); + /* floorsweeping */ + void (*sm_id)(struct gf100_gr *, int gpc, int tpc, int sm); + void (*tpc_nr)(struct gf100_gr *, int gpc); + void (*r4060a8)(struct gf100_gr *); + void (*rop_mapping)(struct gf100_gr *); + void (*alpha_beta_tables)(struct gf100_gr *); + void (*max_ways_evict)(struct gf100_gr *); + void (*dist_skip_table)(struct gf100_gr *); + void (*r406500)(struct gf100_gr *); + void (*gpc_tpc_nr)(struct gf100_gr *); + void (*r419f78)(struct gf100_gr *); + void (*tpc_mask)(struct gf100_gr *); + void (*smid_config)(struct gf100_gr *); + /* misc other things */ + void (*r400088)(struct gf100_gr *, bool); + void (*r419cb8)(struct gf100_gr *); + void (*r418800)(struct gf100_gr *); + void (*r419eb0)(struct gf100_gr *); + void (*r419e00)(struct gf100_gr *); + void (*r418e94)(struct gf100_gr *); + void (*r419a3c)(struct gf100_gr *); + void (*r408840)(struct gf100_gr *); }; extern const struct gf100_grctx_func gf100_grctx; @@ -57,11 +85,14 @@ void gf100_grctx_generate_bundle(struct gf100_grctx *); void gf100_grctx_generate_pagepool(struct gf100_grctx *); void gf100_grctx_generate_attrib(struct gf100_grctx *); void gf100_grctx_generate_unkn(struct gf100_gr *); -void gf100_grctx_generate_tpcid(struct gf100_gr *); -void gf100_grctx_generate_r406028(struct gf100_gr *); +void gf100_grctx_generate_floorsweep(struct gf100_gr *); +void gf100_grctx_generate_sm_id(struct gf100_gr *, int, int, int); +void gf100_grctx_generate_tpc_nr(struct gf100_gr *, int); void gf100_grctx_generate_r4060a8(struct gf100_gr *); -void gf100_grctx_generate_r418bb8(struct gf100_gr *); -void gf100_grctx_generate_r406800(struct gf100_gr *); +void gf100_grctx_generate_rop_mapping(struct gf100_gr *); +void gf100_grctx_generate_alpha_beta_tables(struct gf100_gr *); +void gf100_grctx_generate_max_ways_evict(struct gf100_gr *); +void gf100_grctx_generate_r419cb8(struct gf100_gr *); extern const struct gf100_grctx_func gf108_grctx; void gf108_grctx_generate_attrib(struct gf100_grctx *); @@ -72,22 +103,25 @@ extern const struct gf100_grctx_func gf110_grctx; extern const struct gf100_grctx_func gf117_grctx; void gf117_grctx_generate_attrib(struct gf100_grctx *); +void gf117_grctx_generate_rop_mapping(struct gf100_gr *); +void gf117_grctx_generate_dist_skip_table(struct gf100_gr *); extern const struct gf100_grctx_func gf119_grctx; extern const struct gf100_grctx_func gk104_grctx; +void gk104_grctx_generate_alpha_beta_tables(struct gf100_gr *); +void gk104_grctx_generate_gpc_tpc_nr(struct gf100_gr *); + extern const struct gf100_grctx_func gk20a_grctx; -void gk104_grctx_generate_main(struct gf100_gr *, struct gf100_grctx *); void gk104_grctx_generate_bundle(struct gf100_grctx *); void gk104_grctx_generate_pagepool(struct gf100_grctx *); +void gk104_grctx_generate_patch_ltc(struct gf100_grctx *); void gk104_grctx_generate_unkn(struct gf100_gr *); -void gk104_grctx_generate_r418bb8(struct gf100_gr *); - -void gm107_grctx_generate_bundle(struct gf100_grctx *); -void gm107_grctx_generate_pagepool(struct gf100_grctx *); -void gm107_grctx_generate_attrib(struct gf100_grctx *); +void gk104_grctx_generate_r418800(struct gf100_gr *); extern const struct gf100_grctx_func gk110_grctx; +void gk110_grctx_generate_r419eb0(struct gf100_gr *); + extern const struct gf100_grctx_func gk110b_grctx; extern const struct gf100_grctx_func gk208_grctx; @@ -95,22 +129,30 @@ extern const struct gf100_grctx_func gm107_grctx; void gm107_grctx_generate_bundle(struct gf100_grctx *); void gm107_grctx_generate_pagepool(struct gf100_grctx *); void gm107_grctx_generate_attrib(struct gf100_grctx *); +void gm107_grctx_generate_sm_id(struct gf100_gr *, int, int, int); extern const struct gf100_grctx_func gm200_grctx; -void gm200_grctx_generate_tpcid(struct gf100_gr *); -void gm200_grctx_generate_405b60(struct gf100_gr *); +void gm200_grctx_generate_dist_skip_table(struct gf100_gr *); +void gm200_grctx_generate_r406500(struct gf100_gr *); +void gm200_grctx_generate_tpc_mask(struct gf100_gr *); +void gm200_grctx_generate_smid_config(struct gf100_gr *); +void gm200_grctx_generate_r419a3c(struct gf100_gr *); extern const struct gf100_grctx_func gm20b_grctx; extern const struct gf100_grctx_func gp100_grctx; -void gp100_grctx_generate_main(struct gf100_gr *, struct gf100_grctx *); void gp100_grctx_generate_pagepool(struct gf100_grctx *); +void gp100_grctx_generate_smid_config(struct gf100_gr *); extern const struct gf100_grctx_func gp102_grctx; void gp102_grctx_generate_attrib(struct gf100_grctx *); +extern const struct gf100_grctx_func gp104_grctx; + extern const struct gf100_grctx_func gp107_grctx; +extern const struct gf100_grctx_func gv100_grctx; + /* context init value lists */ extern const struct gf100_gr_pack gf100_grctx_pack_icmd[]; @@ -128,7 +170,8 @@ extern const struct gf100_gr_init gf100_grctx_init_memfmt_0[]; extern const struct gf100_gr_init gf100_grctx_init_rstr2d_0[]; extern const struct gf100_gr_init gf100_grctx_init_scc_0[]; -extern const struct gf100_gr_pack gf100_grctx_pack_gpc[]; +extern const struct gf100_gr_pack gf100_grctx_pack_gpc_0[]; +extern const struct gf100_gr_pack gf100_grctx_pack_gpc_1[]; extern const struct gf100_gr_init gf100_grctx_init_gpc_unk_0[]; extern const struct gf100_gr_init gf100_grctx_init_prop_0[]; extern const struct gf100_gr_init gf100_grctx_init_gpc_unk_1[]; @@ -177,6 +220,8 @@ extern const struct gf100_gr_init gf117_grctx_init_pe_0[]; extern const struct gf100_gr_init gf117_grctx_init_wwdx_0[]; +extern const struct gf100_gr_pack gf117_grctx_pack_gpc_1[]; + extern const struct gf100_gr_init gk104_grctx_init_memfmt_0[]; extern const struct gf100_gr_init gk104_grctx_init_ds_0[]; extern const struct gf100_gr_init gk104_grctx_init_scc_0[]; @@ -186,7 +231,6 @@ extern const struct gf100_gr_init gk104_grctx_init_gpm_0[]; extern const struct gf100_gr_init gk104_grctx_init_pes_0[]; extern const struct gf100_gr_pack gk104_grctx_pack_hub[]; -extern const struct gf100_gr_pack gk104_grctx_pack_gpc[]; extern const struct gf100_gr_pack gk104_grctx_pack_tpc[]; extern const struct gf100_gr_pack gk104_grctx_pack_ppc[]; extern const struct gf100_gr_pack gk104_grctx_pack_icmd[]; @@ -200,7 +244,8 @@ extern const struct gf100_gr_pack gk110_grctx_pack_hub[]; extern const struct gf100_gr_init gk110_grctx_init_pri_0[]; extern const struct gf100_gr_init gk110_grctx_init_cwd_0[]; -extern const struct gf100_gr_pack gk110_grctx_pack_gpc[]; +extern const struct gf100_gr_pack gk110_grctx_pack_gpc_0[]; +extern const struct gf100_gr_pack gk110_grctx_pack_gpc_1[]; extern const struct gf100_gr_init gk110_grctx_init_gpc_unk_2[]; extern const struct gf100_gr_init gk110_grctx_init_tex_0[]; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf104.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf104.c index 54fd74e9cca0..7a0564b6e3c7 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf104.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf104.c @@ -84,7 +84,8 @@ gf104_grctx = { .main = gf100_grctx_generate_main, .unkn = gf100_grctx_generate_unkn, .hub = gf100_grctx_pack_hub, - .gpc = gf100_grctx_pack_gpc, + .gpc_0 = gf100_grctx_pack_gpc_0, + .gpc_1 = gf100_grctx_pack_gpc_1, .zcull = gf100_grctx_pack_zcull, .tpc = gf104_grctx_pack_tpc, .icmd = gf100_grctx_pack_icmd, @@ -96,4 +97,11 @@ gf104_grctx = { .attrib = gf100_grctx_generate_attrib, .attrib_nr_max = 0x324, .attrib_nr = 0x218, + .sm_id = gf100_grctx_generate_sm_id, + .tpc_nr = gf100_grctx_generate_tpc_nr, + .r4060a8 = gf100_grctx_generate_r4060a8, + .rop_mapping = gf100_grctx_generate_rop_mapping, + .alpha_beta_tables = gf100_grctx_generate_alpha_beta_tables, + .max_ways_evict = gf100_grctx_generate_max_ways_evict, + .r419cb8 = gf100_grctx_generate_r419cb8, }; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf108.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf108.c index 82f71b10c06e..dda2c32e6232 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf108.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf108.c @@ -667,12 +667,17 @@ gf108_grctx_init_gpm_0[] = { }; static const struct gf100_gr_pack -gf108_grctx_pack_gpc[] = { +gf108_grctx_pack_gpc_0[] = { { gf100_grctx_init_gpc_unk_0 }, { gf100_grctx_init_prop_0 }, { gf100_grctx_init_gpc_unk_1 }, { gf108_grctx_init_setup_0 }, { gf100_grctx_init_zcull_0 }, + {} +}; + +static const struct gf100_gr_pack +gf108_grctx_pack_gpc_1[] = { { gf100_grctx_init_crstr_0 }, { gf108_grctx_init_gpm_0 }, { gf100_grctx_init_gcc_0 }, @@ -780,7 +785,8 @@ gf108_grctx = { .main = gf100_grctx_generate_main, .unkn = gf108_grctx_generate_unkn, .hub = gf108_grctx_pack_hub, - .gpc = gf108_grctx_pack_gpc, + .gpc_0 = gf108_grctx_pack_gpc_0, + .gpc_1 = gf108_grctx_pack_gpc_1, .zcull = gf100_grctx_pack_zcull, .tpc = gf108_grctx_pack_tpc, .icmd = gf108_grctx_pack_icmd, @@ -794,4 +800,11 @@ gf108_grctx = { .attrib_nr = 0x218, .alpha_nr_max = 0x324, .alpha_nr = 0x218, + .sm_id = gf100_grctx_generate_sm_id, + .tpc_nr = gf100_grctx_generate_tpc_nr, + .r4060a8 = gf100_grctx_generate_r4060a8, + .rop_mapping = gf100_grctx_generate_rop_mapping, + .alpha_beta_tables = gf100_grctx_generate_alpha_beta_tables, + .max_ways_evict = gf100_grctx_generate_max_ways_evict, + .r419cb8 = gf100_grctx_generate_r419cb8, }; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf110.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf110.c index 7df398b53f8f..f5cca5e6a4f2 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf110.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf110.c @@ -314,15 +314,12 @@ gf110_grctx_init_setup_0[] = { }; static const struct gf100_gr_pack -gf110_grctx_pack_gpc[] = { +gf110_grctx_pack_gpc_0[] = { { gf100_grctx_init_gpc_unk_0 }, { gf100_grctx_init_prop_0 }, { gf100_grctx_init_gpc_unk_1 }, { gf110_grctx_init_setup_0 }, { gf100_grctx_init_zcull_0 }, - { gf100_grctx_init_crstr_0 }, - { gf100_grctx_init_gpm_0 }, - { gf100_grctx_init_gcc_0 }, {} }; @@ -335,7 +332,8 @@ gf110_grctx = { .main = gf100_grctx_generate_main, .unkn = gf100_grctx_generate_unkn, .hub = gf100_grctx_pack_hub, - .gpc = gf110_grctx_pack_gpc, + .gpc_0 = gf110_grctx_pack_gpc_0, + .gpc_1 = gf100_grctx_pack_gpc_1, .zcull = gf100_grctx_pack_zcull, .tpc = gf100_grctx_pack_tpc, .icmd = gf110_grctx_pack_icmd, @@ -347,4 +345,11 @@ gf110_grctx = { .attrib = gf100_grctx_generate_attrib, .attrib_nr_max = 0x324, .attrib_nr = 0x218, + .sm_id = gf100_grctx_generate_sm_id, + .tpc_nr = gf100_grctx_generate_tpc_nr, + .r4060a8 = gf100_grctx_generate_r4060a8, + .rop_mapping = gf100_grctx_generate_rop_mapping, + .alpha_beta_tables = gf100_grctx_generate_alpha_beta_tables, + .max_ways_evict = gf100_grctx_generate_max_ways_evict, + .r419cb8 = gf100_grctx_generate_r419cb8, }; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf117.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf117.c index 19301d88577d..276c282d19aa 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf117.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf117.c @@ -84,12 +84,17 @@ gf117_grctx_init_setup_0[] = { }; static const struct gf100_gr_pack -gf117_grctx_pack_gpc[] = { +gf117_grctx_pack_gpc_0[] = { { gf100_grctx_init_gpc_unk_0 }, { gf119_grctx_init_prop_0 }, { gf119_grctx_init_gpc_unk_1 }, { gf117_grctx_init_setup_0 }, { gf100_grctx_init_zcull_0 }, + {} +}; + +const struct gf100_gr_pack +gf117_grctx_pack_gpc_1[] = { { gf119_grctx_init_crstr_0 }, { gf108_grctx_init_gpm_0 }, { gf100_grctx_init_gcc_0 }, @@ -180,6 +185,62 @@ gf117_grctx_pack_ppc[] = { ******************************************************************************/ void +gf117_grctx_generate_dist_skip_table(struct gf100_gr *gr) +{ + struct nvkm_device *device = gr->base.engine.subdev.device; + int i; + + for (i = 0; i < 8; i++) + nvkm_wr32(device, 0x4064d0 + (i * 0x04), 0x00000000); +} + +void +gf117_grctx_generate_rop_mapping(struct gf100_gr *gr) +{ + struct nvkm_device *device = gr->base.engine.subdev.device; + u32 data[6] = {}, data2[2] = {}; + u8 shift, ntpcv; + int i; + + /* Pack tile map into register format. */ + for (i = 0; i < 32; i++) + data[i / 6] |= (gr->tile[i] & 0x07) << ((i % 6) * 5); + + /* Magic. */ + shift = 0; + ntpcv = gr->tpc_total; + while (!(ntpcv & (1 << 4))) { + ntpcv <<= 1; + shift++; + } + + data2[0] = (ntpcv << 16); + data2[0] |= (shift << 21); + data2[0] |= (((1 << (0 + 5)) % ntpcv) << 24); + for (i = 1; i < 7; i++) + data2[1] |= ((1 << (i + 5)) % ntpcv) << ((i - 1) * 5); + + /* GPC_BROADCAST */ + nvkm_wr32(device, 0x418bb8, (gr->tpc_total << 8) | + gr->screen_tile_row_offset); + for (i = 0; i < 6; i++) + nvkm_wr32(device, 0x418b08 + (i * 4), data[i]); + + /* GPC_BROADCAST.TP_BROADCAST */ + nvkm_wr32(device, 0x41bfd0, (gr->tpc_total << 8) | + gr->screen_tile_row_offset | data2[0]); + nvkm_wr32(device, 0x41bfe4, data2[1]); + for (i = 0; i < 6; i++) + nvkm_wr32(device, 0x41bf00 + (i * 4), data[i]); + + /* UNK78xx */ + nvkm_wr32(device, 0x4078bc, (gr->tpc_total << 8) | + gr->screen_tile_row_offset); + for (i = 0; i < 6; i++) + nvkm_wr32(device, 0x40780c + (i * 4), data[i]); +} + +void gf117_grctx_generate_attrib(struct gf100_grctx *info) { struct gf100_gr *gr = info->gr; @@ -217,50 +278,13 @@ gf117_grctx_generate_attrib(struct gf100_grctx *info) } } -static void -gf117_grctx_generate_main(struct gf100_gr *gr, struct gf100_grctx *info) -{ - struct nvkm_device *device = gr->base.engine.subdev.device; - const struct gf100_grctx_func *grctx = gr->func->grctx; - u32 idle_timeout; - int i; - - nvkm_mc_unk260(device, 0); - - gf100_gr_mmio(gr, grctx->hub); - gf100_gr_mmio(gr, grctx->gpc); - gf100_gr_mmio(gr, grctx->zcull); - gf100_gr_mmio(gr, grctx->tpc); - gf100_gr_mmio(gr, grctx->ppc); - - idle_timeout = nvkm_mask(device, 0x404154, 0xffffffff, 0x00000000); - - grctx->bundle(info); - grctx->pagepool(info); - grctx->attrib(info); - grctx->unkn(gr); - - gf100_grctx_generate_tpcid(gr); - gf100_grctx_generate_r406028(gr); - gf100_grctx_generate_r4060a8(gr); - gk104_grctx_generate_r418bb8(gr); - gf100_grctx_generate_r406800(gr); - - for (i = 0; i < 8; i++) - nvkm_wr32(device, 0x4064d0 + (i * 0x04), 0x00000000); - - gf100_gr_icmd(gr, grctx->icmd); - nvkm_wr32(device, 0x404154, idle_timeout); - gf100_gr_mthd(gr, grctx->mthd); - nvkm_mc_unk260(device, 1); -} - const struct gf100_grctx_func gf117_grctx = { - .main = gf117_grctx_generate_main, + .main = gf100_grctx_generate_main, .unkn = gk104_grctx_generate_unkn, .hub = gf117_grctx_pack_hub, - .gpc = gf117_grctx_pack_gpc, + .gpc_0 = gf117_grctx_pack_gpc_0, + .gpc_1 = gf117_grctx_pack_gpc_1, .zcull = gf100_grctx_pack_zcull, .tpc = gf117_grctx_pack_tpc, .ppc = gf117_grctx_pack_ppc, @@ -275,4 +299,12 @@ gf117_grctx = { .attrib_nr = 0x218, .alpha_nr_max = 0x7ff, .alpha_nr = 0x324, + .sm_id = gf100_grctx_generate_sm_id, + .tpc_nr = gf100_grctx_generate_tpc_nr, + .r4060a8 = gf100_grctx_generate_r4060a8, + .rop_mapping = gf117_grctx_generate_rop_mapping, + .alpha_beta_tables = gf100_grctx_generate_alpha_beta_tables, + .max_ways_evict = gf100_grctx_generate_max_ways_evict, + .dist_skip_table = gf117_grctx_generate_dist_skip_table, + .r419cb8 = gf100_grctx_generate_r419cb8, }; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf119.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf119.c index 605185b078be..0cfe46366af6 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf119.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf119.c @@ -431,15 +431,12 @@ gf119_grctx_init_crstr_0[] = { }; static const struct gf100_gr_pack -gf119_grctx_pack_gpc[] = { +gf119_grctx_pack_gpc_0[] = { { gf100_grctx_init_gpc_unk_0 }, { gf119_grctx_init_prop_0 }, { gf119_grctx_init_gpc_unk_1 }, { gf119_grctx_init_setup_0 }, { gf100_grctx_init_zcull_0 }, - { gf119_grctx_init_crstr_0 }, - { gf108_grctx_init_gpm_0 }, - { gf100_grctx_init_gcc_0 }, {} }; @@ -503,7 +500,8 @@ gf119_grctx = { .main = gf100_grctx_generate_main, .unkn = gf108_grctx_generate_unkn, .hub = gf119_grctx_pack_hub, - .gpc = gf119_grctx_pack_gpc, + .gpc_0 = gf119_grctx_pack_gpc_0, + .gpc_1 = gf117_grctx_pack_gpc_1, .zcull = gf100_grctx_pack_zcull, .tpc = gf119_grctx_pack_tpc, .icmd = gf119_grctx_pack_icmd, @@ -517,4 +515,11 @@ gf119_grctx = { .attrib_nr = 0x218, .alpha_nr_max = 0x324, .alpha_nr = 0x218, + .sm_id = gf100_grctx_generate_sm_id, + .tpc_nr = gf100_grctx_generate_tpc_nr, + .r4060a8 = gf100_grctx_generate_r4060a8, + .rop_mapping = gf100_grctx_generate_rop_mapping, + .alpha_beta_tables = gf100_grctx_generate_alpha_beta_tables, + .max_ways_evict = gf100_grctx_generate_max_ways_evict, + .r419cb8 = gf100_grctx_generate_r419cb8, }; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgk104.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgk104.c index 825c8fd500bc..304e9d268bad 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgk104.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgk104.c @@ -739,13 +739,18 @@ gk104_grctx_init_gpm_0[] = { {} }; -const struct gf100_gr_pack -gk104_grctx_pack_gpc[] = { +static const struct gf100_gr_pack +gk104_grctx_pack_gpc_0[] = { { gf100_grctx_init_gpc_unk_0 }, { gf119_grctx_init_prop_0 }, { gf119_grctx_init_gpc_unk_1 }, { gk104_grctx_init_setup_0 }, { gf100_grctx_init_zcull_0 }, + {} +}; + +static const struct gf100_gr_pack +gk104_grctx_pack_gpc_1[] = { { gf119_grctx_init_crstr_0 }, { gk104_grctx_init_gpm_0 }, { gf100_grctx_init_gcc_0 }, @@ -841,6 +846,32 @@ gk104_grctx_pack_ppc[] = { ******************************************************************************/ void +gk104_grctx_generate_r418800(struct gf100_gr *gr) +{ + struct nvkm_device *device = gr->base.engine.subdev.device; + /*XXX: Not real sure where to apply these, there doesn't seem + * to be any pattern to which chipsets it's done on. + * + * Perhaps a VBIOS tweak? + */ + if (0) { + nvkm_mask(device, 0x418800, 0x00200000, 0x00200000); + nvkm_mask(device, 0x41be10, 0x00800000, 0x00800000); + } +} + +void +gk104_grctx_generate_patch_ltc(struct gf100_grctx *info) +{ + struct nvkm_device *device = info->gr->base.engine.subdev.device; + u32 data0 = nvkm_rd32(device, 0x17e91c); + u32 data1 = nvkm_rd32(device, 0x17e920); + /*XXX: Figure out how to modify this correctly! */ + mmio_wr32(info, 0x17e91c, data0); + mmio_wr32(info, 0x17e920, data1); +} + +void gk104_grctx_generate_bundle(struct gf100_grctx *info) { const struct gf100_grctx_func *grctx = info->gr->func->grctx; @@ -881,114 +912,74 @@ gk104_grctx_generate_unkn(struct gf100_gr *gr) nvkm_mask(device, 0x419c00, 0x00000008, 0x00000008); } -void -gk104_grctx_generate_r418bb8(struct gf100_gr *gr) +static void +gk104_grctx_generate_r419f78(struct gf100_gr *gr) { struct nvkm_device *device = gr->base.engine.subdev.device; - u32 data[6] = {}, data2[2] = {}; - u8 tpcnr[GPC_MAX]; - u8 shift, ntpcv; - int gpc, tpc, i; - - /* calculate first set of magics */ - memcpy(tpcnr, gr->tpc_nr, sizeof(gr->tpc_nr)); - - gpc = -1; - for (tpc = 0; tpc < gr->tpc_total; tpc++) { - do { - gpc = (gpc + 1) % gr->gpc_nr; - } while (!tpcnr[gpc]); - tpcnr[gpc]--; - - data[tpc / 6] |= gpc << ((tpc % 6) * 5); - } - - for (; tpc < 32; tpc++) - data[tpc / 6] |= 7 << ((tpc % 6) * 5); - - /* and the second... */ - shift = 0; - ntpcv = gr->tpc_total; - while (!(ntpcv & (1 << 4))) { - ntpcv <<= 1; - shift++; - } - - data2[0] = (ntpcv << 16); - data2[0] |= (shift << 21); - data2[0] |= (((1 << (0 + 5)) % ntpcv) << 24); - for (i = 1; i < 7; i++) - data2[1] |= ((1 << (i + 5)) % ntpcv) << ((i - 1) * 5); - - /* GPC_BROADCAST */ - nvkm_wr32(device, 0x418bb8, (gr->tpc_total << 8) | - gr->screen_tile_row_offset); - for (i = 0; i < 6; i++) - nvkm_wr32(device, 0x418b08 + (i * 4), data[i]); - - /* GPC_BROADCAST.TP_BROADCAST */ - nvkm_wr32(device, 0x41bfd0, (gr->tpc_total << 8) | - gr->screen_tile_row_offset | data2[0]); - nvkm_wr32(device, 0x41bfe4, data2[1]); - for (i = 0; i < 6; i++) - nvkm_wr32(device, 0x41bf00 + (i * 4), data[i]); - - /* UNK78xx */ - nvkm_wr32(device, 0x4078bc, (gr->tpc_total << 8) | - gr->screen_tile_row_offset); - for (i = 0; i < 6; i++) - nvkm_wr32(device, 0x40780c + (i * 4), data[i]); + nvkm_mask(device, 0x419f78, 0x00000001, 0x00000000); } void -gk104_grctx_generate_main(struct gf100_gr *gr, struct gf100_grctx *info) +gk104_grctx_generate_gpc_tpc_nr(struct gf100_gr *gr) { struct nvkm_device *device = gr->base.engine.subdev.device; - const struct gf100_grctx_func *grctx = gr->func->grctx; - u32 idle_timeout; - int i; - - nvkm_mc_unk260(device, 0); - - gf100_gr_mmio(gr, grctx->hub); - gf100_gr_mmio(gr, grctx->gpc); - gf100_gr_mmio(gr, grctx->zcull); - gf100_gr_mmio(gr, grctx->tpc); - gf100_gr_mmio(gr, grctx->ppc); - - idle_timeout = nvkm_mask(device, 0x404154, 0xffffffff, 0x00000000); - - grctx->bundle(info); - grctx->pagepool(info); - grctx->attrib(info); - grctx->unkn(gr); - - gf100_grctx_generate_tpcid(gr); - gf100_grctx_generate_r406028(gr); - gk104_grctx_generate_r418bb8(gr); - gf100_grctx_generate_r406800(gr); - - for (i = 0; i < 8; i++) - nvkm_wr32(device, 0x4064d0 + (i * 0x04), 0x00000000); - nvkm_wr32(device, 0x405b00, (gr->tpc_total << 8) | gr->gpc_nr); - nvkm_mask(device, 0x419f78, 0x00000001, 0x00000000); - - gf100_gr_icmd(gr, grctx->icmd); - nvkm_wr32(device, 0x404154, idle_timeout); - gf100_gr_mthd(gr, grctx->mthd); - nvkm_mc_unk260(device, 1); +} - nvkm_mask(device, 0x418800, 0x00200000, 0x00200000); - nvkm_mask(device, 0x41be10, 0x00800000, 0x00800000); +void +gk104_grctx_generate_alpha_beta_tables(struct gf100_gr *gr) +{ + struct nvkm_device *device = gr->base.engine.subdev.device; + int i, j, gpc, ppc; + + for (i = 0; i < 32; i++) { + u32 atarget = max_t(u32, gr->tpc_total * i / 32, 1); + u32 btarget = gr->tpc_total - atarget; + bool alpha = atarget < btarget; + u64 amask = 0, bmask = 0; + + for (gpc = 0; gpc < gr->gpc_nr; gpc++) { + for (ppc = 0; ppc < gr->func->ppc_nr; ppc++) { + u32 ppc_tpcs = gr->ppc_tpc_nr[gpc][ppc]; + u32 abits, bbits, pmask; + + if (alpha) { + abits = atarget ? ppc_tpcs : 0; + bbits = ppc_tpcs - abits; + } else { + bbits = btarget ? ppc_tpcs : 0; + abits = ppc_tpcs - bbits; + } + + pmask = gr->ppc_tpc_mask[gpc][ppc]; + while (ppc_tpcs-- > abits) + pmask &= pmask - 1; + amask |= (u64)pmask << (gpc * 8); + + pmask ^= gr->ppc_tpc_mask[gpc][ppc]; + bmask |= (u64)pmask << (gpc * 8); + + atarget -= min(abits, atarget); + btarget -= min(bbits, btarget); + if ((abits > 0) || (bbits > 0)) + alpha = !alpha; + } + } + + for (j = 0; j < gr->gpc_nr; j += 4, amask >>= 32, bmask >>= 32) { + nvkm_wr32(device, 0x406800 + (i * 0x20) + j, amask); + nvkm_wr32(device, 0x406c00 + (i * 0x20) + j, bmask); + } + } } const struct gf100_grctx_func gk104_grctx = { - .main = gk104_grctx_generate_main, + .main = gf100_grctx_generate_main, .unkn = gk104_grctx_generate_unkn, .hub = gk104_grctx_pack_hub, - .gpc = gk104_grctx_pack_gpc, + .gpc_0 = gk104_grctx_pack_gpc_0, + .gpc_1 = gk104_grctx_pack_gpc_1, .zcull = gf100_grctx_pack_zcull, .tpc = gk104_grctx_pack_tpc, .ppc = gk104_grctx_pack_ppc, @@ -1005,4 +996,13 @@ gk104_grctx = { .attrib_nr = 0x218, .alpha_nr_max = 0x7ff, .alpha_nr = 0x648, + .patch_ltc = gk104_grctx_generate_patch_ltc, + .sm_id = gf100_grctx_generate_sm_id, + .tpc_nr = gf100_grctx_generate_tpc_nr, + .rop_mapping = gf117_grctx_generate_rop_mapping, + .alpha_beta_tables = gk104_grctx_generate_alpha_beta_tables, + .dist_skip_table = gf117_grctx_generate_dist_skip_table, + .gpc_tpc_nr = gk104_grctx_generate_gpc_tpc_nr, + .r419f78 = gk104_grctx_generate_r419f78, + .r418800 = gk104_grctx_generate_r418800, }; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgk110.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgk110.c index 7b95ec2fe453..86547cfc38dc 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgk110.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgk110.c @@ -704,12 +704,17 @@ gk110_grctx_init_gpc_unk_2[] = { }; const struct gf100_gr_pack -gk110_grctx_pack_gpc[] = { +gk110_grctx_pack_gpc_0[] = { { gf100_grctx_init_gpc_unk_0 }, { gf119_grctx_init_prop_0 }, { gf119_grctx_init_gpc_unk_1 }, { gk110_grctx_init_setup_0 }, { gf100_grctx_init_zcull_0 }, + {} +}; + +const struct gf100_gr_pack +gk110_grctx_pack_gpc_1[] = { { gf119_grctx_init_crstr_0 }, { gk104_grctx_init_gpm_0 }, { gk110_grctx_init_gpc_unk_2 }, @@ -808,12 +813,20 @@ gk110_grctx_pack_ppc[] = { * PGRAPH context implementation ******************************************************************************/ +void +gk110_grctx_generate_r419eb0(struct gf100_gr *gr) +{ + struct nvkm_device *device = gr->base.engine.subdev.device; + nvkm_mask(device, 0x419eb0, 0x00001000, 0x00001000); +} + const struct gf100_grctx_func gk110_grctx = { - .main = gk104_grctx_generate_main, + .main = gf100_grctx_generate_main, .unkn = gk104_grctx_generate_unkn, .hub = gk110_grctx_pack_hub, - .gpc = gk110_grctx_pack_gpc, + .gpc_0 = gk110_grctx_pack_gpc_0, + .gpc_1 = gk110_grctx_pack_gpc_1, .zcull = gf100_grctx_pack_zcull, .tpc = gk110_grctx_pack_tpc, .ppc = gk110_grctx_pack_ppc, @@ -830,4 +843,13 @@ gk110_grctx = { .attrib_nr = 0x218, .alpha_nr_max = 0x7ff, .alpha_nr = 0x648, + .patch_ltc = gk104_grctx_generate_patch_ltc, + .sm_id = gf100_grctx_generate_sm_id, + .tpc_nr = gf100_grctx_generate_tpc_nr, + .rop_mapping = gf117_grctx_generate_rop_mapping, + .alpha_beta_tables = gk104_grctx_generate_alpha_beta_tables, + .dist_skip_table = gf117_grctx_generate_dist_skip_table, + .gpc_tpc_nr = gk104_grctx_generate_gpc_tpc_nr, + .r418800 = gk104_grctx_generate_r418800, + .r419eb0 = gk110_grctx_generate_r419eb0, }; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgk110b.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgk110b.c index 048b1152da44..ebb947bd1446 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgk110b.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgk110b.c @@ -71,10 +71,11 @@ gk110b_grctx_pack_tpc[] = { const struct gf100_grctx_func gk110b_grctx = { - .main = gk104_grctx_generate_main, + .main = gf100_grctx_generate_main, .unkn = gk104_grctx_generate_unkn, .hub = gk110_grctx_pack_hub, - .gpc = gk110_grctx_pack_gpc, + .gpc_0 = gk110_grctx_pack_gpc_0, + .gpc_1 = gk110_grctx_pack_gpc_1, .zcull = gf100_grctx_pack_zcull, .tpc = gk110b_grctx_pack_tpc, .ppc = gk110_grctx_pack_ppc, @@ -91,4 +92,13 @@ gk110b_grctx = { .attrib_nr = 0x218, .alpha_nr_max = 0x7ff, .alpha_nr = 0x648, + .patch_ltc = gk104_grctx_generate_patch_ltc, + .sm_id = gf100_grctx_generate_sm_id, + .tpc_nr = gf100_grctx_generate_tpc_nr, + .rop_mapping = gf117_grctx_generate_rop_mapping, + .alpha_beta_tables = gk104_grctx_generate_alpha_beta_tables, + .dist_skip_table = gf117_grctx_generate_dist_skip_table, + .gpc_tpc_nr = gk104_grctx_generate_gpc_tpc_nr, + .r418800 = gk104_grctx_generate_r418800, + .r419eb0 = gk110_grctx_generate_r419eb0, }; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgk208.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgk208.c index 67b7a1b43617..4d40512b5c99 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgk208.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgk208.c @@ -443,12 +443,17 @@ gk208_grctx_init_gpm_0[] = { }; static const struct gf100_gr_pack -gk208_grctx_pack_gpc[] = { +gk208_grctx_pack_gpc_0[] = { { gf100_grctx_init_gpc_unk_0 }, { gk208_grctx_init_prop_0 }, { gk208_grctx_init_gpc_unk_1 }, { gk208_grctx_init_setup_0 }, { gf100_grctx_init_zcull_0 }, + {} +}; + +static const struct gf100_gr_pack +gk208_grctx_pack_gpc_1[] = { { gk208_grctx_init_crstr_0 }, { gk208_grctx_init_gpm_0 }, { gk110_grctx_init_gpc_unk_2 }, @@ -532,10 +537,11 @@ gk208_grctx_pack_ppc[] = { const struct gf100_grctx_func gk208_grctx = { - .main = gk104_grctx_generate_main, + .main = gf100_grctx_generate_main, .unkn = gk104_grctx_generate_unkn, .hub = gk208_grctx_pack_hub, - .gpc = gk208_grctx_pack_gpc, + .gpc_0 = gk208_grctx_pack_gpc_0, + .gpc_1 = gk208_grctx_pack_gpc_1, .zcull = gf100_grctx_pack_zcull, .tpc = gk208_grctx_pack_tpc, .ppc = gk208_grctx_pack_ppc, @@ -552,4 +558,12 @@ gk208_grctx = { .attrib_nr = 0x218, .alpha_nr_max = 0x7ff, .alpha_nr = 0x648, + .patch_ltc = gk104_grctx_generate_patch_ltc, + .sm_id = gf100_grctx_generate_sm_id, + .tpc_nr = gf100_grctx_generate_tpc_nr, + .rop_mapping = gf117_grctx_generate_rop_mapping, + .alpha_beta_tables = gk104_grctx_generate_alpha_beta_tables, + .dist_skip_table = gf117_grctx_generate_dist_skip_table, + .gpc_tpc_nr = gk104_grctx_generate_gpc_tpc_nr, + .r418800 = gk104_grctx_generate_r418800, }; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgk20a.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgk20a.c index da7c35a6a3d2..896d473dcc0f 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgk20a.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgk20a.c @@ -42,10 +42,7 @@ gk20a_grctx_generate_main(struct gf100_gr *gr, struct gf100_grctx *info) grctx->unkn(gr); - gf100_grctx_generate_tpcid(gr); - gf100_grctx_generate_r406028(gr); - gk104_grctx_generate_r418bb8(gr); - gf100_grctx_generate_r406800(gr); + gf100_grctx_generate_floorsweep(gr); for (i = 0; i < 8; i++) nvkm_wr32(device, 0x4064d0 + (i * 0x04), 0x00000000); @@ -82,4 +79,8 @@ gk20a_grctx = { .attrib_nr = 0x240, .alpha_nr_max = 0x648 + (0x648 / 2), .alpha_nr = 0x648, + .sm_id = gf100_grctx_generate_sm_id, + .tpc_nr = gf100_grctx_generate_tpc_nr, + .rop_mapping = gf117_grctx_generate_rop_mapping, + .alpha_beta_tables = gk104_grctx_generate_alpha_beta_tables, }; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgm107.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgm107.c index 9b43d4ce3eaa..0b3964e6b36e 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgm107.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgm107.c @@ -744,12 +744,17 @@ gm107_grctx_init_gpc_unk_2[] = { }; static const struct gf100_gr_pack -gm107_grctx_pack_gpc[] = { +gm107_grctx_pack_gpc_0[] = { { gm107_grctx_init_gpc_unk_0 }, { gk208_grctx_init_prop_0 }, { gm107_grctx_init_gpc_unk_1 }, { gm107_grctx_init_setup_0 }, { gf100_grctx_init_zcull_0 }, + {} +}; + +static const struct gf100_gr_pack +gm107_grctx_pack_gpc_1[] = { { gk208_grctx_init_crstr_0 }, { gk104_grctx_init_gpm_0 }, { gm107_grctx_init_gpc_unk_2 }, @@ -860,6 +865,16 @@ gm107_grctx_pack_ppc[] = { * PGRAPH context implementation ******************************************************************************/ +static void +gm107_grctx_generate_r419e00(struct gf100_gr *gr) +{ + struct nvkm_device *device = gr->base.engine.subdev.device; + nvkm_mask(device, 0x419e00, 0x00808080, 0x00808080); + nvkm_mask(device, 0x419ccc, 0x80000000, 0x80000000); + nvkm_mask(device, 0x419f80, 0x80000000, 0x80000000); + nvkm_mask(device, 0x419f88, 0x80000000, 0x80000000); +} + void gm107_grctx_generate_bundle(struct gf100_grctx *info) { @@ -931,75 +946,27 @@ gm107_grctx_generate_attrib(struct gf100_grctx *info) } static void -gm107_grctx_generate_tpcid(struct gf100_gr *gr) +gm107_grctx_generate_r406500(struct gf100_gr *gr) { - struct nvkm_device *device = gr->base.engine.subdev.device; - int gpc, tpc, id; - - for (tpc = 0, id = 0; tpc < 4; tpc++) { - for (gpc = 0; gpc < gr->gpc_nr; gpc++) { - if (tpc < gr->tpc_nr[gpc]) { - nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x698), id); - nvkm_wr32(device, GPC_UNIT(gpc, 0x0c10 + tpc * 4), id); - nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x088), id); - id++; - } - - nvkm_wr32(device, GPC_UNIT(gpc, 0x0c08), gr->tpc_nr[gpc]); - nvkm_wr32(device, GPC_UNIT(gpc, 0x0c8c), gr->tpc_nr[gpc]); - } - } + nvkm_wr32(gr->base.engine.subdev.device, 0x406500, 0x00000001); } -static void -gm107_grctx_generate_main(struct gf100_gr *gr, struct gf100_grctx *info) +void +gm107_grctx_generate_sm_id(struct gf100_gr *gr, int gpc, int tpc, int sm) { struct nvkm_device *device = gr->base.engine.subdev.device; - const struct gf100_grctx_func *grctx = gr->func->grctx; - u32 idle_timeout; - int i; - - gf100_gr_mmio(gr, grctx->hub); - gf100_gr_mmio(gr, grctx->gpc); - gf100_gr_mmio(gr, grctx->zcull); - gf100_gr_mmio(gr, grctx->tpc); - gf100_gr_mmio(gr, grctx->ppc); - - idle_timeout = nvkm_mask(device, 0x404154, 0xffffffff, 0x00000000); - - grctx->bundle(info); - grctx->pagepool(info); - grctx->attrib(info); - grctx->unkn(gr); - - gm107_grctx_generate_tpcid(gr); - gf100_grctx_generate_r406028(gr); - gk104_grctx_generate_r418bb8(gr); - gf100_grctx_generate_r406800(gr); - - nvkm_wr32(device, 0x4064d0, 0x00000001); - for (i = 1; i < 8; i++) - nvkm_wr32(device, 0x4064d0 + (i * 0x04), 0x00000000); - nvkm_wr32(device, 0x406500, 0x00000001); - - nvkm_wr32(device, 0x405b00, (gr->tpc_total << 8) | gr->gpc_nr); - - gf100_gr_icmd(gr, grctx->icmd); - nvkm_wr32(device, 0x404154, idle_timeout); - gf100_gr_mthd(gr, grctx->mthd); - - nvkm_mask(device, 0x419e00, 0x00808080, 0x00808080); - nvkm_mask(device, 0x419ccc, 0x80000000, 0x80000000); - nvkm_mask(device, 0x419f80, 0x80000000, 0x80000000); - nvkm_mask(device, 0x419f88, 0x80000000, 0x80000000); + nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x698), sm); + nvkm_wr32(device, GPC_UNIT(gpc, 0x0c10 + tpc * 4), sm); + nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x088), sm); } const struct gf100_grctx_func gm107_grctx = { - .main = gm107_grctx_generate_main, + .main = gf100_grctx_generate_main, .unkn = gk104_grctx_generate_unkn, .hub = gm107_grctx_pack_hub, - .gpc = gm107_grctx_pack_gpc, + .gpc_0 = gm107_grctx_pack_gpc_0, + .gpc_1 = gm107_grctx_pack_gpc_1, .zcull = gf100_grctx_pack_zcull, .tpc = gm107_grctx_pack_tpc, .ppc = gm107_grctx_pack_ppc, @@ -1016,4 +983,12 @@ gm107_grctx = { .attrib_nr = 0xaa0, .alpha_nr_max = 0x1800, .alpha_nr = 0x1000, + .sm_id = gm107_grctx_generate_sm_id, + .tpc_nr = gf100_grctx_generate_tpc_nr, + .rop_mapping = gf117_grctx_generate_rop_mapping, + .alpha_beta_tables = gk104_grctx_generate_alpha_beta_tables, + .dist_skip_table = gf117_grctx_generate_dist_skip_table, + .r406500 = gm107_grctx_generate_r406500, + .gpc_tpc_nr = gk104_grctx_generate_gpc_tpc_nr, + .r419e00 = gm107_grctx_generate_r419e00, }; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgm200.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgm200.c index db209d33f486..013d05a0f0f6 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgm200.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgm200.c @@ -28,47 +28,34 @@ ******************************************************************************/ void -gm200_grctx_generate_tpcid(struct gf100_gr *gr) +gm200_grctx_generate_r419a3c(struct gf100_gr *gr) { struct nvkm_device *device = gr->base.engine.subdev.device; - int gpc, tpc, id; + nvkm_mask(device, 0x419a3c, 0x00000014, 0x00000000); +} - for (tpc = 0, id = 0; tpc < TPC_MAX_PER_GPC; tpc++) { - for (gpc = 0; gpc < gr->gpc_nr; gpc++) { - if (tpc < gr->tpc_nr[gpc]) { - nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x698), id); - nvkm_wr32(device, GPC_UNIT(gpc, 0x0c10 + tpc * 4), id); - nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x088), id); - id++; - } - } - } +static void +gm200_grctx_generate_r418e94(struct gf100_gr *gr) +{ + struct nvkm_device *device = gr->base.engine.subdev.device; + nvkm_mask(device, 0x418e94, 0xffffffff, 0xc4230000); + nvkm_mask(device, 0x418e4c, 0xffffffff, 0x70000000); } void -gm200_grctx_generate_405b60(struct gf100_gr *gr) +gm200_grctx_generate_smid_config(struct gf100_gr *gr) { struct nvkm_device *device = gr->base.engine.subdev.device; const u32 dist_nr = DIV_ROUND_UP(gr->tpc_total, 4); u32 dist[TPC_MAX / 4] = {}; u32 gpcs[GPC_MAX] = {}; - u8 tpcnr[GPC_MAX]; - int tpc, gpc, i; + u8 sm, i; - memcpy(tpcnr, gr->tpc_nr, sizeof(gr->tpc_nr)); - - /* won't result in the same distribution as the binary driver where - * some of the gpcs have more tpcs than others, but this shall do - * for the moment. the code for earlier gpus has this issue too. - */ - for (gpc = -1, i = 0; i < gr->tpc_total; i++) { - do { - gpc = (gpc + 1) % gr->gpc_nr; - } while(!tpcnr[gpc]); - tpc = gr->tpc_nr[gpc] - tpcnr[gpc]--; - - dist[i / 4] |= ((gpc << 4) | tpc) << ((i % 4) * 8); - gpcs[gpc] |= i << (tpc * 8); + for (sm = 0; sm < gr->sm_nr; sm++) { + const u8 gpc = gr->sm[sm].gpc; + const u8 tpc = gr->sm[sm].tpc; + dist[sm / 4] |= ((gpc << 4) | tpc) << ((sm % 4) * 8); + gpcs[gpc] |= sm << (tpc * 8); } for (i = 0; i < dist_nr; i++) @@ -77,50 +64,46 @@ gm200_grctx_generate_405b60(struct gf100_gr *gr) nvkm_wr32(device, 0x405ba0 + (i * 4), gpcs[i]); } -static void -gm200_grctx_generate_main(struct gf100_gr *gr, struct gf100_grctx *info) +void +gm200_grctx_generate_tpc_mask(struct gf100_gr *gr) { - struct nvkm_device *device = gr->base.engine.subdev.device; - const struct gf100_grctx_func *grctx = gr->func->grctx; - u32 idle_timeout, tmp; - int i; - - gf100_gr_mmio(gr, gr->fuc_sw_ctx); - - idle_timeout = nvkm_mask(device, 0x404154, 0xffffffff, 0x00000000); - - grctx->bundle(info); - grctx->pagepool(info); - grctx->attrib(info); - grctx->unkn(gr); - - gm200_grctx_generate_tpcid(gr); - gf100_grctx_generate_r406028(gr); - gk104_grctx_generate_r418bb8(gr); - - for (i = 0; i < 8; i++) - nvkm_wr32(device, 0x4064d0 + (i * 0x04), 0x00000000); - nvkm_wr32(device, 0x406500, 0x00000000); - - nvkm_wr32(device, 0x405b00, (gr->tpc_total << 8) | gr->gpc_nr); - + u32 tmp, i; for (tmp = 0, i = 0; i < gr->gpc_nr; i++) - tmp |= ((1 << gr->tpc_nr[i]) - 1) << (i * 4); - nvkm_wr32(device, 0x4041c4, tmp); + tmp |= ((1 << gr->tpc_nr[i]) - 1) << (i * gr->func->tpc_nr); + nvkm_wr32(gr->base.engine.subdev.device, 0x4041c4, tmp); +} - gm200_grctx_generate_405b60(gr); +void +gm200_grctx_generate_r406500(struct gf100_gr *gr) +{ + nvkm_wr32(gr->base.engine.subdev.device, 0x406500, 0x00000000); +} - gf100_gr_icmd(gr, gr->fuc_bundle); - nvkm_wr32(device, 0x404154, idle_timeout); - gf100_gr_mthd(gr, gr->fuc_method); +void +gm200_grctx_generate_dist_skip_table(struct gf100_gr *gr) +{ + struct nvkm_device *device = gr->base.engine.subdev.device; + u32 data[8] = {}; + int gpc, ppc, i; + + for (gpc = 0; gpc < gr->gpc_nr; gpc++) { + for (ppc = 0; ppc < gr->ppc_nr[gpc]; ppc++) { + u8 ppc_tpcs = gr->ppc_tpc_nr[gpc][ppc]; + u8 ppc_tpcm = gr->ppc_tpc_mask[gpc][ppc]; + while (ppc_tpcs-- > gr->ppc_tpc_min) + ppc_tpcm &= ppc_tpcm - 1; + ppc_tpcm ^= gr->ppc_tpc_mask[gpc][ppc]; + ((u8 *)data)[gpc] |= ppc_tpcm; + } + } - nvkm_mask(device, 0x418e94, 0xffffffff, 0xc4230000); - nvkm_mask(device, 0x418e4c, 0xffffffff, 0x70000000); + for (i = 0; i < ARRAY_SIZE(data); i++) + nvkm_wr32(device, 0x4064d0 + (i * 0x04), data[i]); } const struct gf100_grctx_func gm200_grctx = { - .main = gm200_grctx_generate_main, + .main = gf100_grctx_generate_main, .unkn = gk104_grctx_generate_unkn, .bundle = gm107_grctx_generate_bundle, .bundle_size = 0x3000, @@ -133,4 +116,13 @@ gm200_grctx = { .attrib_nr = 0x400, .alpha_nr_max = 0x1800, .alpha_nr = 0x1000, + .sm_id = gm107_grctx_generate_sm_id, + .rop_mapping = gf117_grctx_generate_rop_mapping, + .dist_skip_table = gm200_grctx_generate_dist_skip_table, + .r406500 = gm200_grctx_generate_r406500, + .gpc_tpc_nr = gk104_grctx_generate_gpc_tpc_nr, + .tpc_mask = gm200_grctx_generate_tpc_mask, + .smid_config = gm200_grctx_generate_smid_config, + .r418e94 = gm200_grctx_generate_r418e94, + .r419a3c = gm200_grctx_generate_r419a3c, }; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgm20b.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgm20b.c index e5702e3e0a5a..a1d9e114ebeb 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgm20b.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgm20b.c @@ -22,20 +22,6 @@ #include "ctxgf100.h" static void -gm20b_grctx_generate_r406028(struct gf100_gr *gr) -{ - struct nvkm_device *device = gr->base.engine.subdev.device; - u32 tpc_per_gpc = 0; - int i; - - for (i = 0; i < gr->gpc_nr; i++) - tpc_per_gpc |= gr->tpc_nr[i] << (4 * i); - - nvkm_wr32(device, 0x406028, tpc_per_gpc); - nvkm_wr32(device, 0x405870, tpc_per_gpc); -} - -static void gm20b_grctx_generate_main(struct gf100_gr *gr, struct gf100_grctx *info) { struct nvkm_device *device = gr->base.engine.subdev.device; @@ -53,9 +39,7 @@ gm20b_grctx_generate_main(struct gf100_gr *gr, struct gf100_grctx *info) grctx->unkn(gr); - gm200_grctx_generate_tpcid(gr); - gm20b_grctx_generate_r406028(gr); - gk104_grctx_generate_r418bb8(gr); + gf100_grctx_generate_floorsweep(gr); for (i = 0; i < 8; i++) nvkm_wr32(device, 0x4064d0 + (i * 0x04), 0x00000000); @@ -68,7 +52,7 @@ gm20b_grctx_generate_main(struct gf100_gr *gr, struct gf100_grctx *info) tmp |= ((1 << gr->tpc_nr[i]) - 1) << (i * 4); nvkm_wr32(device, 0x4041c4, tmp); - gm200_grctx_generate_405b60(gr); + gm200_grctx_generate_smid_config(gr); gf100_gr_wait_idle(gr); @@ -98,4 +82,6 @@ gm20b_grctx = { .attrib_nr = 0x400, .alpha_nr_max = 0xc00, .alpha_nr = 0x800, + .sm_id = gm107_grctx_generate_sm_id, + .rop_mapping = gf117_grctx_generate_rop_mapping, }; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgp100.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgp100.c index 88ea322d956c..0b3326262e12 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgp100.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgp100.c @@ -36,7 +36,7 @@ gp100_grctx_generate_pagepool(struct gf100_grctx *info) const int s = 8; const int b = mmio_vram(info, grctx->pagepool_size, (1 << s), true); mmio_refn(info, 0x40800c, 0x00000000, s, b); - mmio_wr32(info, 0x408010, 0x80000000); + mmio_wr32(info, 0x408010, 0x8007d800); mmio_refn(info, 0x419004, 0x00000000, s, b); mmio_wr32(info, 0x419008, 0x00000000); } @@ -48,14 +48,17 @@ gp100_grctx_generate_attrib(struct gf100_grctx *info) const struct gf100_grctx_func *grctx = gr->func->grctx; const u32 alpha = grctx->alpha_nr; const u32 attrib = grctx->attrib_nr; - const u32 pertpc = 0x20 * (grctx->attrib_nr_max + grctx->alpha_nr_max); - const u32 size = roundup(gr->tpc_total * pertpc, 0x80); const int s = 12; - const int b = mmio_vram(info, size, (1 << s), false); const int max_batches = 0xffff; + u32 size = grctx->alpha_nr_max * gr->tpc_total; u32 ao = 0; - u32 bo = ao + grctx->alpha_nr_max * gr->tpc_total; - int gpc, ppc, n = 0; + u32 bo = ao + size; + int gpc, ppc, b, n = 0; + + for (gpc = 0; gpc < gr->gpc_nr; gpc++) + size += grctx->attrib_nr_max * gr->ppc_nr[gpc] * gr->ppc_tpc_max; + size = ((size * 0x20) + 128) & ~127; + b = mmio_vram(info, size, (1 << s), false); mmio_refn(info, 0x418810, 0x80000000, s, b); mmio_refn(info, 0x419848, 0x10000000, s, b); @@ -69,7 +72,7 @@ gp100_grctx_generate_attrib(struct gf100_grctx *info) for (gpc = 0; gpc < gr->gpc_nr; gpc++) { for (ppc = 0; ppc < gr->ppc_nr[gpc]; ppc++, n++) { const u32 as = alpha * gr->ppc_tpc_nr[gpc][ppc]; - const u32 bs = attrib * gr->ppc_tpc_nr[gpc][ppc]; + const u32 bs = attrib * gr->ppc_tpc_max; const u32 u = 0x418ea0 + (n * 0x04); const u32 o = PPC_UNIT(gpc, ppc, 0); if (!(gr->ppc_mask[gpc] & (1 << ppc))) @@ -77,7 +80,7 @@ gp100_grctx_generate_attrib(struct gf100_grctx *info) mmio_wr32(info, o + 0xc0, bs); mmio_wr32(info, o + 0xf4, bo); mmio_wr32(info, o + 0xf0, bs); - bo += grctx->attrib_nr_max * gr->ppc_tpc_nr[gpc][ppc]; + bo += grctx->attrib_nr_max * gr->ppc_tpc_max; mmio_wr32(info, o + 0xe4, as); mmio_wr32(info, o + 0xf8, ao); ao += grctx->alpha_nr_max * gr->ppc_tpc_nr[gpc][ppc]; @@ -89,79 +92,30 @@ gp100_grctx_generate_attrib(struct gf100_grctx *info) mmio_wr32(info, 0x41befc, 0x00000000); } -static void -gp100_grctx_generate_405b60(struct gf100_gr *gr) +void +gp100_grctx_generate_smid_config(struct gf100_gr *gr) { struct nvkm_device *device = gr->base.engine.subdev.device; const u32 dist_nr = DIV_ROUND_UP(gr->tpc_total, 4); - u32 dist[TPC_MAX / 4] = {}; - u32 gpcs[GPC_MAX * 2] = {}; - u8 tpcnr[GPC_MAX]; - int tpc, gpc, i; - - memcpy(tpcnr, gr->tpc_nr, sizeof(gr->tpc_nr)); - - /* won't result in the same distribution as the binary driver where - * some of the gpcs have more tpcs than others, but this shall do - * for the moment. the code for earlier gpus has this issue too. - */ - for (gpc = -1, i = 0; i < gr->tpc_total; i++) { - do { - gpc = (gpc + 1) % gr->gpc_nr; - } while(!tpcnr[gpc]); - tpc = gr->tpc_nr[gpc] - tpcnr[gpc]--; - - dist[i / 4] |= ((gpc << 4) | tpc) << ((i % 4) * 8); - gpcs[gpc + (gr->gpc_nr * (tpc / 4))] |= i << (tpc * 8); + u32 dist[TPC_MAX / 4] = {}, gpcs[16] = {}; + u8 sm, i; + + for (sm = 0; sm < gr->sm_nr; sm++) { + const u8 gpc = gr->sm[sm].gpc; + const u8 tpc = gr->sm[sm].tpc; + dist[sm / 4] |= ((gpc << 4) | tpc) << ((sm % 4) * 8); + gpcs[gpc + (gr->func->gpc_nr * (tpc / 4))] |= sm << ((tpc % 4) * 8); } for (i = 0; i < dist_nr; i++) nvkm_wr32(device, 0x405b60 + (i * 4), dist[i]); - for (i = 0; i < gr->gpc_nr * 2; i++) + for (i = 0; i < ARRAY_SIZE(gpcs); i++) nvkm_wr32(device, 0x405ba0 + (i * 4), gpcs[i]); } -void -gp100_grctx_generate_main(struct gf100_gr *gr, struct gf100_grctx *info) -{ - struct nvkm_device *device = gr->base.engine.subdev.device; - const struct gf100_grctx_func *grctx = gr->func->grctx; - u32 idle_timeout, tmp; - int i; - - gf100_gr_mmio(gr, gr->fuc_sw_ctx); - - idle_timeout = nvkm_mask(device, 0x404154, 0xffffffff, 0x00000000); - - grctx->pagepool(info); - grctx->bundle(info); - grctx->attrib(info); - grctx->unkn(gr); - - gm200_grctx_generate_tpcid(gr); - gf100_grctx_generate_r406028(gr); - gk104_grctx_generate_r418bb8(gr); - - for (i = 0; i < 8; i++) - nvkm_wr32(device, 0x4064d0 + (i * 0x04), 0x00000000); - nvkm_wr32(device, 0x406500, 0x00000000); - - nvkm_wr32(device, 0x405b00, (gr->tpc_total << 8) | gr->gpc_nr); - - for (tmp = 0, i = 0; i < gr->gpc_nr; i++) - tmp |= ((1 << gr->tpc_nr[i]) - 1) << (i * 5); - nvkm_wr32(device, 0x4041c4, tmp); - - gp100_grctx_generate_405b60(gr); - - gf100_gr_icmd(gr, gr->fuc_bundle); - nvkm_wr32(device, 0x404154, idle_timeout); - gf100_gr_mthd(gr, gr->fuc_method); -} - const struct gf100_grctx_func gp100_grctx = { - .main = gp100_grctx_generate_main, + .main = gf100_grctx_generate_main, .unkn = gk104_grctx_generate_unkn, .bundle = gm107_grctx_generate_bundle, .bundle_size = 0x3000, @@ -174,4 +128,12 @@ gp100_grctx = { .attrib_nr = 0x440, .alpha_nr_max = 0xc00, .alpha_nr = 0x800, + .sm_id = gm107_grctx_generate_sm_id, + .rop_mapping = gf117_grctx_generate_rop_mapping, + .dist_skip_table = gm200_grctx_generate_dist_skip_table, + .r406500 = gm200_grctx_generate_r406500, + .gpc_tpc_nr = gk104_grctx_generate_gpc_tpc_nr, + .tpc_mask = gm200_grctx_generate_tpc_mask, + .smid_config = gp100_grctx_generate_smid_config, + .r419a3c = gm200_grctx_generate_r419a3c, }; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgp102.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgp102.c index 7a66b4c2eb18..daee17bf7d0d 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgp102.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgp102.c @@ -29,6 +29,13 @@ * PGRAPH context implementation ******************************************************************************/ +static void +gp102_grctx_generate_r408840(struct gf100_gr *gr) +{ + struct nvkm_device *device = gr->base.engine.subdev.device; + nvkm_mask(device, 0x408840, 0x00000003, 0x00000000); +} + void gp102_grctx_generate_attrib(struct gf100_grctx *info) { @@ -36,14 +43,18 @@ gp102_grctx_generate_attrib(struct gf100_grctx *info) const struct gf100_grctx_func *grctx = gr->func->grctx; const u32 alpha = grctx->alpha_nr; const u32 attrib = grctx->attrib_nr; - const u32 pertpc = 0x20 * (grctx->attrib_nr_max + grctx->alpha_nr_max); - const u32 size = roundup(gr->tpc_total * pertpc, 0x80); + const u32 gfxp = grctx->gfxp_nr; const int s = 12; - const int b = mmio_vram(info, size, (1 << s), false); const int max_batches = 0xffff; + u32 size = grctx->alpha_nr_max * gr->tpc_total; u32 ao = 0; - u32 bo = ao + grctx->alpha_nr_max * gr->tpc_total; - int gpc, ppc, n = 0; + u32 bo = ao + size; + int gpc, ppc, b, n = 0; + + for (gpc = 0; gpc < gr->gpc_nr; gpc++) + size += grctx->gfxp_nr * gr->ppc_nr[gpc] * gr->ppc_tpc_max; + size = ((size * 0x20) + 128) & ~127; + b = mmio_vram(info, size, (1 << s), false); mmio_refn(info, 0x418810, 0x80000000, s, b); mmio_refn(info, 0x419848, 0x10000000, s, b); @@ -57,17 +68,18 @@ gp102_grctx_generate_attrib(struct gf100_grctx *info) for (gpc = 0; gpc < gr->gpc_nr; gpc++) { for (ppc = 0; ppc < gr->ppc_nr[gpc]; ppc++, n++) { const u32 as = alpha * gr->ppc_tpc_nr[gpc][ppc]; - const u32 bs = attrib * gr->ppc_tpc_nr[gpc][ppc]; + const u32 bs = attrib * gr->ppc_tpc_max; + const u32 gs = gfxp * gr->ppc_tpc_max; const u32 u = 0x418ea0 + (n * 0x04); const u32 o = PPC_UNIT(gpc, ppc, 0); const u32 p = GPC_UNIT(gpc, 0xc44 + (ppc * 4)); if (!(gr->ppc_mask[gpc] & (1 << ppc))) continue; - mmio_wr32(info, o + 0xc0, bs); + mmio_wr32(info, o + 0xc0, gs); mmio_wr32(info, p, bs); mmio_wr32(info, o + 0xf4, bo); mmio_wr32(info, o + 0xf0, bs); - bo += grctx->attrib_nr_max * gr->ppc_tpc_nr[gpc][ppc]; + bo += gs; mmio_wr32(info, o + 0xe4, as); mmio_wr32(info, o + 0xf8, ao); ao += grctx->alpha_nr_max * gr->ppc_tpc_nr[gpc][ppc]; @@ -81,7 +93,7 @@ gp102_grctx_generate_attrib(struct gf100_grctx *info) const struct gf100_grctx_func gp102_grctx = { - .main = gp100_grctx_generate_main, + .main = gf100_grctx_generate_main, .unkn = gk104_grctx_generate_unkn, .bundle = gm107_grctx_generate_bundle, .bundle_size = 0x3000, @@ -90,8 +102,18 @@ gp102_grctx = { .pagepool = gp100_grctx_generate_pagepool, .pagepool_size = 0x20000, .attrib = gp102_grctx_generate_attrib, - .attrib_nr_max = 0x5d4, + .attrib_nr_max = 0x4b0, .attrib_nr = 0x320, .alpha_nr_max = 0xc00, .alpha_nr = 0x800, + .gfxp_nr = 0xba8, + .sm_id = gm107_grctx_generate_sm_id, + .rop_mapping = gf117_grctx_generate_rop_mapping, + .dist_skip_table = gm200_grctx_generate_dist_skip_table, + .r406500 = gm200_grctx_generate_r406500, + .gpc_tpc_nr = gk104_grctx_generate_gpc_tpc_nr, + .tpc_mask = gm200_grctx_generate_tpc_mask, + .smid_config = gp100_grctx_generate_smid_config, + .r419a3c = gm200_grctx_generate_r419a3c, + .r408840 = gp102_grctx_generate_r408840, }; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgp104.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgp104.c new file mode 100644 index 000000000000..3b85e3d326b2 --- /dev/null +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgp104.c @@ -0,0 +1,48 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "ctxgf100.h" + +const struct gf100_grctx_func +gp104_grctx = { + .main = gf100_grctx_generate_main, + .unkn = gk104_grctx_generate_unkn, + .bundle = gm107_grctx_generate_bundle, + .bundle_size = 0x3000, + .bundle_min_gpm_fifo_depth = 0x180, + .bundle_token_limit = 0x900, + .pagepool = gp100_grctx_generate_pagepool, + .pagepool_size = 0x20000, + .attrib = gp102_grctx_generate_attrib, + .attrib_nr_max = 0x4b0, + .attrib_nr = 0x320, + .alpha_nr_max = 0xc00, + .alpha_nr = 0x800, + .gfxp_nr = 0xba8, + .sm_id = gm107_grctx_generate_sm_id, + .rop_mapping = gf117_grctx_generate_rop_mapping, + .dist_skip_table = gm200_grctx_generate_dist_skip_table, + .r406500 = gm200_grctx_generate_r406500, + .gpc_tpc_nr = gk104_grctx_generate_gpc_tpc_nr, + .tpc_mask = gm200_grctx_generate_tpc_mask, + .smid_config = gp100_grctx_generate_smid_config, + .r419a3c = gm200_grctx_generate_r419a3c, +}; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgp107.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgp107.c index 8da91a0b3bd2..5060c5ee5ce0 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgp107.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgp107.c @@ -31,7 +31,7 @@ const struct gf100_grctx_func gp107_grctx = { - .main = gp100_grctx_generate_main, + .main = gf100_grctx_generate_main, .unkn = gk104_grctx_generate_unkn, .bundle = gm107_grctx_generate_bundle, .bundle_size = 0x3000, @@ -44,4 +44,13 @@ gp107_grctx = { .attrib_nr = 0x540, .alpha_nr_max = 0xc00, .alpha_nr = 0x800, + .gfxp_nr = 0xe94, + .sm_id = gm107_grctx_generate_sm_id, + .rop_mapping = gf117_grctx_generate_rop_mapping, + .dist_skip_table = gm200_grctx_generate_dist_skip_table, + .r406500 = gm200_grctx_generate_r406500, + .gpc_tpc_nr = gk104_grctx_generate_gpc_tpc_nr, + .tpc_mask = gm200_grctx_generate_tpc_mask, + .smid_config = gp100_grctx_generate_smid_config, + .r419a3c = gm200_grctx_generate_r419a3c, }; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgv100.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgv100.c new file mode 100644 index 000000000000..0990765ef191 --- /dev/null +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgv100.c @@ -0,0 +1,215 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "ctxgf100.h" + +/******************************************************************************* + * PGRAPH context implementation + ******************************************************************************/ + +static const struct gf100_gr_init +gv100_grctx_init_sw_veid_bundle_init_0[] = { + { 0x00001000, 64, 0x00100000, 0x00000008 }, + { 0x00000941, 64, 0x00100000, 0x00000000 }, + { 0x0000097e, 64, 0x00100000, 0x00000000 }, + { 0x0000097f, 64, 0x00100000, 0x00000100 }, + { 0x0000035c, 64, 0x00100000, 0x00000000 }, + { 0x0000035d, 64, 0x00100000, 0x00000000 }, + { 0x00000a08, 64, 0x00100000, 0x00000000 }, + { 0x00000a09, 64, 0x00100000, 0x00000000 }, + { 0x00000a0a, 64, 0x00100000, 0x00000000 }, + { 0x00000352, 64, 0x00100000, 0x00000000 }, + { 0x00000353, 64, 0x00100000, 0x00000000 }, + { 0x00000358, 64, 0x00100000, 0x00000000 }, + { 0x00000359, 64, 0x00100000, 0x00000000 }, + { 0x00000370, 64, 0x00100000, 0x00000000 }, + { 0x00000371, 64, 0x00100000, 0x00000000 }, + { 0x00000372, 64, 0x00100000, 0x000fffff }, + { 0x00000366, 64, 0x00100000, 0x00000000 }, + { 0x00000367, 64, 0x00100000, 0x00000000 }, + { 0x00000368, 64, 0x00100000, 0x00000fff }, + { 0x00000623, 64, 0x00100000, 0x00000000 }, + { 0x00000624, 64, 0x00100000, 0x00000000 }, + { 0x0001e100, 1, 0x00000001, 0x02000001 }, + {} +}; + +static const struct gf100_gr_pack +gv100_grctx_pack_sw_veid_bundle_init[] = { + { gv100_grctx_init_sw_veid_bundle_init_0 }, + {} +}; + +static void +gv100_grctx_generate_attrib(struct gf100_grctx *info) +{ + struct gf100_gr *gr = info->gr; + const struct gf100_grctx_func *grctx = gr->func->grctx; + const u32 alpha = grctx->alpha_nr; + const u32 attrib = grctx->attrib_nr; + const u32 gfxp = grctx->gfxp_nr; + const int s = 12; + const int max_batches = 0xffff; + u32 size = grctx->alpha_nr_max * gr->tpc_total; + u32 ao = 0; + u32 bo = ao + size; + int gpc, ppc, b, n = 0; + + size += grctx->gfxp_nr * gr->tpc_total; + size = ((size * 0x20) + 128) & ~127; + b = mmio_vram(info, size, (1 << s), false); + + mmio_refn(info, 0x418810, 0x80000000, s, b); + mmio_refn(info, 0x419848, 0x10000000, s, b); + mmio_refn(info, 0x419c2c, 0x10000000, s, b); + mmio_refn(info, 0x419e00, 0x00000000, s, b); + mmio_wr32(info, 0x419e04, 0x80000000 | size >> 7); + mmio_wr32(info, 0x405830, attrib); + mmio_wr32(info, 0x40585c, alpha); + mmio_wr32(info, 0x4064c4, ((alpha / 4) << 16) | max_batches); + + for (gpc = 0; gpc < gr->gpc_nr; gpc++) { + for (ppc = 0; ppc < gr->ppc_nr[gpc]; ppc++, n++) { + const u32 as = alpha * gr->ppc_tpc_nr[gpc][ppc]; + const u32 bs = attrib * gr->ppc_tpc_nr[gpc][ppc]; + const u32 gs = gfxp * gr->ppc_tpc_nr[gpc][ppc]; + const u32 u = 0x418ea0 + (n * 0x04); + const u32 o = PPC_UNIT(gpc, ppc, 0); + if (!(gr->ppc_mask[gpc] & (1 << ppc))) + continue; + mmio_wr32(info, o + 0xc0, gs); + mmio_wr32(info, o + 0xf4, bo); + mmio_wr32(info, o + 0xf0, bs); + bo += gs; + mmio_wr32(info, o + 0xe4, as); + mmio_wr32(info, o + 0xf8, ao); + ao += grctx->alpha_nr_max * gr->ppc_tpc_nr[gpc][ppc]; + mmio_wr32(info, u, bs); + } + } + + mmio_wr32(info, 0x4181e4, 0x00000100); + mmio_wr32(info, 0x41befc, 0x00000100); +} + +static void +gv100_grctx_generate_rop_mapping(struct gf100_gr *gr) +{ + struct nvkm_device *device = gr->base.engine.subdev.device; + u32 data; + int i, j; + + /* Pack tile map into register format. */ + nvkm_wr32(device, 0x418bb8, (gr->tpc_total << 8) | + gr->screen_tile_row_offset); + for (i = 0; i < 11; i++) { + for (data = 0, j = 0; j < 6; j++) + data |= (gr->tile[i * 6 + j] & 0x1f) << (j * 5); + nvkm_wr32(device, 0x418b08 + (i * 4), data); + nvkm_wr32(device, 0x41bf00 + (i * 4), data); + nvkm_wr32(device, 0x40780c + (i * 4), data); + } + + /* GPC_BROADCAST.TP_BROADCAST */ + nvkm_wr32(device, 0x41bfd0, (gr->tpc_total << 8) | + gr->screen_tile_row_offset); + for (i = 0, j = 1; i < 5; i++, j += 4) { + u8 v19 = (1 << (j + 0)) % gr->tpc_total; + u8 v20 = (1 << (j + 1)) % gr->tpc_total; + u8 v21 = (1 << (j + 2)) % gr->tpc_total; + u8 v22 = (1 << (j + 3)) % gr->tpc_total; + nvkm_wr32(device, 0x41bfb0 + (i * 4), (v22 << 24) | + (v21 << 16) | + (v20 << 8) | + v19); + } + + /* UNK78xx */ + nvkm_wr32(device, 0x4078bc, (gr->tpc_total << 8) | + gr->screen_tile_row_offset); +} + +static void +gv100_grctx_generate_r400088(struct gf100_gr *gr, bool on) +{ + struct nvkm_device *device = gr->base.engine.subdev.device; + nvkm_mask(device, 0x400088, 0x00060000, on ? 0x00060000 : 0x00000000); +} + +static void +gv100_grctx_generate_sm_id(struct gf100_gr *gr, int gpc, int tpc, int sm) +{ + struct nvkm_device *device = gr->base.engine.subdev.device; + nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x608), sm); + nvkm_wr32(device, GPC_UNIT(gpc, 0x0c10 + tpc * 4), sm); + nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x088), sm); +} + +static void +gv100_grctx_generate_unkn(struct gf100_gr *gr) +{ + struct nvkm_device *device = gr->base.engine.subdev.device; + nvkm_mask(device, 0x41980c, 0x00000010, 0x00000010); + nvkm_mask(device, 0x41be08, 0x00000004, 0x00000004); + nvkm_mask(device, 0x4064c0, 0x80000000, 0x80000000); + nvkm_mask(device, 0x405800, 0x08000000, 0x08000000); + nvkm_mask(device, 0x419c00, 0x00000008, 0x00000008); +} + +static void +gv100_grctx_unkn88c(struct gf100_gr *gr, bool on) +{ + struct nvkm_device *device = gr->base.engine.subdev.device; + const u32 mask = 0x00000010, data = on ? mask : 0x00000000; + nvkm_mask(device, 0x40988c, mask, data); + nvkm_rd32(device, 0x40988c); + nvkm_mask(device, 0x41a88c, mask, data); + nvkm_rd32(device, 0x41a88c); + nvkm_mask(device, 0x408a14, mask, data); + nvkm_rd32(device, 0x408a14); +} + +const struct gf100_grctx_func +gv100_grctx = { + .unkn88c = gv100_grctx_unkn88c, + .main = gf100_grctx_generate_main, + .unkn = gv100_grctx_generate_unkn, + .sw_veid_bundle_init = gv100_grctx_pack_sw_veid_bundle_init, + .bundle = gm107_grctx_generate_bundle, + .bundle_size = 0x3000, + .bundle_min_gpm_fifo_depth = 0x180, + .bundle_token_limit = 0x1680, + .pagepool = gp100_grctx_generate_pagepool, + .pagepool_size = 0x20000, + .attrib = gv100_grctx_generate_attrib, + .attrib_nr_max = 0x6c0, + .attrib_nr = 0x480, + .alpha_nr_max = 0xc00, + .alpha_nr = 0x800, + .gfxp_nr = 0xd10, + .sm_id = gv100_grctx_generate_sm_id, + .rop_mapping = gv100_grctx_generate_rop_mapping, + .dist_skip_table = gm200_grctx_generate_dist_skip_table, + .r406500 = gm200_grctx_generate_r406500, + .gpc_tpc_nr = gk104_grctx_generate_gpc_tpc_nr, + .smid_config = gp100_grctx_generate_smid_config, + .r400088 = gv100_grctx_generate_r400088, +}; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.c index 2f8dc107047d..70d3d41e616c 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.c @@ -32,6 +32,7 @@ #include <subdev/fb.h> #include <subdev/mc.h> #include <subdev/pmu.h> +#include <subdev/therm.h> #include <subdev/timer.h> #include <engine/fifo.h> @@ -91,7 +92,7 @@ gf100_gr_zbc_color_get(struct gf100_gr *gr, int format, memcpy(gr->zbc_color[zbc].l2, l2, sizeof(gr->zbc_color[zbc].l2)); gr->zbc_color[zbc].format = format; nvkm_ltc_zbc_color_get(ltc, zbc, l2); - gf100_gr_zbc_clear_color(gr, zbc); + gr->func->zbc->clear_color(gr, zbc); return zbc; } @@ -136,10 +137,16 @@ gf100_gr_zbc_depth_get(struct gf100_gr *gr, int format, gr->zbc_depth[zbc].ds = ds; gr->zbc_depth[zbc].l2 = l2; nvkm_ltc_zbc_depth_get(ltc, zbc, l2); - gf100_gr_zbc_clear_depth(gr, zbc); + gr->func->zbc->clear_depth(gr, zbc); return zbc; } +const struct gf100_gr_func_zbc +gf100_gr_zbc = { + .clear_color = gf100_gr_zbc_clear_color, + .clear_depth = gf100_gr_zbc_clear_depth, +}; + /******************************************************************************* * Graphics object classes ******************************************************************************/ @@ -743,21 +750,31 @@ gf100_gr_zbc_init(struct gf100_gr *gr) const u32 f32_1[] = { 0x3f800000, 0x3f800000, 0x3f800000, 0x3f800000, 0x3f800000, 0x3f800000, 0x3f800000, 0x3f800000 }; struct nvkm_ltc *ltc = gr->base.engine.subdev.device->ltc; - int index; + int index, c = ltc->zbc_min, d = ltc->zbc_min, s = ltc->zbc_min; if (!gr->zbc_color[0].format) { - gf100_gr_zbc_color_get(gr, 1, & zero[0], &zero[4]); - gf100_gr_zbc_color_get(gr, 2, & one[0], &one[4]); - gf100_gr_zbc_color_get(gr, 4, &f32_0[0], &f32_0[4]); - gf100_gr_zbc_color_get(gr, 4, &f32_1[0], &f32_1[4]); - gf100_gr_zbc_depth_get(gr, 1, 0x00000000, 0x00000000); - gf100_gr_zbc_depth_get(gr, 1, 0x3f800000, 0x3f800000); - } - - for (index = ltc->zbc_min; index <= ltc->zbc_max; index++) - gf100_gr_zbc_clear_color(gr, index); - for (index = ltc->zbc_min; index <= ltc->zbc_max; index++) - gf100_gr_zbc_clear_depth(gr, index); + gf100_gr_zbc_color_get(gr, 1, & zero[0], &zero[4]); c++; + gf100_gr_zbc_color_get(gr, 2, & one[0], &one[4]); c++; + gf100_gr_zbc_color_get(gr, 4, &f32_0[0], &f32_0[4]); c++; + gf100_gr_zbc_color_get(gr, 4, &f32_1[0], &f32_1[4]); c++; + gf100_gr_zbc_depth_get(gr, 1, 0x00000000, 0x00000000); d++; + gf100_gr_zbc_depth_get(gr, 1, 0x3f800000, 0x3f800000); d++; + if (gr->func->zbc->stencil_get) { + gr->func->zbc->stencil_get(gr, 1, 0x00, 0x00); s++; + gr->func->zbc->stencil_get(gr, 1, 0x01, 0x01); s++; + gr->func->zbc->stencil_get(gr, 1, 0xff, 0xff); s++; + } + } + + for (index = c; index <= ltc->zbc_max; index++) + gr->func->zbc->clear_color(gr, index); + for (index = d; index <= ltc->zbc_max; index++) + gr->func->zbc->clear_depth(gr, index); + + if (gr->func->zbc->clear_stencil) { + for (index = s; index <= ltc->zbc_max; index++) + gr->func->zbc->clear_stencil(gr, index); + } } /** @@ -970,7 +987,7 @@ gf100_gr_trap_gpc_rop(struct gf100_gr *gr, int gpc) nvkm_wr32(device, GPC_UNIT(gpc, 0x0420), 0xc0000000); } -static const struct nvkm_enum gf100_mp_warp_error[] = { +const struct nvkm_enum gf100_mp_warp_error[] = { { 0x01, "STACK_ERROR" }, { 0x02, "API_STACK_ERROR" }, { 0x03, "RET_EMPTY_STACK_ERROR" }, @@ -995,7 +1012,7 @@ static const struct nvkm_enum gf100_mp_warp_error[] = { {} }; -static const struct nvkm_bitfield gf100_mp_global_error[] = { +const struct nvkm_bitfield gf100_mp_global_error[] = { { 0x00000001, "SM_TO_SM_FAULT" }, { 0x00000002, "L1_ERROR" }, { 0x00000004, "MULTIPLE_WARP_ERRORS" }, @@ -1009,7 +1026,7 @@ static const struct nvkm_bitfield gf100_mp_global_error[] = { {} }; -static void +void gf100_gr_trap_mp(struct gf100_gr *gr, int gpc, int tpc) { struct nvkm_subdev *subdev = &gr->base.engine.subdev; @@ -1045,7 +1062,7 @@ gf100_gr_trap_tpc(struct gf100_gr *gr, int gpc, int tpc) } if (stat & 0x00000002) { - gf100_gr_trap_mp(gr, gpc, tpc); + gr->func->trap_mp(gr, gpc, tpc); stat &= ~0x00000002; } @@ -1611,7 +1628,8 @@ gf100_gr_init_ctxctl_int(struct gf100_gr *gr) /* load register lists */ gf100_gr_init_csdata(gr, grctx->hub, 0x409000, 0x000, 0x000000); - gf100_gr_init_csdata(gr, grctx->gpc, 0x41a000, 0x000, 0x418000); + gf100_gr_init_csdata(gr, grctx->gpc_0, 0x41a000, 0x000, 0x418000); + gf100_gr_init_csdata(gr, grctx->gpc_1, 0x41a000, 0x000, 0x418000); gf100_gr_init_csdata(gr, grctx->tpc, 0x41a000, 0x004, 0x419800); gf100_gr_init_csdata(gr, grctx->ppc, 0x41a000, 0x008, 0x41be00); @@ -1651,6 +1669,97 @@ gf100_gr_init_ctxctl(struct gf100_gr *gr) return ret; } +void +gf100_gr_oneinit_sm_id(struct gf100_gr *gr) +{ + int tpc, gpc; + for (tpc = 0; tpc < gr->tpc_max; tpc++) { + for (gpc = 0; gpc < gr->gpc_nr; gpc++) { + if (tpc < gr->tpc_nr[gpc]) { + gr->sm[gr->sm_nr].gpc = gpc; + gr->sm[gr->sm_nr].tpc = tpc; + gr->sm_nr++; + } + } + } +} + +void +gf100_gr_oneinit_tiles(struct gf100_gr *gr) +{ + static const u8 primes[] = { + 3, 5, 7, 11, 13, 17, 19, 23, 29, 31, 37, 41, 43, 47, 53, 59, 61 + }; + int init_frac[GPC_MAX], init_err[GPC_MAX], run_err[GPC_MAX], i, j; + u32 mul_factor, comm_denom; + u8 gpc_map[GPC_MAX]; + bool sorted; + + switch (gr->tpc_total) { + case 15: gr->screen_tile_row_offset = 0x06; break; + case 14: gr->screen_tile_row_offset = 0x05; break; + case 13: gr->screen_tile_row_offset = 0x02; break; + case 11: gr->screen_tile_row_offset = 0x07; break; + case 10: gr->screen_tile_row_offset = 0x06; break; + case 7: + case 5: gr->screen_tile_row_offset = 0x01; break; + case 3: gr->screen_tile_row_offset = 0x02; break; + case 2: + case 1: gr->screen_tile_row_offset = 0x01; break; + default: gr->screen_tile_row_offset = 0x03; + for (i = 0; i < ARRAY_SIZE(primes); i++) { + if (gr->tpc_total % primes[i]) { + gr->screen_tile_row_offset = primes[i]; + break; + } + } + break; + } + + /* Sort GPCs by TPC count, highest-to-lowest. */ + for (i = 0; i < gr->gpc_nr; i++) + gpc_map[i] = i; + sorted = false; + + while (!sorted) { + for (sorted = true, i = 0; i < gr->gpc_nr - 1; i++) { + if (gr->tpc_nr[gpc_map[i + 1]] > + gr->tpc_nr[gpc_map[i + 0]]) { + u8 swap = gpc_map[i]; + gpc_map[i + 0] = gpc_map[i + 1]; + gpc_map[i + 1] = swap; + sorted = false; + } + } + } + + /* Determine tile->GPC mapping */ + mul_factor = gr->gpc_nr * gr->tpc_max; + if (mul_factor & 1) + mul_factor = 2; + else + mul_factor = 1; + + comm_denom = gr->gpc_nr * gr->tpc_max * mul_factor; + + for (i = 0; i < gr->gpc_nr; i++) { + init_frac[i] = gr->tpc_nr[gpc_map[i]] * gr->gpc_nr * mul_factor; + init_err[i] = i * gr->tpc_max * mul_factor - comm_denom/2; + run_err[i] = init_frac[i] + init_err[i]; + } + + for (i = 0; i < gr->tpc_total;) { + for (j = 0; j < gr->gpc_nr; j++) { + if ((run_err[j] * 2) >= comm_denom) { + gr->tile[i++] = gpc_map[j]; + run_err[j] += init_frac[j] - comm_denom; + } else { + run_err[j] += init_frac[j]; + } + } + } +} + static int gf100_gr_oneinit(struct nvkm_gr *base) { @@ -1674,55 +1783,27 @@ gf100_gr_oneinit(struct nvkm_gr *base) gr->gpc_nr = nvkm_rd32(device, 0x409604) & 0x0000001f; for (i = 0; i < gr->gpc_nr; i++) { gr->tpc_nr[i] = nvkm_rd32(device, GPC_UNIT(i, 0x2608)); + gr->tpc_max = max(gr->tpc_max, gr->tpc_nr[i]); gr->tpc_total += gr->tpc_nr[i]; gr->ppc_nr[i] = gr->func->ppc_nr; for (j = 0; j < gr->ppc_nr[i]; j++) { - u8 mask = nvkm_rd32(device, GPC_UNIT(i, 0x0c30 + (j * 4))); - if (mask) - gr->ppc_mask[i] |= (1 << j); - gr->ppc_tpc_nr[i][j] = hweight8(mask); - } - } - - /*XXX: these need figuring out... though it might not even matter */ - switch (device->chipset) { - case 0xc0: - if (gr->tpc_total == 11) { /* 465, 3/4/4/0, 4 */ - gr->screen_tile_row_offset = 0x07; - } else - if (gr->tpc_total == 14) { /* 470, 3/3/4/4, 5 */ - gr->screen_tile_row_offset = 0x05; - } else - if (gr->tpc_total == 15) { /* 480, 3/4/4/4, 6 */ - gr->screen_tile_row_offset = 0x06; + gr->ppc_tpc_mask[i][j] = + nvkm_rd32(device, GPC_UNIT(i, 0x0c30 + (j * 4))); + if (gr->ppc_tpc_mask[i][j] == 0) + continue; + gr->ppc_mask[i] |= (1 << j); + gr->ppc_tpc_nr[i][j] = hweight8(gr->ppc_tpc_mask[i][j]); + if (gr->ppc_tpc_min == 0 || + gr->ppc_tpc_min > gr->ppc_tpc_nr[i][j]) + gr->ppc_tpc_min = gr->ppc_tpc_nr[i][j]; + if (gr->ppc_tpc_max < gr->ppc_tpc_nr[i][j]) + gr->ppc_tpc_max = gr->ppc_tpc_nr[i][j]; } - break; - case 0xc3: /* 450, 4/0/0/0, 2 */ - gr->screen_tile_row_offset = 0x03; - break; - case 0xc4: /* 460, 3/4/0/0, 4 */ - gr->screen_tile_row_offset = 0x01; - break; - case 0xc1: /* 2/0/0/0, 1 */ - gr->screen_tile_row_offset = 0x01; - break; - case 0xc8: /* 4/4/3/4, 5 */ - gr->screen_tile_row_offset = 0x06; - break; - case 0xce: /* 4/4/0/0, 4 */ - gr->screen_tile_row_offset = 0x03; - break; - case 0xcf: /* 4/0/0/0, 3 */ - gr->screen_tile_row_offset = 0x03; - break; - case 0xd7: - case 0xd9: /* 1/0/0/0, 1 */ - case 0xea: /* gk20a */ - case 0x12b: /* gm20b */ - gr->screen_tile_row_offset = 0x01; - break; } + memset(gr->tile, 0xff, sizeof(gr->tile)); + gr->func->oneinit_tiles(gr); + gr->func->oneinit_sm_id(gr); return 0; } @@ -1914,13 +1995,68 @@ gf100_gr_new_(const struct gf100_gr_func *func, struct nvkm_device *device, } void +gf100_gr_init_400054(struct gf100_gr *gr) +{ + nvkm_wr32(gr->base.engine.subdev.device, 0x400054, 0x34ce3464); +} + +void +gf100_gr_init_shader_exceptions(struct gf100_gr *gr, int gpc, int tpc) +{ + struct nvkm_device *device = gr->base.engine.subdev.device; + nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x644), 0x001ffffe); + nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x64c), 0x0000000f); +} + +void +gf100_gr_init_tex_hww_esr(struct gf100_gr *gr, int gpc, int tpc) +{ + struct nvkm_device *device = gr->base.engine.subdev.device; + nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x224), 0xc0000000); +} + +void +gf100_gr_init_419eb4(struct gf100_gr *gr) +{ + struct nvkm_device *device = gr->base.engine.subdev.device; + nvkm_mask(device, 0x419eb4, 0x00001000, 0x00001000); +} + +void +gf100_gr_init_419cc0(struct gf100_gr *gr) +{ + struct nvkm_device *device = gr->base.engine.subdev.device; + int gpc, tpc; + + nvkm_mask(device, 0x419cc0, 0x00000008, 0x00000008); + + for (gpc = 0; gpc < gr->gpc_nr; gpc++) { + for (tpc = 0; tpc < gr->tpc_nr[gpc]; tpc++) + nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x48c), 0xc0000000); + } +} + +void +gf100_gr_init_40601c(struct gf100_gr *gr) +{ + nvkm_wr32(gr->base.engine.subdev.device, 0x40601c, 0xc0000000); +} + +void +gf100_gr_init_fecs_exceptions(struct gf100_gr *gr) +{ + const u32 data = gr->firmware ? 0x000e0000 : 0x000e0001; + nvkm_wr32(gr->base.engine.subdev.device, 0x409c24, data); +} + +void gf100_gr_init_gpc_mmu(struct gf100_gr *gr) { struct nvkm_device *device = gr->base.engine.subdev.device; struct nvkm_fb *fb = device->fb; nvkm_wr32(device, 0x418880, nvkm_rd32(device, 0x100c80) & 0x00000001); - nvkm_wr32(device, 0x4188a4, 0x00000000); + nvkm_wr32(device, 0x4188a4, 0x03000000); nvkm_wr32(device, 0x418888, 0x00000000); nvkm_wr32(device, 0x41888c, 0x00000000); nvkm_wr32(device, 0x418890, 0x00000000); @@ -1929,37 +2065,30 @@ gf100_gr_init_gpc_mmu(struct gf100_gr *gr) nvkm_wr32(device, 0x4188b8, nvkm_memory_addr(fb->mmu_rd) >> 8); } -int -gf100_gr_init(struct gf100_gr *gr) +void +gf100_gr_init_num_active_ltcs(struct gf100_gr *gr) { struct nvkm_device *device = gr->base.engine.subdev.device; - const u32 magicgpc918 = DIV_ROUND_UP(0x00800000, gr->tpc_total); - u32 data[TPC_MAX / 8] = {}; - u8 tpcnr[GPC_MAX]; - int gpc, tpc, rop; - int i; - - gr->func->init_gpc_mmu(gr); - - gf100_gr_mmio(gr, gr->func->mmio); - - nvkm_mask(device, TPC_UNIT(0, 0, 0x05c), 0x00000001, 0x00000001); - - memcpy(tpcnr, gr->tpc_nr, sizeof(gr->tpc_nr)); - for (i = 0, gpc = -1; i < gr->tpc_total; i++) { - do { - gpc = (gpc + 1) % gr->gpc_nr; - } while (!tpcnr[gpc]); - tpc = gr->tpc_nr[gpc] - tpcnr[gpc]--; + nvkm_wr32(device, GPC_BCAST(0x08ac), nvkm_rd32(device, 0x100800)); +} - data[i / 8] |= tpc << ((i % 8) * 4); +void +gf100_gr_init_zcull(struct gf100_gr *gr) +{ + struct nvkm_device *device = gr->base.engine.subdev.device; + const u32 magicgpc918 = DIV_ROUND_UP(0x00800000, gr->tpc_total); + const u8 tile_nr = ALIGN(gr->tpc_total, 32); + u8 bank[GPC_MAX] = {}, gpc, i, j; + u32 data; + + for (i = 0; i < tile_nr; i += 8) { + for (data = 0, j = 0; j < 8 && i + j < gr->tpc_total; j++) { + data |= bank[gr->tile[i + j]] << (j * 4); + bank[gr->tile[i + j]]++; + } + nvkm_wr32(device, GPC_BCAST(0x0980 + ((i / 8) * 4)), data); } - nvkm_wr32(device, GPC_BCAST(0x0980), data[0]); - nvkm_wr32(device, GPC_BCAST(0x0984), data[1]); - nvkm_wr32(device, GPC_BCAST(0x0988), data[2]); - nvkm_wr32(device, GPC_BCAST(0x098c), data[3]); - for (gpc = 0; gpc < gr->gpc_nr; gpc++) { nvkm_wr32(device, GPC_UNIT(gpc, 0x0914), gr->screen_tile_row_offset << 8 | gr->tpc_nr[gpc]); @@ -1968,29 +2097,88 @@ gf100_gr_init(struct gf100_gr *gr) nvkm_wr32(device, GPC_UNIT(gpc, 0x0918), magicgpc918); } - if (device->chipset != 0xd7) - nvkm_wr32(device, GPC_BCAST(0x1bd4), magicgpc918); + nvkm_wr32(device, GPC_BCAST(0x1bd4), magicgpc918); +} + +void +gf100_gr_init_vsc_stream_master(struct gf100_gr *gr) +{ + struct nvkm_device *device = gr->base.engine.subdev.device; + nvkm_mask(device, TPC_UNIT(0, 0, 0x05c), 0x00000001, 0x00000001); +} + +int +gf100_gr_init(struct gf100_gr *gr) +{ + struct nvkm_device *device = gr->base.engine.subdev.device; + int gpc, tpc, rop; + + if (gr->func->init_419bd8) + gr->func->init_419bd8(gr); + + gr->func->init_gpc_mmu(gr); + + if (gr->fuc_sw_nonctx) + gf100_gr_mmio(gr, gr->fuc_sw_nonctx); else - nvkm_wr32(device, GPC_BCAST(0x3fd4), magicgpc918); + gf100_gr_mmio(gr, gr->func->mmio); - nvkm_wr32(device, GPC_BCAST(0x08ac), nvkm_rd32(device, 0x100800)); + gf100_gr_wait_idle(gr); + + if (gr->func->init_r405a14) + gr->func->init_r405a14(gr); + + if (gr->func->clkgate_pack) + nvkm_therm_clkgate_init(device->therm, gr->func->clkgate_pack); + + if (gr->func->init_bios) + gr->func->init_bios(gr); + + gr->func->init_vsc_stream_master(gr); + gr->func->init_zcull(gr); + gr->func->init_num_active_ltcs(gr); + if (gr->func->init_rop_active_fbps) + gr->func->init_rop_active_fbps(gr); + if (gr->func->init_bios_2) + gr->func->init_bios_2(gr); + if (gr->func->init_swdx_pes_mask) + gr->func->init_swdx_pes_mask(gr); nvkm_wr32(device, 0x400500, 0x00010001); nvkm_wr32(device, 0x400100, 0xffffffff); nvkm_wr32(device, 0x40013c, 0xffffffff); + nvkm_wr32(device, 0x400124, 0x00000002); + + gr->func->init_fecs_exceptions(gr); + if (gr->func->init_ds_hww_esr_2) + gr->func->init_ds_hww_esr_2(gr); - nvkm_wr32(device, 0x409c24, 0x000f0000); nvkm_wr32(device, 0x404000, 0xc0000000); nvkm_wr32(device, 0x404600, 0xc0000000); nvkm_wr32(device, 0x408030, 0xc0000000); - nvkm_wr32(device, 0x40601c, 0xc0000000); + + if (gr->func->init_40601c) + gr->func->init_40601c(gr); + nvkm_wr32(device, 0x404490, 0xc0000000); nvkm_wr32(device, 0x406018, 0xc0000000); + + if (gr->func->init_sked_hww_esr) + gr->func->init_sked_hww_esr(gr); + nvkm_wr32(device, 0x405840, 0xc0000000); nvkm_wr32(device, 0x405844, 0x00ffffff); - nvkm_mask(device, 0x419cc0, 0x00000008, 0x00000008); - nvkm_mask(device, 0x419eb4, 0x00001000, 0x00001000); + + if (gr->func->init_419cc0) + gr->func->init_419cc0(gr); + if (gr->func->init_419eb4) + gr->func->init_419eb4(gr); + if (gr->func->init_419c9c) + gr->func->init_419c9c(gr); + + if (gr->func->init_ppc_exceptions) + gr->func->init_ppc_exceptions(gr); for (gpc = 0; gpc < gr->gpc_nr; gpc++) { nvkm_wr32(device, GPC_UNIT(gpc, 0x0420), 0xc0000000); @@ -2000,19 +2188,20 @@ gf100_gr_init(struct gf100_gr *gr) for (tpc = 0; tpc < gr->tpc_nr[gpc]; tpc++) { nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x508), 0xffffffff); nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x50c), 0xffffffff); - nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x224), 0xc0000000); - nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x48c), 0xc0000000); + if (gr->func->init_tex_hww_esr) + gr->func->init_tex_hww_esr(gr, gpc, tpc); nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x084), 0xc0000000); - nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x644), 0x001ffffe); - nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x64c), 0x0000000f); + if (gr->func->init_504430) + gr->func->init_504430(gr, gpc, tpc); + gr->func->init_shader_exceptions(gr, gpc, tpc); } nvkm_wr32(device, GPC_UNIT(gpc, 0x2c90), 0xffffffff); nvkm_wr32(device, GPC_UNIT(gpc, 0x2c94), 0xffffffff); } for (rop = 0; rop < gr->rop_nr; rop++) { - nvkm_wr32(device, ROP_UNIT(rop, 0x144), 0xc0000000); - nvkm_wr32(device, ROP_UNIT(rop, 0x070), 0xc0000000); + nvkm_wr32(device, ROP_UNIT(rop, 0x144), 0x40000000); + nvkm_wr32(device, ROP_UNIT(rop, 0x070), 0x40000000); nvkm_wr32(device, ROP_UNIT(rop, 0x204), 0xffffffff); nvkm_wr32(device, ROP_UNIT(rop, 0x208), 0xffffffff); } @@ -2024,10 +2213,14 @@ gf100_gr_init(struct gf100_gr *gr) nvkm_wr32(device, 0x40011c, 0xffffffff); nvkm_wr32(device, 0x400134, 0xffffffff); - nvkm_wr32(device, 0x400054, 0x34ce3464); + if (gr->func->init_400054) + gr->func->init_400054(gr); gf100_gr_zbc_init(gr); + if (gr->func->init_4188a4) + gr->func->init_4188a4(gr); + return gf100_gr_init_ctxctl(gr); } @@ -2053,13 +2246,27 @@ gf100_gr_gpccs_ucode = { static const struct gf100_gr_func gf100_gr = { + .oneinit_tiles = gf100_gr_oneinit_tiles, + .oneinit_sm_id = gf100_gr_oneinit_sm_id, .init = gf100_gr_init, .init_gpc_mmu = gf100_gr_init_gpc_mmu, + .init_vsc_stream_master = gf100_gr_init_vsc_stream_master, + .init_zcull = gf100_gr_init_zcull, + .init_num_active_ltcs = gf100_gr_init_num_active_ltcs, + .init_fecs_exceptions = gf100_gr_init_fecs_exceptions, + .init_40601c = gf100_gr_init_40601c, + .init_419cc0 = gf100_gr_init_419cc0, + .init_419eb4 = gf100_gr_init_419eb4, + .init_tex_hww_esr = gf100_gr_init_tex_hww_esr, + .init_shader_exceptions = gf100_gr_init_shader_exceptions, + .init_400054 = gf100_gr_init_400054, + .trap_mp = gf100_gr_trap_mp, .mmio = gf100_gr_pack_mmio, .fecs.ucode = &gf100_gr_fecs_ucode, .gpccs.ucode = &gf100_gr_gpccs_ucode, .rops = gf100_gr_rops, .grctx = &gf100_grctx, + .zbc = &gf100_gr_zbc, .sclass = { { -1, -1, FERMI_TWOD_A }, { -1, -1, FERMI_MEMORY_TO_MEMORY_FORMAT_A }, diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.h b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.h index c8ec3fd97155..dc46cf0131db 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.h +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.h @@ -72,6 +72,12 @@ struct gf100_gr_zbc_depth { u32 l2; }; +struct gf100_gr_zbc_stencil { + u32 format; + u32 ds; + u32 l2; +}; + struct gf100_gr { const struct gf100_gr_func *func; struct nvkm_gr base; @@ -95,21 +101,33 @@ struct gf100_gr { struct gf100_gr_zbc_color zbc_color[NVKM_LTC_MAX_ZBC_CNT]; struct gf100_gr_zbc_depth zbc_depth[NVKM_LTC_MAX_ZBC_CNT]; + struct gf100_gr_zbc_stencil zbc_stencil[NVKM_LTC_MAX_ZBC_CNT]; u8 rop_nr; u8 gpc_nr; u8 tpc_nr[GPC_MAX]; + u8 tpc_max; u8 tpc_total; u8 ppc_nr[GPC_MAX]; u8 ppc_mask[GPC_MAX]; + u8 ppc_tpc_mask[GPC_MAX][4]; u8 ppc_tpc_nr[GPC_MAX][4]; + u8 ppc_tpc_min; + u8 ppc_tpc_max; + + u8 screen_tile_row_offset; + u8 tile[TPC_MAX]; + + struct { + u8 gpc; + u8 tpc; + } sm[TPC_MAX]; + u8 sm_nr; struct gf100_gr_data mmio_data[4]; struct gf100_gr_mmio mmio_list[4096/8]; u32 size; u32 *data; - - u8 screen_tile_row_offset; }; int gf100_gr_ctor(const struct gf100_gr_func *, struct nvkm_device *, @@ -118,14 +136,43 @@ int gf100_gr_new_(const struct gf100_gr_func *, struct nvkm_device *, int, struct nvkm_gr **); void *gf100_gr_dtor(struct nvkm_gr *); +struct gf100_gr_func_zbc { + void (*clear_color)(struct gf100_gr *, int zbc); + void (*clear_depth)(struct gf100_gr *, int zbc); + int (*stencil_get)(struct gf100_gr *, int format, + const u32 ds, const u32 l2); + void (*clear_stencil)(struct gf100_gr *, int zbc); +}; + struct gf100_gr_func { void (*dtor)(struct gf100_gr *); + void (*oneinit_tiles)(struct gf100_gr *); + void (*oneinit_sm_id)(struct gf100_gr *); int (*init)(struct gf100_gr *); + void (*init_419bd8)(struct gf100_gr *); void (*init_gpc_mmu)(struct gf100_gr *); + void (*init_r405a14)(struct gf100_gr *); + void (*init_bios)(struct gf100_gr *); + void (*init_vsc_stream_master)(struct gf100_gr *); + void (*init_zcull)(struct gf100_gr *); + void (*init_num_active_ltcs)(struct gf100_gr *); void (*init_rop_active_fbps)(struct gf100_gr *); - void (*init_ppc_exceptions)(struct gf100_gr *); + void (*init_bios_2)(struct gf100_gr *); void (*init_swdx_pes_mask)(struct gf100_gr *); - void (*init_num_active_ltcs)(struct gf100_gr *); + void (*init_fecs_exceptions)(struct gf100_gr *); + void (*init_ds_hww_esr_2)(struct gf100_gr *); + void (*init_40601c)(struct gf100_gr *); + void (*init_sked_hww_esr)(struct gf100_gr *); + void (*init_419cc0)(struct gf100_gr *); + void (*init_419eb4)(struct gf100_gr *); + void (*init_419c9c)(struct gf100_gr *); + void (*init_ppc_exceptions)(struct gf100_gr *); + void (*init_tex_hww_esr)(struct gf100_gr *, int gpc, int tpc); + void (*init_504430)(struct gf100_gr *, int gpc, int tpc); + void (*init_shader_exceptions)(struct gf100_gr *, int gpc, int tpc); + void (*init_400054)(struct gf100_gr *); + void (*init_4188a4)(struct gf100_gr *); + void (*trap_mp)(struct gf100_gr *, int gpc, int tpc); void (*set_hww_esr_report_mask)(struct gf100_gr *); const struct gf100_gr_pack *mmio; struct { @@ -135,26 +182,60 @@ struct gf100_gr_func { struct gf100_gr_ucode *ucode; } gpccs; int (*rops)(struct gf100_gr *); + int gpc_nr; + int tpc_nr; int ppc_nr; const struct gf100_grctx_func *grctx; const struct nvkm_therm_clkgate_pack *clkgate_pack; + const struct gf100_gr_func_zbc *zbc; struct nvkm_sclass sclass[]; }; -int gf100_gr_init(struct gf100_gr *); int gf100_gr_rops(struct gf100_gr *); - -int gk104_gr_init(struct gf100_gr *); +void gf100_gr_oneinit_tiles(struct gf100_gr *); +void gf100_gr_oneinit_sm_id(struct gf100_gr *); +int gf100_gr_init(struct gf100_gr *); +void gf100_gr_init_vsc_stream_master(struct gf100_gr *); +void gf100_gr_init_zcull(struct gf100_gr *); +void gf100_gr_init_num_active_ltcs(struct gf100_gr *); +void gf100_gr_init_fecs_exceptions(struct gf100_gr *); +void gf100_gr_init_40601c(struct gf100_gr *); +void gf100_gr_init_419cc0(struct gf100_gr *); +void gf100_gr_init_419eb4(struct gf100_gr *); +void gf100_gr_init_tex_hww_esr(struct gf100_gr *, int, int); +void gf100_gr_init_shader_exceptions(struct gf100_gr *, int, int); +void gf100_gr_init_400054(struct gf100_gr *); +extern const struct gf100_gr_func_zbc gf100_gr_zbc; + +void gf117_gr_init_zcull(struct gf100_gr *); + +void gk104_gr_init_vsc_stream_master(struct gf100_gr *); void gk104_gr_init_rop_active_fbps(struct gf100_gr *); void gk104_gr_init_ppc_exceptions(struct gf100_gr *); +void gk104_gr_init_sked_hww_esr(struct gf100_gr *); + +void gk110_gr_init_419eb4(struct gf100_gr *); + +void gm107_gr_init_504430(struct gf100_gr *, int, int); +void gm107_gr_init_shader_exceptions(struct gf100_gr *, int, int); +void gm107_gr_init_400054(struct gf100_gr *); int gk20a_gr_init(struct gf100_gr *); -int gm200_gr_init(struct gf100_gr *); +void gm200_gr_oneinit_tiles(struct gf100_gr *); +void gm200_gr_oneinit_sm_id(struct gf100_gr *); int gm200_gr_rops(struct gf100_gr *); +void gm200_gr_init_num_active_ltcs(struct gf100_gr *); +void gm200_gr_init_ds_hww_esr_2(struct gf100_gr *); -int gp100_gr_init(struct gf100_gr *); void gp100_gr_init_rop_active_fbps(struct gf100_gr *); +void gp100_gr_init_fecs_exceptions(struct gf100_gr *); +void gp100_gr_init_shader_exceptions(struct gf100_gr *, int, int); +void gp100_gr_zbc_clear_color(struct gf100_gr *, int); +void gp100_gr_zbc_clear_depth(struct gf100_gr *, int); + +void gp102_gr_init_swdx_pes_mask(struct gf100_gr *); +extern const struct gf100_gr_func_zbc gp102_gr_zbc; #define gf100_gr_chan(p) container_of((p), struct gf100_gr_chan, object) #include <core/object.h> @@ -187,7 +268,7 @@ extern const struct nvkm_object_func gf100_fermi; struct gf100_gr_init { u32 addr; u8 count; - u8 pitch; + u32 pitch; u32 data; }; @@ -257,6 +338,9 @@ extern const struct gf100_gr_init gf100_gr_init_be_0[]; extern const struct gf100_gr_init gf100_gr_init_fe_1[]; extern const struct gf100_gr_init gf100_gr_init_pe_1[]; void gf100_gr_init_gpc_mmu(struct gf100_gr *); +void gf100_gr_trap_mp(struct gf100_gr *, int, int); +extern const struct nvkm_bitfield gf100_mp_global_error[]; +extern const struct nvkm_enum gf100_mp_warp_error[]; extern const struct gf100_gr_init gf104_gr_init_ds_0[]; extern const struct gf100_gr_init gf104_gr_init_tex_0[]; @@ -279,6 +363,7 @@ extern const struct gf100_gr_init gf117_gr_init_wwdx_0[]; extern const struct gf100_gr_init gf117_gr_init_cbm_0[]; extern const struct gf100_gr_init gk104_gr_init_main_0[]; +extern const struct gf100_gr_init gk104_gr_init_gpc_unk_2[]; extern const struct gf100_gr_init gk104_gr_init_tpccs_0[]; extern const struct gf100_gr_init gk104_gr_init_pe_0[]; extern const struct gf100_gr_init gk104_gr_init_be_0[]; @@ -306,8 +391,4 @@ extern const struct gf100_gr_init gm107_gr_init_cbm_0[]; void gm107_gr_init_bios(struct gf100_gr *); void gm200_gr_init_gpc_mmu(struct gf100_gr *); - -void gp100_gr_init_num_active_ltcs(struct gf100_gr *gr); - -void gp102_gr_init_swdx_pes_mask(struct gf100_gr *); #endif diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf104.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf104.c index ec0f11983b23..42c2fd9fc04e 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf104.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf104.c @@ -114,13 +114,27 @@ gf104_gr_pack_mmio[] = { static const struct gf100_gr_func gf104_gr = { + .oneinit_tiles = gf100_gr_oneinit_tiles, + .oneinit_sm_id = gf100_gr_oneinit_sm_id, .init = gf100_gr_init, .init_gpc_mmu = gf100_gr_init_gpc_mmu, + .init_vsc_stream_master = gf100_gr_init_vsc_stream_master, + .init_zcull = gf100_gr_init_zcull, + .init_num_active_ltcs = gf100_gr_init_num_active_ltcs, + .init_fecs_exceptions = gf100_gr_init_fecs_exceptions, + .init_40601c = gf100_gr_init_40601c, + .init_419cc0 = gf100_gr_init_419cc0, + .init_419eb4 = gf100_gr_init_419eb4, + .init_tex_hww_esr = gf100_gr_init_tex_hww_esr, + .init_shader_exceptions = gf100_gr_init_shader_exceptions, + .init_400054 = gf100_gr_init_400054, + .trap_mp = gf100_gr_trap_mp, .mmio = gf104_gr_pack_mmio, .fecs.ucode = &gf100_gr_fecs_ucode, .gpccs.ucode = &gf100_gr_gpccs_ucode, .rops = gf100_gr_rops, .grctx = &gf104_grctx, + .zbc = &gf100_gr_zbc, .sclass = { { -1, -1, FERMI_TWOD_A }, { -1, -1, FERMI_MEMORY_TO_MEMORY_FORMAT_A }, diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf108.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf108.c index cc152eb74123..4731a460adc7 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf108.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf108.c @@ -103,15 +103,36 @@ gf108_gr_pack_mmio[] = { * PGRAPH engine/subdev functions ******************************************************************************/ +static void +gf108_gr_init_r405a14(struct gf100_gr *gr) +{ + nvkm_wr32(gr->base.engine.subdev.device, 0x405a14, 0x80000000); +} + static const struct gf100_gr_func gf108_gr = { + .oneinit_tiles = gf100_gr_oneinit_tiles, + .oneinit_sm_id = gf100_gr_oneinit_sm_id, .init = gf100_gr_init, .init_gpc_mmu = gf100_gr_init_gpc_mmu, + .init_r405a14 = gf108_gr_init_r405a14, + .init_vsc_stream_master = gf100_gr_init_vsc_stream_master, + .init_zcull = gf100_gr_init_zcull, + .init_num_active_ltcs = gf100_gr_init_num_active_ltcs, + .init_fecs_exceptions = gf100_gr_init_fecs_exceptions, + .init_40601c = gf100_gr_init_40601c, + .init_419cc0 = gf100_gr_init_419cc0, + .init_419eb4 = gf100_gr_init_419eb4, + .init_tex_hww_esr = gf100_gr_init_tex_hww_esr, + .init_shader_exceptions = gf100_gr_init_shader_exceptions, + .init_400054 = gf100_gr_init_400054, + .trap_mp = gf100_gr_trap_mp, .mmio = gf108_gr_pack_mmio, .fecs.ucode = &gf100_gr_fecs_ucode, .gpccs.ucode = &gf100_gr_gpccs_ucode, .rops = gf100_gr_rops, .grctx = &gf108_grctx, + .zbc = &gf100_gr_zbc, .sclass = { { -1, -1, FERMI_TWOD_A }, { -1, -1, FERMI_MEMORY_TO_MEMORY_FORMAT_A }, diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf110.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf110.c index 10d2d73ca8c3..cdf759c8cd7f 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf110.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf110.c @@ -86,13 +86,27 @@ gf110_gr_pack_mmio[] = { static const struct gf100_gr_func gf110_gr = { + .oneinit_tiles = gf100_gr_oneinit_tiles, + .oneinit_sm_id = gf100_gr_oneinit_sm_id, .init = gf100_gr_init, .init_gpc_mmu = gf100_gr_init_gpc_mmu, + .init_vsc_stream_master = gf100_gr_init_vsc_stream_master, + .init_zcull = gf100_gr_init_zcull, + .init_num_active_ltcs = gf100_gr_init_num_active_ltcs, + .init_fecs_exceptions = gf100_gr_init_fecs_exceptions, + .init_40601c = gf100_gr_init_40601c, + .init_419cc0 = gf100_gr_init_419cc0, + .init_419eb4 = gf100_gr_init_419eb4, + .init_tex_hww_esr = gf100_gr_init_tex_hww_esr, + .init_shader_exceptions = gf100_gr_init_shader_exceptions, + .init_400054 = gf100_gr_init_400054, + .trap_mp = gf100_gr_trap_mp, .mmio = gf110_gr_pack_mmio, .fecs.ucode = &gf100_gr_fecs_ucode, .gpccs.ucode = &gf100_gr_gpccs_ucode, .rops = gf100_gr_rops, .grctx = &gf110_grctx, + .zbc = &gf100_gr_zbc, .sclass = { { -1, -1, FERMI_TWOD_A }, { -1, -1, FERMI_MEMORY_TO_MEMORY_FORMAT_A }, diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf117.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf117.c index ac09a07c4150..a4158f84c649 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf117.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf117.c @@ -120,16 +120,58 @@ gf117_gr_gpccs_ucode = { .data.size = sizeof(gf117_grgpc_data), }; +void +gf117_gr_init_zcull(struct gf100_gr *gr) +{ + struct nvkm_device *device = gr->base.engine.subdev.device; + const u32 magicgpc918 = DIV_ROUND_UP(0x00800000, gr->tpc_total); + const u8 tile_nr = ALIGN(gr->tpc_total, 32); + u8 bank[GPC_MAX] = {}, gpc, i, j; + u32 data; + + for (i = 0; i < tile_nr; i += 8) { + for (data = 0, j = 0; j < 8 && i + j < gr->tpc_total; j++) { + data |= bank[gr->tile[i + j]] << (j * 4); + bank[gr->tile[i + j]]++; + } + nvkm_wr32(device, GPC_BCAST(0x0980 + ((i / 8) * 4)), data); + } + + for (gpc = 0; gpc < gr->gpc_nr; gpc++) { + nvkm_wr32(device, GPC_UNIT(gpc, 0x0914), + gr->screen_tile_row_offset << 8 | gr->tpc_nr[gpc]); + nvkm_wr32(device, GPC_UNIT(gpc, 0x0910), 0x00040000 | + gr->tpc_total); + nvkm_wr32(device, GPC_UNIT(gpc, 0x0918), magicgpc918); + } + + nvkm_wr32(device, GPC_BCAST(0x3fd4), magicgpc918); +} + static const struct gf100_gr_func gf117_gr = { + .oneinit_tiles = gf100_gr_oneinit_tiles, + .oneinit_sm_id = gf100_gr_oneinit_sm_id, .init = gf100_gr_init, .init_gpc_mmu = gf100_gr_init_gpc_mmu, + .init_vsc_stream_master = gf100_gr_init_vsc_stream_master, + .init_zcull = gf117_gr_init_zcull, + .init_num_active_ltcs = gf100_gr_init_num_active_ltcs, + .init_fecs_exceptions = gf100_gr_init_fecs_exceptions, + .init_40601c = gf100_gr_init_40601c, + .init_419cc0 = gf100_gr_init_419cc0, + .init_419eb4 = gf100_gr_init_419eb4, + .init_tex_hww_esr = gf100_gr_init_tex_hww_esr, + .init_shader_exceptions = gf100_gr_init_shader_exceptions, + .init_400054 = gf100_gr_init_400054, + .trap_mp = gf100_gr_trap_mp, .mmio = gf117_gr_pack_mmio, .fecs.ucode = &gf117_gr_fecs_ucode, .gpccs.ucode = &gf117_gr_gpccs_ucode, .rops = gf100_gr_rops, .ppc_nr = 1, .grctx = &gf117_grctx, + .zbc = &gf100_gr_zbc, .sclass = { { -1, -1, FERMI_TWOD_A }, { -1, -1, FERMI_MEMORY_TO_MEMORY_FORMAT_A }, diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf119.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf119.c index 7f449ec6f760..4197844870b3 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf119.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf119.c @@ -177,13 +177,27 @@ gf119_gr_pack_mmio[] = { static const struct gf100_gr_func gf119_gr = { + .oneinit_tiles = gf100_gr_oneinit_tiles, + .oneinit_sm_id = gf100_gr_oneinit_sm_id, .init = gf100_gr_init, .init_gpc_mmu = gf100_gr_init_gpc_mmu, + .init_vsc_stream_master = gf100_gr_init_vsc_stream_master, + .init_zcull = gf100_gr_init_zcull, + .init_num_active_ltcs = gf100_gr_init_num_active_ltcs, + .init_fecs_exceptions = gf100_gr_init_fecs_exceptions, + .init_40601c = gf100_gr_init_40601c, + .init_419cc0 = gf100_gr_init_419cc0, + .init_419eb4 = gf100_gr_init_419eb4, + .init_tex_hww_esr = gf100_gr_init_tex_hww_esr, + .init_shader_exceptions = gf100_gr_init_shader_exceptions, + .init_400054 = gf100_gr_init_400054, + .trap_mp = gf100_gr_trap_mp, .mmio = gf119_gr_pack_mmio, .fecs.ucode = &gf100_gr_fecs_ucode, .gpccs.ucode = &gf100_gr_gpccs_ucode, .rops = gf100_gr_rops, .grctx = &gf119_grctx, + .zbc = &gf100_gr_zbc, .sclass = { { -1, -1, FERMI_TWOD_A }, { -1, -1, FERMI_MEMORY_TO_MEMORY_FORMAT_A }, diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gk104.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gk104.c index 1b52fcb2c49a..477fee3e3715 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gk104.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gk104.c @@ -83,6 +83,12 @@ gk104_gr_init_gpc_unk_1[] = { }; const struct gf100_gr_init +gk104_gr_init_gpc_unk_2[] = { + { 0x418884, 1, 0x04, 0x00000000 }, + {} +}; + +const struct gf100_gr_init gk104_gr_init_tpccs_0[] = { { 0x419d0c, 1, 0x04, 0x00000000 }, { 0x419d10, 1, 0x04, 0x00000014 }, @@ -160,6 +166,7 @@ gk104_gr_pack_mmio[] = { { gf119_gr_init_gpm_0 }, { gk104_gr_init_gpc_unk_1 }, { gf100_gr_init_gcc_0 }, + { gk104_gr_init_gpc_unk_2 }, { gk104_gr_init_tpccs_0 }, { gf119_gr_init_tex_0 }, { gk104_gr_init_pe_0 }, @@ -381,6 +388,21 @@ gk104_clkgate_pack[] = { ******************************************************************************/ void +gk104_gr_init_sked_hww_esr(struct gf100_gr *gr) +{ + nvkm_wr32(gr->base.engine.subdev.device, 0x407020, 0x40000000); +} + +static void +gk104_gr_init_fecs_exceptions(struct gf100_gr *gr) +{ + struct nvkm_device *device = gr->base.engine.subdev.device; + nvkm_wr32(device, 0x409ffc, 0x00000000); + nvkm_wr32(device, 0x409c14, 0x00003e3e); + nvkm_wr32(device, 0x409c24, 0x000f0001); +} + +void gk104_gr_init_rop_active_fbps(struct gf100_gr *gr) { struct nvkm_device *device = gr->base.engine.subdev.device; @@ -404,112 +426,11 @@ gk104_gr_init_ppc_exceptions(struct gf100_gr *gr) } } -int -gk104_gr_init(struct gf100_gr *gr) +void +gk104_gr_init_vsc_stream_master(struct gf100_gr *gr) { struct nvkm_device *device = gr->base.engine.subdev.device; - const u32 magicgpc918 = DIV_ROUND_UP(0x00800000, gr->tpc_total); - u32 data[TPC_MAX / 8] = {}; - u8 tpcnr[GPC_MAX]; - int gpc, tpc, rop; - int i; - - gr->func->init_gpc_mmu(gr); - - gf100_gr_mmio(gr, gr->func->mmio); - if (gr->func->clkgate_pack) - nvkm_therm_clkgate_init(gr->base.engine.subdev.device->therm, - gr->func->clkgate_pack); - nvkm_wr32(device, GPC_UNIT(0, 0x3018), 0x00000001); - - memset(data, 0x00, sizeof(data)); - memcpy(tpcnr, gr->tpc_nr, sizeof(gr->tpc_nr)); - for (i = 0, gpc = -1; i < gr->tpc_total; i++) { - do { - gpc = (gpc + 1) % gr->gpc_nr; - } while (!tpcnr[gpc]); - tpc = gr->tpc_nr[gpc] - tpcnr[gpc]--; - - data[i / 8] |= tpc << ((i % 8) * 4); - } - - nvkm_wr32(device, GPC_BCAST(0x0980), data[0]); - nvkm_wr32(device, GPC_BCAST(0x0984), data[1]); - nvkm_wr32(device, GPC_BCAST(0x0988), data[2]); - nvkm_wr32(device, GPC_BCAST(0x098c), data[3]); - - for (gpc = 0; gpc < gr->gpc_nr; gpc++) { - nvkm_wr32(device, GPC_UNIT(gpc, 0x0914), - gr->screen_tile_row_offset << 8 | gr->tpc_nr[gpc]); - nvkm_wr32(device, GPC_UNIT(gpc, 0x0910), 0x00040000 | - gr->tpc_total); - nvkm_wr32(device, GPC_UNIT(gpc, 0x0918), magicgpc918); - } - - nvkm_wr32(device, GPC_BCAST(0x3fd4), magicgpc918); - nvkm_wr32(device, GPC_BCAST(0x08ac), nvkm_rd32(device, 0x100800)); - - gr->func->init_rop_active_fbps(gr); - - nvkm_wr32(device, 0x400500, 0x00010001); - - nvkm_wr32(device, 0x400100, 0xffffffff); - nvkm_wr32(device, 0x40013c, 0xffffffff); - - nvkm_wr32(device, 0x409ffc, 0x00000000); - nvkm_wr32(device, 0x409c14, 0x00003e3e); - nvkm_wr32(device, 0x409c24, 0x000f0001); - nvkm_wr32(device, 0x404000, 0xc0000000); - nvkm_wr32(device, 0x404600, 0xc0000000); - nvkm_wr32(device, 0x408030, 0xc0000000); - nvkm_wr32(device, 0x404490, 0xc0000000); - nvkm_wr32(device, 0x406018, 0xc0000000); - nvkm_wr32(device, 0x407020, 0x40000000); - nvkm_wr32(device, 0x405840, 0xc0000000); - nvkm_wr32(device, 0x405844, 0x00ffffff); - nvkm_mask(device, 0x419cc0, 0x00000008, 0x00000008); - nvkm_mask(device, 0x419eb4, 0x00001000, 0x00001000); - - gr->func->init_ppc_exceptions(gr); - - for (gpc = 0; gpc < gr->gpc_nr; gpc++) { - nvkm_wr32(device, GPC_UNIT(gpc, 0x0420), 0xc0000000); - nvkm_wr32(device, GPC_UNIT(gpc, 0x0900), 0xc0000000); - nvkm_wr32(device, GPC_UNIT(gpc, 0x1028), 0xc0000000); - nvkm_wr32(device, GPC_UNIT(gpc, 0x0824), 0xc0000000); - for (tpc = 0; tpc < gr->tpc_nr[gpc]; tpc++) { - nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x508), 0xffffffff); - nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x50c), 0xffffffff); - nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x224), 0xc0000000); - nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x48c), 0xc0000000); - nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x084), 0xc0000000); - nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x644), 0x001ffffe); - nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x64c), 0x0000000f); - } - nvkm_wr32(device, GPC_UNIT(gpc, 0x2c90), 0xffffffff); - nvkm_wr32(device, GPC_UNIT(gpc, 0x2c94), 0xffffffff); - } - - for (rop = 0; rop < gr->rop_nr; rop++) { - nvkm_wr32(device, ROP_UNIT(rop, 0x144), 0xc0000000); - nvkm_wr32(device, ROP_UNIT(rop, 0x070), 0xc0000000); - nvkm_wr32(device, ROP_UNIT(rop, 0x204), 0xffffffff); - nvkm_wr32(device, ROP_UNIT(rop, 0x208), 0xffffffff); - } - - nvkm_wr32(device, 0x400108, 0xffffffff); - nvkm_wr32(device, 0x400138, 0xffffffff); - nvkm_wr32(device, 0x400118, 0xffffffff); - nvkm_wr32(device, 0x400130, 0xffffffff); - nvkm_wr32(device, 0x40011c, 0xffffffff); - nvkm_wr32(device, 0x400134, 0xffffffff); - - nvkm_wr32(device, 0x400054, 0x34ce3464); - - gf100_gr_zbc_init(gr); - - return gf100_gr_init_ctxctl(gr); } #include "fuc/hubgk104.fuc3.h" @@ -534,10 +455,23 @@ gk104_gr_gpccs_ucode = { static const struct gf100_gr_func gk104_gr = { - .init = gk104_gr_init, + .oneinit_tiles = gf100_gr_oneinit_tiles, + .oneinit_sm_id = gf100_gr_oneinit_sm_id, + .init = gf100_gr_init, .init_gpc_mmu = gf100_gr_init_gpc_mmu, + .init_vsc_stream_master = gk104_gr_init_vsc_stream_master, + .init_zcull = gf117_gr_init_zcull, + .init_num_active_ltcs = gf100_gr_init_num_active_ltcs, .init_rop_active_fbps = gk104_gr_init_rop_active_fbps, + .init_fecs_exceptions = gk104_gr_init_fecs_exceptions, + .init_sked_hww_esr = gk104_gr_init_sked_hww_esr, + .init_419cc0 = gf100_gr_init_419cc0, + .init_419eb4 = gf100_gr_init_419eb4, .init_ppc_exceptions = gk104_gr_init_ppc_exceptions, + .init_tex_hww_esr = gf100_gr_init_tex_hww_esr, + .init_shader_exceptions = gf100_gr_init_shader_exceptions, + .init_400054 = gf100_gr_init_400054, + .trap_mp = gf100_gr_trap_mp, .mmio = gk104_gr_pack_mmio, .fecs.ucode = &gk104_gr_fecs_ucode, .gpccs.ucode = &gk104_gr_gpccs_ucode, @@ -545,6 +479,7 @@ gk104_gr = { .ppc_nr = 1, .grctx = &gk104_grctx, .clkgate_pack = gk104_clkgate_pack, + .zbc = &gf100_gr_zbc, .sclass = { { -1, -1, FERMI_TWOD_A }, { -1, -1, KEPLER_INLINE_TO_MEMORY_A }, diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gk110.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gk110.c index 4da916a9fc73..7cd628c84e07 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gk110.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gk110.c @@ -143,6 +143,7 @@ gk110_gr_pack_mmio[] = { { gf119_gr_init_gpm_0 }, { gk110_gr_init_gpc_unk_1 }, { gf100_gr_init_gcc_0 }, + { gk104_gr_init_gpc_unk_2 }, { gk104_gr_init_tpccs_0 }, { gk110_gr_init_tex_0 }, { gk104_gr_init_pe_0 }, @@ -334,12 +335,39 @@ gk110_gr_gpccs_ucode = { .data.size = sizeof(gk110_grgpc_data), }; +void +gk110_gr_init_419eb4(struct gf100_gr *gr) +{ + struct nvkm_device *device = gr->base.engine.subdev.device; + nvkm_mask(device, 0x419eb4, 0x00001000, 0x00001000); + nvkm_mask(device, 0x419eb4, 0x00002000, 0x00002000); + nvkm_mask(device, 0x419eb4, 0x00004000, 0x00004000); + nvkm_mask(device, 0x419eb4, 0x00008000, 0x00008000); + nvkm_mask(device, 0x419eb4, 0x00001000, 0x00000000); + nvkm_mask(device, 0x419eb4, 0x00002000, 0x00000000); + nvkm_mask(device, 0x419eb4, 0x00004000, 0x00000000); + nvkm_mask(device, 0x419eb4, 0x00008000, 0x00000000); +} + static const struct gf100_gr_func gk110_gr = { - .init = gk104_gr_init, + .oneinit_tiles = gf100_gr_oneinit_tiles, + .oneinit_sm_id = gf100_gr_oneinit_sm_id, + .init = gf100_gr_init, .init_gpc_mmu = gf100_gr_init_gpc_mmu, + .init_vsc_stream_master = gk104_gr_init_vsc_stream_master, + .init_zcull = gf117_gr_init_zcull, + .init_num_active_ltcs = gf100_gr_init_num_active_ltcs, .init_rop_active_fbps = gk104_gr_init_rop_active_fbps, + .init_fecs_exceptions = gf100_gr_init_fecs_exceptions, + .init_sked_hww_esr = gk104_gr_init_sked_hww_esr, + .init_419cc0 = gf100_gr_init_419cc0, + .init_419eb4 = gk110_gr_init_419eb4, .init_ppc_exceptions = gk104_gr_init_ppc_exceptions, + .init_tex_hww_esr = gf100_gr_init_tex_hww_esr, + .init_shader_exceptions = gf100_gr_init_shader_exceptions, + .init_400054 = gf100_gr_init_400054, + .trap_mp = gf100_gr_trap_mp, .mmio = gk110_gr_pack_mmio, .fecs.ucode = &gk110_gr_fecs_ucode, .gpccs.ucode = &gk110_gr_gpccs_ucode, @@ -347,6 +375,7 @@ gk110_gr = { .ppc_nr = 2, .grctx = &gk110_grctx, .clkgate_pack = gk110_clkgate_pack, + .zbc = &gf100_gr_zbc, .sclass = { { -1, -1, FERMI_TWOD_A }, { -1, -1, KEPLER_INLINE_TO_MEMORY_B }, diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gk110b.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gk110b.c index 1912c0bfd7ee..a38faa215635 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gk110b.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gk110b.c @@ -82,6 +82,7 @@ gk110b_gr_pack_mmio[] = { { gf119_gr_init_gpm_0 }, { gk110_gr_init_gpc_unk_1 }, { gf100_gr_init_gcc_0 }, + { gk104_gr_init_gpc_unk_2 }, { gk104_gr_init_tpccs_0 }, { gk110_gr_init_tex_0 }, { gk104_gr_init_pe_0 }, @@ -102,16 +103,30 @@ gk110b_gr_pack_mmio[] = { static const struct gf100_gr_func gk110b_gr = { - .init = gk104_gr_init, + .oneinit_tiles = gf100_gr_oneinit_tiles, + .oneinit_sm_id = gf100_gr_oneinit_sm_id, + .init = gf100_gr_init, .init_gpc_mmu = gf100_gr_init_gpc_mmu, + .init_vsc_stream_master = gk104_gr_init_vsc_stream_master, + .init_zcull = gf117_gr_init_zcull, + .init_num_active_ltcs = gf100_gr_init_num_active_ltcs, .init_rop_active_fbps = gk104_gr_init_rop_active_fbps, + .init_fecs_exceptions = gf100_gr_init_fecs_exceptions, + .init_sked_hww_esr = gk104_gr_init_sked_hww_esr, + .init_419cc0 = gf100_gr_init_419cc0, + .init_419eb4 = gk110_gr_init_419eb4, .init_ppc_exceptions = gk104_gr_init_ppc_exceptions, + .init_tex_hww_esr = gf100_gr_init_tex_hww_esr, + .init_shader_exceptions = gf100_gr_init_shader_exceptions, + .init_400054 = gf100_gr_init_400054, + .trap_mp = gf100_gr_trap_mp, .mmio = gk110b_gr_pack_mmio, .fecs.ucode = &gk110_gr_fecs_ucode, .gpccs.ucode = &gk110_gr_gpccs_ucode, .rops = gf100_gr_rops, .ppc_nr = 2, .grctx = &gk110b_grctx, + .zbc = &gf100_gr_zbc, .sclass = { { -1, -1, FERMI_TWOD_A }, { -1, -1, KEPLER_INLINE_TO_MEMORY_B }, diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gk208.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gk208.c index 1fc258163f25..58456660e603 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gk208.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gk208.c @@ -121,6 +121,7 @@ gk208_gr_pack_mmio[] = { { gf119_gr_init_gpm_0 }, { gk110_gr_init_gpc_unk_1 }, { gf100_gr_init_gcc_0 }, + { gk104_gr_init_gpc_unk_2 }, { gk104_gr_init_tpccs_0 }, { gk208_gr_init_tex_0 }, { gk104_gr_init_pe_0 }, @@ -161,16 +162,29 @@ gk208_gr_gpccs_ucode = { static const struct gf100_gr_func gk208_gr = { - .init = gk104_gr_init, + .oneinit_tiles = gf100_gr_oneinit_tiles, + .oneinit_sm_id = gf100_gr_oneinit_sm_id, + .init = gf100_gr_init, .init_gpc_mmu = gf100_gr_init_gpc_mmu, + .init_vsc_stream_master = gk104_gr_init_vsc_stream_master, + .init_zcull = gf117_gr_init_zcull, + .init_num_active_ltcs = gf100_gr_init_num_active_ltcs, .init_rop_active_fbps = gk104_gr_init_rop_active_fbps, + .init_fecs_exceptions = gf100_gr_init_fecs_exceptions, + .init_sked_hww_esr = gk104_gr_init_sked_hww_esr, + .init_419cc0 = gf100_gr_init_419cc0, .init_ppc_exceptions = gk104_gr_init_ppc_exceptions, + .init_tex_hww_esr = gf100_gr_init_tex_hww_esr, + .init_shader_exceptions = gf100_gr_init_shader_exceptions, + .init_400054 = gf100_gr_init_400054, + .trap_mp = gf100_gr_trap_mp, .mmio = gk208_gr_pack_mmio, .fecs.ucode = &gk208_gr_fecs_ucode, .gpccs.ucode = &gk208_gr_gpccs_ucode, .rops = gf100_gr_rops, .ppc_nr = 1, .grctx = &gk208_grctx, + .zbc = &gf100_gr_zbc, .sclass = { { -1, -1, FERMI_TWOD_A }, { -1, -1, KEPLER_INLINE_TO_MEMORY_B }, diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gk20a.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gk20a.c index de8b806b88fd..500cb08dd608 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gk20a.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gk20a.c @@ -219,11 +219,7 @@ int gk20a_gr_init(struct gf100_gr *gr) { struct nvkm_device *device = gr->base.engine.subdev.device; - const u32 magicgpc918 = DIV_ROUND_UP(0x00800000, gr->tpc_total); - u32 data[TPC_MAX / 8] = {}; - u8 tpcnr[GPC_MAX]; - int gpc, tpc; - int ret, i; + int ret; /* Clear SCC RAM */ nvkm_wr32(device, 0x40802c, 0x1); @@ -246,31 +242,7 @@ gk20a_gr_init(struct gf100_gr *gr) nvkm_mask(device, 0x503018, 0x1, 0x1); /* Zcull init */ - memset(data, 0x00, sizeof(data)); - memcpy(tpcnr, gr->tpc_nr, sizeof(gr->tpc_nr)); - for (i = 0, gpc = -1; i < gr->tpc_total; i++) { - do { - gpc = (gpc + 1) % gr->gpc_nr; - } while (!tpcnr[gpc]); - tpc = gr->tpc_nr[gpc] - tpcnr[gpc]--; - - data[i / 8] |= tpc << ((i % 8) * 4); - } - - nvkm_wr32(device, GPC_BCAST(0x0980), data[0]); - nvkm_wr32(device, GPC_BCAST(0x0984), data[1]); - nvkm_wr32(device, GPC_BCAST(0x0988), data[2]); - nvkm_wr32(device, GPC_BCAST(0x098c), data[3]); - - for (gpc = 0; gpc < gr->gpc_nr; gpc++) { - nvkm_wr32(device, GPC_UNIT(gpc, 0x0914), - gr->screen_tile_row_offset << 8 | gr->tpc_nr[gpc]); - nvkm_wr32(device, GPC_UNIT(gpc, 0x0910), 0x00040000 | - gr->tpc_total); - nvkm_wr32(device, GPC_UNIT(gpc, 0x0918), magicgpc918); - } - - nvkm_wr32(device, GPC_BCAST(0x3fd4), magicgpc918); + gr->func->init_zcull(gr); gr->func->init_rop_active_fbps(gr); @@ -310,12 +282,17 @@ gk20a_gr_init(struct gf100_gr *gr) static const struct gf100_gr_func gk20a_gr = { + .oneinit_tiles = gf100_gr_oneinit_tiles, + .oneinit_sm_id = gf100_gr_oneinit_sm_id, .init = gk20a_gr_init, + .init_zcull = gf117_gr_init_zcull, .init_rop_active_fbps = gk104_gr_init_rop_active_fbps, + .trap_mp = gf100_gr_trap_mp, .set_hww_esr_report_mask = gk20a_gr_set_hww_esr_report_mask, .rops = gf100_gr_rops, .ppc_nr = 1, .grctx = &gk20a_grctx, + .zbc = &gf100_gr_zbc, .sclass = { { -1, -1, FERMI_TWOD_A }, { -1, -1, KEPLER_INLINE_TO_MEMORY_A }, diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gm107.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gm107.c index 2c67fac576d1..92e31d397207 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gm107.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gm107.c @@ -25,6 +25,8 @@ #include "ctxgf100.h" #include <subdev/bios.h> +#include <subdev/bios/bit.h> +#include <subdev/bios/init.h> #include <subdev/bios/P0260.h> #include <subdev/fb.h> @@ -36,6 +38,10 @@ static const struct gf100_gr_init gm107_gr_init_main_0[] = { + { 0x40880c, 1, 0x04, 0x00000000 }, + { 0x408910, 1, 0x04, 0x00000000 }, + { 0x408984, 1, 0x04, 0x00000000 }, + { 0x41a8a0, 1, 0x04, 0x00000000 }, { 0x400080, 1, 0x04, 0x003003c2 }, { 0x400088, 1, 0x04, 0x0001bfe7 }, { 0x40008c, 1, 0x04, 0x00060000 }, @@ -210,14 +216,13 @@ gm107_gr_init_cbm_0[] = { static const struct gf100_gr_init gm107_gr_init_be_0[] = { { 0x408890, 1, 0x04, 0x000000ff }, - { 0x40880c, 1, 0x04, 0x00000000 }, { 0x408850, 1, 0x04, 0x00000004 }, { 0x408878, 1, 0x04, 0x00c81603 }, { 0x40887c, 1, 0x04, 0x80543432 }, { 0x408880, 1, 0x04, 0x0010581e }, { 0x408884, 1, 0x04, 0x00001205 }, { 0x408974, 1, 0x04, 0x000000ff }, - { 0x408910, 9, 0x04, 0x00000000 }, + { 0x408914, 8, 0x04, 0x00000000 }, { 0x408950, 1, 0x04, 0x00000000 }, { 0x408954, 1, 0x04, 0x0000ffff }, { 0x408958, 1, 0x04, 0x00000034 }, @@ -227,7 +232,6 @@ gm107_gr_init_be_0[] = { { 0x408968, 1, 0x04, 0x02808833 }, { 0x40896c, 1, 0x04, 0x01f02438 }, { 0x408970, 1, 0x04, 0x00012c00 }, - { 0x408984, 1, 0x04, 0x00000000 }, { 0x408988, 1, 0x04, 0x08040201 }, { 0x40898c, 1, 0x04, 0x80402010 }, {} @@ -260,6 +264,7 @@ gm107_gr_pack_mmio[] = { { gf100_gr_init_gpm_0 }, { gm107_gr_init_gpc_unk_1 }, { gf100_gr_init_gcc_0 }, + { gk104_gr_init_gpc_unk_2 }, { gm107_gr_init_tpccs_0 }, { gm107_gr_init_tex_0 }, { gm107_gr_init_pe_0 }, @@ -280,6 +285,52 @@ gm107_gr_pack_mmio[] = { ******************************************************************************/ void +gm107_gr_init_400054(struct gf100_gr *gr) +{ + nvkm_wr32(gr->base.engine.subdev.device, 0x400054, 0x2c350f63); +} + +void +gm107_gr_init_shader_exceptions(struct gf100_gr *gr, int gpc, int tpc) +{ + struct nvkm_device *device = gr->base.engine.subdev.device; + nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x644), 0x00dffffe); + nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x64c), 0x00000005); +} + +void +gm107_gr_init_504430(struct gf100_gr *gr, int gpc, int tpc) +{ + struct nvkm_device *device = gr->base.engine.subdev.device; + nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x430), 0xc0000000); +} + +static void +gm107_gr_init_bios_2(struct gf100_gr *gr) +{ + struct nvkm_subdev *subdev = &gr->base.engine.subdev; + struct nvkm_device *device = subdev->device; + struct nvkm_bios *bios = device->bios; + struct bit_entry bit_P; + if (!bit_entry(bios, 'P', &bit_P) && + bit_P.version == 2 && bit_P.length >= 0x2c) { + u32 data = nvbios_rd32(bios, bit_P.offset + 0x28); + if (data) { + u8 ver = nvbios_rd08(bios, data + 0x00); + u8 hdr = nvbios_rd08(bios, data + 0x01); + if (ver == 0x20 && hdr >= 8) { + data = nvbios_rd32(bios, data + 0x04); + if (data) { + u32 save = nvkm_rd32(device, 0x619444); + nvbios_init(subdev, data); + nvkm_wr32(device, 0x619444, save); + } + } + } + } +} + +void gm107_gr_init_bios(struct gf100_gr *gr) { static const struct { @@ -308,115 +359,17 @@ gm107_gr_init_bios(struct gf100_gr *gr) } } -static int -gm107_gr_init(struct gf100_gr *gr) +static void +gm107_gr_init_gpc_mmu(struct gf100_gr *gr) { struct nvkm_device *device = gr->base.engine.subdev.device; struct nvkm_fb *fb = device->fb; - const u32 magicgpc918 = DIV_ROUND_UP(0x00800000, gr->tpc_total); - u32 data[TPC_MAX / 8] = {}; - u8 tpcnr[GPC_MAX]; - int gpc, tpc, rop; - int i; nvkm_wr32(device, GPC_BCAST(0x0880), 0x00000000); nvkm_wr32(device, GPC_BCAST(0x0890), 0x00000000); nvkm_wr32(device, GPC_BCAST(0x0894), 0x00000000); nvkm_wr32(device, GPC_BCAST(0x08b4), nvkm_memory_addr(fb->mmu_wr) >> 8); nvkm_wr32(device, GPC_BCAST(0x08b8), nvkm_memory_addr(fb->mmu_rd) >> 8); - - gf100_gr_mmio(gr, gr->func->mmio); - - gm107_gr_init_bios(gr); - - nvkm_wr32(device, GPC_UNIT(0, 0x3018), 0x00000001); - - memset(data, 0x00, sizeof(data)); - memcpy(tpcnr, gr->tpc_nr, sizeof(gr->tpc_nr)); - for (i = 0, gpc = -1; i < gr->tpc_total; i++) { - do { - gpc = (gpc + 1) % gr->gpc_nr; - } while (!tpcnr[gpc]); - tpc = gr->tpc_nr[gpc] - tpcnr[gpc]--; - - data[i / 8] |= tpc << ((i % 8) * 4); - } - - nvkm_wr32(device, GPC_BCAST(0x0980), data[0]); - nvkm_wr32(device, GPC_BCAST(0x0984), data[1]); - nvkm_wr32(device, GPC_BCAST(0x0988), data[2]); - nvkm_wr32(device, GPC_BCAST(0x098c), data[3]); - - for (gpc = 0; gpc < gr->gpc_nr; gpc++) { - nvkm_wr32(device, GPC_UNIT(gpc, 0x0914), - gr->screen_tile_row_offset << 8 | gr->tpc_nr[gpc]); - nvkm_wr32(device, GPC_UNIT(gpc, 0x0910), 0x00040000 | - gr->tpc_total); - nvkm_wr32(device, GPC_UNIT(gpc, 0x0918), magicgpc918); - } - - nvkm_wr32(device, GPC_BCAST(0x3fd4), magicgpc918); - nvkm_wr32(device, GPC_BCAST(0x08ac), nvkm_rd32(device, 0x100800)); - - gr->func->init_rop_active_fbps(gr); - - nvkm_wr32(device, 0x400500, 0x00010001); - - nvkm_wr32(device, 0x400100, 0xffffffff); - nvkm_wr32(device, 0x40013c, 0xffffffff); - nvkm_wr32(device, 0x400124, 0x00000002); - nvkm_wr32(device, 0x409c24, 0x000e0000); - - nvkm_wr32(device, 0x404000, 0xc0000000); - nvkm_wr32(device, 0x404600, 0xc0000000); - nvkm_wr32(device, 0x408030, 0xc0000000); - nvkm_wr32(device, 0x404490, 0xc0000000); - nvkm_wr32(device, 0x406018, 0xc0000000); - nvkm_wr32(device, 0x407020, 0x40000000); - nvkm_wr32(device, 0x405840, 0xc0000000); - nvkm_wr32(device, 0x405844, 0x00ffffff); - nvkm_mask(device, 0x419cc0, 0x00000008, 0x00000008); - - gr->func->init_ppc_exceptions(gr); - - for (gpc = 0; gpc < gr->gpc_nr; gpc++) { - nvkm_wr32(device, GPC_UNIT(gpc, 0x0420), 0xc0000000); - nvkm_wr32(device, GPC_UNIT(gpc, 0x0900), 0xc0000000); - nvkm_wr32(device, GPC_UNIT(gpc, 0x1028), 0xc0000000); - nvkm_wr32(device, GPC_UNIT(gpc, 0x0824), 0xc0000000); - for (tpc = 0; tpc < gr->tpc_nr[gpc]; tpc++) { - nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x508), 0xffffffff); - nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x50c), 0xffffffff); - nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x224), 0xc0000000); - nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x48c), 0xc0000000); - nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x084), 0xc0000000); - nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x430), 0xc0000000); - nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x644), 0x00dffffe); - nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x64c), 0x00000005); - } - nvkm_wr32(device, GPC_UNIT(gpc, 0x2c90), 0xffffffff); - nvkm_wr32(device, GPC_UNIT(gpc, 0x2c94), 0xffffffff); - } - - for (rop = 0; rop < gr->rop_nr; rop++) { - nvkm_wr32(device, ROP_UNIT(rop, 0x144), 0x40000000); - nvkm_wr32(device, ROP_UNIT(rop, 0x070), 0x40000000); - nvkm_wr32(device, ROP_UNIT(rop, 0x204), 0xffffffff); - nvkm_wr32(device, ROP_UNIT(rop, 0x208), 0xffffffff); - } - - nvkm_wr32(device, 0x400108, 0xffffffff); - nvkm_wr32(device, 0x400138, 0xffffffff); - nvkm_wr32(device, 0x400118, 0xffffffff); - nvkm_wr32(device, 0x400130, 0xffffffff); - nvkm_wr32(device, 0x40011c, 0xffffffff); - nvkm_wr32(device, 0x400134, 0xffffffff); - - nvkm_wr32(device, 0x400054, 0x2c350f63); - - gf100_gr_zbc_init(gr); - - return gf100_gr_init_ctxctl(gr); } #include "fuc/hubgm107.fuc5.h" @@ -441,15 +394,32 @@ gm107_gr_gpccs_ucode = { static const struct gf100_gr_func gm107_gr = { - .init = gm107_gr_init, + .oneinit_tiles = gf100_gr_oneinit_tiles, + .oneinit_sm_id = gf100_gr_oneinit_sm_id, + .init = gf100_gr_init, + .init_gpc_mmu = gm107_gr_init_gpc_mmu, + .init_bios = gm107_gr_init_bios, + .init_vsc_stream_master = gk104_gr_init_vsc_stream_master, + .init_zcull = gf117_gr_init_zcull, + .init_num_active_ltcs = gf100_gr_init_num_active_ltcs, .init_rop_active_fbps = gk104_gr_init_rop_active_fbps, + .init_bios_2 = gm107_gr_init_bios_2, + .init_fecs_exceptions = gf100_gr_init_fecs_exceptions, + .init_sked_hww_esr = gk104_gr_init_sked_hww_esr, + .init_419cc0 = gf100_gr_init_419cc0, .init_ppc_exceptions = gk104_gr_init_ppc_exceptions, + .init_tex_hww_esr = gf100_gr_init_tex_hww_esr, + .init_504430 = gm107_gr_init_504430, + .init_shader_exceptions = gm107_gr_init_shader_exceptions, + .init_400054 = gm107_gr_init_400054, + .trap_mp = gf100_gr_trap_mp, .mmio = gm107_gr_pack_mmio, .fecs.ucode = &gm107_gr_fecs_ucode, .gpccs.ucode = &gm107_gr_gpccs_ucode, .rops = gf100_gr_rops, .ppc_nr = 2, .grctx = &gm107_grctx, + .zbc = &gf100_gr_zbc, .sclass = { { -1, -1, FERMI_TWOD_A }, { -1, -1, KEPLER_INLINE_TO_MEMORY_B }, diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gm200.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gm200.c index 6435f1257572..eff30662b984 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gm200.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gm200.c @@ -39,6 +39,22 @@ gm200_gr_rops(struct gf100_gr *gr) } void +gm200_gr_init_ds_hww_esr_2(struct gf100_gr *gr) +{ + struct nvkm_device *device = gr->base.engine.subdev.device; + nvkm_wr32(device, 0x405848, 0xc0000000); + nvkm_mask(device, 0x40584c, 0x00000001, 0x00000001); +} + +void +gm200_gr_init_num_active_ltcs(struct gf100_gr *gr) +{ + struct nvkm_device *device = gr->base.engine.subdev.device; + nvkm_wr32(device, GPC_BCAST(0x08ac), nvkm_rd32(device, 0x100800)); + nvkm_wr32(device, GPC_BCAST(0x033c), nvkm_rd32(device, 0x100804)); +} + +void gm200_gr_init_gpc_mmu(struct gf100_gr *gr) { struct nvkm_device *device = gr->base.engine.subdev.device; @@ -61,111 +77,51 @@ gm200_gr_init_rop_active_fbps(struct gf100_gr *gr) nvkm_mask(device, 0x408958, 0x0000000f, fbp_count); /* crop */ } -int -gm200_gr_init(struct gf100_gr *gr) -{ - struct nvkm_device *device = gr->base.engine.subdev.device; - const u32 magicgpc918 = DIV_ROUND_UP(0x00800000, gr->tpc_total); - u32 data[TPC_MAX / 8] = {}; - u8 tpcnr[GPC_MAX]; - int gpc, tpc, rop; - int i; - - gr->func->init_gpc_mmu(gr); - - gf100_gr_mmio(gr, gr->fuc_sw_nonctx); - - gm107_gr_init_bios(gr); - - nvkm_wr32(device, GPC_UNIT(0, 0x3018), 0x00000001); - - memset(data, 0x00, sizeof(data)); - memcpy(tpcnr, gr->tpc_nr, sizeof(gr->tpc_nr)); - for (i = 0, gpc = -1; i < gr->tpc_total; i++) { - do { - gpc = (gpc + 1) % gr->gpc_nr; - } while (!tpcnr[gpc]); - tpc = gr->tpc_nr[gpc] - tpcnr[gpc]--; - - data[i / 8] |= tpc << ((i % 8) * 4); - } - - nvkm_wr32(device, GPC_BCAST(0x0980), data[0]); - nvkm_wr32(device, GPC_BCAST(0x0984), data[1]); - nvkm_wr32(device, GPC_BCAST(0x0988), data[2]); - nvkm_wr32(device, GPC_BCAST(0x098c), data[3]); - - for (gpc = 0; gpc < gr->gpc_nr; gpc++) { - nvkm_wr32(device, GPC_UNIT(gpc, 0x0914), - gr->screen_tile_row_offset << 8 | gr->tpc_nr[gpc]); - nvkm_wr32(device, GPC_UNIT(gpc, 0x0910), 0x00040000 | - gr->tpc_total); - nvkm_wr32(device, GPC_UNIT(gpc, 0x0918), magicgpc918); - } +static u8 +gm200_gr_tile_map_6_24[] = { + 0, 1, 2, 3, 4, 5, 3, 4, 5, 0, 1, 2, 0, 1, 2, 3, 4, 5, 3, 4, 5, 0, 1, 2, +}; - nvkm_wr32(device, GPC_BCAST(0x3fd4), magicgpc918); - nvkm_wr32(device, GPC_BCAST(0x08ac), nvkm_rd32(device, 0x100800)); - nvkm_wr32(device, GPC_BCAST(0x033c), nvkm_rd32(device, 0x100804)); +static u8 +gm200_gr_tile_map_4_16[] = { + 0, 1, 2, 3, 2, 3, 0, 1, 3, 0, 1, 2, 1, 2, 3, 0, +}; - gr->func->init_rop_active_fbps(gr); +static u8 +gm200_gr_tile_map_2_8[] = { + 0, 1, 1, 0, 0, 1, 1, 0, +}; - nvkm_wr32(device, 0x400500, 0x00010001); - nvkm_wr32(device, 0x400100, 0xffffffff); - nvkm_wr32(device, 0x40013c, 0xffffffff); - nvkm_wr32(device, 0x400124, 0x00000002); - nvkm_wr32(device, 0x409c24, 0x000e0000); - nvkm_wr32(device, 0x405848, 0xc0000000); - nvkm_wr32(device, 0x40584c, 0x00000001); - nvkm_wr32(device, 0x404000, 0xc0000000); - nvkm_wr32(device, 0x404600, 0xc0000000); - nvkm_wr32(device, 0x408030, 0xc0000000); - nvkm_wr32(device, 0x404490, 0xc0000000); - nvkm_wr32(device, 0x406018, 0xc0000000); - nvkm_wr32(device, 0x407020, 0x40000000); - nvkm_wr32(device, 0x405840, 0xc0000000); - nvkm_wr32(device, 0x405844, 0x00ffffff); - nvkm_mask(device, 0x419cc0, 0x00000008, 0x00000008); - - gr->func->init_ppc_exceptions(gr); - - for (gpc = 0; gpc < gr->gpc_nr; gpc++) { - nvkm_wr32(device, GPC_UNIT(gpc, 0x0420), 0xc0000000); - nvkm_wr32(device, GPC_UNIT(gpc, 0x0900), 0xc0000000); - nvkm_wr32(device, GPC_UNIT(gpc, 0x1028), 0xc0000000); - nvkm_wr32(device, GPC_UNIT(gpc, 0x0824), 0xc0000000); - for (tpc = 0; tpc < gr->tpc_nr[gpc]; tpc++) { - nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x508), 0xffffffff); - nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x50c), 0xffffffff); - nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x224), 0xc0000000); - nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x48c), 0xc0000000); - nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x084), 0xc0000000); - nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x430), 0xc0000000); - nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x644), 0x00dffffe); - nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x64c), 0x00000005); - } - nvkm_wr32(device, GPC_UNIT(gpc, 0x2c90), 0xffffffff); - nvkm_wr32(device, GPC_UNIT(gpc, 0x2c94), 0xffffffff); - } +void +gm200_gr_oneinit_sm_id(struct gf100_gr *gr) +{ + /*XXX: There's a different algorithm here I've not yet figured out. */ + gf100_gr_oneinit_sm_id(gr); +} - for (rop = 0; rop < gr->rop_nr; rop++) { - nvkm_wr32(device, ROP_UNIT(rop, 0x144), 0x40000000); - nvkm_wr32(device, ROP_UNIT(rop, 0x070), 0x40000000); - nvkm_wr32(device, ROP_UNIT(rop, 0x204), 0xffffffff); - nvkm_wr32(device, ROP_UNIT(rop, 0x208), 0xffffffff); +void +gm200_gr_oneinit_tiles(struct gf100_gr *gr) +{ + /*XXX: Not sure what this is about. The algorithm from NVGPU + * seems to work for all boards I tried from earlier (and + * later) GPUs except in these specific configurations. + * + * Let's just hardcode them for now. + */ + if (gr->gpc_nr == 2 && gr->tpc_total == 8) { + memcpy(gr->tile, gm200_gr_tile_map_2_8, gr->tpc_total); + gr->screen_tile_row_offset = 1; + } else + if (gr->gpc_nr == 4 && gr->tpc_total == 16) { + memcpy(gr->tile, gm200_gr_tile_map_4_16, gr->tpc_total); + gr->screen_tile_row_offset = 4; + } else + if (gr->gpc_nr == 6 && gr->tpc_total == 24) { + memcpy(gr->tile, gm200_gr_tile_map_6_24, gr->tpc_total); + gr->screen_tile_row_offset = 5; + } else { + gf100_gr_oneinit_tiles(gr); } - - nvkm_wr32(device, 0x400108, 0xffffffff); - nvkm_wr32(device, 0x400138, 0xffffffff); - nvkm_wr32(device, 0x400118, 0xffffffff); - nvkm_wr32(device, 0x400130, 0xffffffff); - nvkm_wr32(device, 0x40011c, 0xffffffff); - nvkm_wr32(device, 0x400134, 0xffffffff); - - nvkm_wr32(device, 0x400054, 0x2c350f63); - - gf100_gr_zbc_init(gr); - - return gf100_gr_init_ctxctl(gr); } int @@ -208,13 +164,30 @@ gm200_gr_new_(const struct gf100_gr_func *func, struct nvkm_device *device, static const struct gf100_gr_func gm200_gr = { - .init = gm200_gr_init, + .oneinit_tiles = gm200_gr_oneinit_tiles, + .oneinit_sm_id = gm200_gr_oneinit_sm_id, + .init = gf100_gr_init, .init_gpc_mmu = gm200_gr_init_gpc_mmu, + .init_bios = gm107_gr_init_bios, + .init_vsc_stream_master = gk104_gr_init_vsc_stream_master, + .init_zcull = gf117_gr_init_zcull, + .init_num_active_ltcs = gm200_gr_init_num_active_ltcs, .init_rop_active_fbps = gm200_gr_init_rop_active_fbps, + .init_fecs_exceptions = gf100_gr_init_fecs_exceptions, + .init_ds_hww_esr_2 = gm200_gr_init_ds_hww_esr_2, + .init_sked_hww_esr = gk104_gr_init_sked_hww_esr, + .init_419cc0 = gf100_gr_init_419cc0, .init_ppc_exceptions = gk104_gr_init_ppc_exceptions, + .init_tex_hww_esr = gf100_gr_init_tex_hww_esr, + .init_504430 = gm107_gr_init_504430, + .init_shader_exceptions = gm107_gr_init_shader_exceptions, + .init_400054 = gm107_gr_init_400054, + .trap_mp = gf100_gr_trap_mp, .rops = gm200_gr_rops, + .tpc_nr = 4, .ppc_nr = 2, .grctx = &gm200_grctx, + .zbc = &gf100_gr_zbc, .sclass = { { -1, -1, FERMI_TWOD_A }, { -1, -1, KEPLER_INLINE_TO_MEMORY_B }, diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gm20b.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gm20b.c index 69479af1d829..a667770ce3cb 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gm20b.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gm20b.c @@ -64,13 +64,18 @@ gm20b_gr_set_hww_esr_report_mask(struct gf100_gr *gr) static const struct gf100_gr_func gm20b_gr = { + .oneinit_tiles = gm200_gr_oneinit_tiles, + .oneinit_sm_id = gm200_gr_oneinit_sm_id, .init = gk20a_gr_init, + .init_zcull = gf117_gr_init_zcull, .init_gpc_mmu = gm20b_gr_init_gpc_mmu, .init_rop_active_fbps = gk104_gr_init_rop_active_fbps, + .trap_mp = gf100_gr_trap_mp, .set_hww_esr_report_mask = gm20b_gr_set_hww_esr_report_mask, .rops = gm200_gr_rops, .ppc_nr = 1, .grctx = &gm20b_grctx, + .zbc = &gf100_gr_zbc, .sclass = { { -1, -1, FERMI_TWOD_A }, { -1, -1, KEPLER_INLINE_TO_MEMORY_B }, diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gp100.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gp100.c index 867a5f7cc5bc..9d0521ce309a 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gp100.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gp100.c @@ -29,143 +29,103 @@ /******************************************************************************* * PGRAPH engine/subdev functions ******************************************************************************/ - void -gp100_gr_init_rop_active_fbps(struct gf100_gr *gr) +gp100_gr_zbc_clear_color(struct gf100_gr *gr, int zbc) { struct nvkm_device *device = gr->base.engine.subdev.device; - /*XXX: otherwise identical to gm200 aside from mask.. do everywhere? */ - const u32 fbp_count = nvkm_rd32(device, 0x12006c) & 0x0000000f; - nvkm_mask(device, 0x408850, 0x0000000f, fbp_count); /* zrop */ - nvkm_mask(device, 0x408958, 0x0000000f, fbp_count); /* crop */ + const int znum = zbc - 1; + const u32 zoff = znum * 4; + + if (gr->zbc_color[zbc].format) { + nvkm_wr32(device, 0x418010 + zoff, gr->zbc_color[zbc].ds[0]); + nvkm_wr32(device, 0x41804c + zoff, gr->zbc_color[zbc].ds[1]); + nvkm_wr32(device, 0x418088 + zoff, gr->zbc_color[zbc].ds[2]); + nvkm_wr32(device, 0x4180c4 + zoff, gr->zbc_color[zbc].ds[3]); + } + + nvkm_mask(device, 0x418100 + ((znum / 4) * 4), + 0x0000007f << ((znum % 4) * 7), + gr->zbc_color[zbc].format << ((znum % 4) * 7)); } void -gp100_gr_init_num_active_ltcs(struct gf100_gr *gr) +gp100_gr_zbc_clear_depth(struct gf100_gr *gr, int zbc) { struct nvkm_device *device = gr->base.engine.subdev.device; - - nvkm_wr32(device, GPC_BCAST(0x08ac), nvkm_rd32(device, 0x100800)); - nvkm_wr32(device, GPC_BCAST(0x033c), nvkm_rd32(device, 0x100804)); + const int znum = zbc - 1; + const u32 zoff = znum * 4; + + if (gr->zbc_depth[zbc].format) + nvkm_wr32(device, 0x418110 + zoff, gr->zbc_depth[zbc].ds); + nvkm_mask(device, 0x41814c + ((znum / 4) * 4), + 0x0000007f << ((znum % 4) * 7), + gr->zbc_depth[zbc].format << ((znum % 4) * 7)); } -int -gp100_gr_init(struct gf100_gr *gr) +static const struct gf100_gr_func_zbc +gp100_gr_zbc = { + .clear_color = gp100_gr_zbc_clear_color, + .clear_depth = gp100_gr_zbc_clear_depth, +}; + +void +gp100_gr_init_shader_exceptions(struct gf100_gr *gr, int gpc, int tpc) { struct nvkm_device *device = gr->base.engine.subdev.device; - const u32 magicgpc918 = DIV_ROUND_UP(0x00800000, gr->tpc_total); - u32 data[TPC_MAX / 8] = {}; - u8 tpcnr[GPC_MAX]; - int gpc, tpc, rop; - int i; - - gr->func->init_gpc_mmu(gr); - - gf100_gr_mmio(gr, gr->fuc_sw_nonctx); - - nvkm_wr32(device, GPC_UNIT(0, 0x3018), 0x00000001); - - memset(data, 0x00, sizeof(data)); - memcpy(tpcnr, gr->tpc_nr, sizeof(gr->tpc_nr)); - for (i = 0, gpc = -1; i < gr->tpc_total; i++) { - do { - gpc = (gpc + 1) % gr->gpc_nr; - } while (!tpcnr[gpc]); - tpc = gr->tpc_nr[gpc] - tpcnr[gpc]--; - - data[i / 8] |= tpc << ((i % 8) * 4); - } - - nvkm_wr32(device, GPC_BCAST(0x0980), data[0]); - nvkm_wr32(device, GPC_BCAST(0x0984), data[1]); - nvkm_wr32(device, GPC_BCAST(0x0988), data[2]); - nvkm_wr32(device, GPC_BCAST(0x098c), data[3]); - - for (gpc = 0; gpc < gr->gpc_nr; gpc++) { - nvkm_wr32(device, GPC_UNIT(gpc, 0x0914), - gr->screen_tile_row_offset << 8 | gr->tpc_nr[gpc]); - nvkm_wr32(device, GPC_UNIT(gpc, 0x0910), 0x00040000 | - gr->tpc_total); - nvkm_wr32(device, GPC_UNIT(gpc, 0x0918), magicgpc918); - } - - nvkm_wr32(device, GPC_BCAST(0x3fd4), magicgpc918); - gr->func->init_num_active_ltcs(gr); - - gr->func->init_rop_active_fbps(gr); - if (gr->func->init_swdx_pes_mask) - gr->func->init_swdx_pes_mask(gr); - - nvkm_wr32(device, 0x400500, 0x00010001); - nvkm_wr32(device, 0x400100, 0xffffffff); - nvkm_wr32(device, 0x40013c, 0xffffffff); - nvkm_wr32(device, 0x400124, 0x00000002); - nvkm_wr32(device, 0x409c24, 0x000f0002); - nvkm_wr32(device, 0x405848, 0xc0000000); - nvkm_mask(device, 0x40584c, 0x00000000, 0x00000001); - nvkm_wr32(device, 0x404000, 0xc0000000); - nvkm_wr32(device, 0x404600, 0xc0000000); - nvkm_wr32(device, 0x408030, 0xc0000000); - nvkm_wr32(device, 0x404490, 0xc0000000); - nvkm_wr32(device, 0x406018, 0xc0000000); - nvkm_wr32(device, 0x407020, 0x40000000); - nvkm_wr32(device, 0x405840, 0xc0000000); - nvkm_wr32(device, 0x405844, 0x00ffffff); - nvkm_mask(device, 0x419cc0, 0x00000008, 0x00000008); + nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x644), 0x00dffffe); + nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x64c), 0x00000105); +} +static void +gp100_gr_init_419c9c(struct gf100_gr *gr) +{ + struct nvkm_device *device = gr->base.engine.subdev.device; nvkm_mask(device, 0x419c9c, 0x00010000, 0x00010000); nvkm_mask(device, 0x419c9c, 0x00020000, 0x00020000); +} - gr->func->init_ppc_exceptions(gr); - - for (gpc = 0; gpc < gr->gpc_nr; gpc++) { - nvkm_wr32(device, GPC_UNIT(gpc, 0x0420), 0xc0000000); - nvkm_wr32(device, GPC_UNIT(gpc, 0x0900), 0xc0000000); - nvkm_wr32(device, GPC_UNIT(gpc, 0x1028), 0xc0000000); - nvkm_wr32(device, GPC_UNIT(gpc, 0x0824), 0xc0000000); - for (tpc = 0; tpc < gr->tpc_nr[gpc]; tpc++) { - nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x508), 0xffffffff); - nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x50c), 0xffffffff); - nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x224), 0xc0000000); - nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x48c), 0xc0000000); - nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x084), 0xc0000000); - nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x430), 0xc0000000); - nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x644), 0x00dffffe); - nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x64c), 0x00000105); - } - nvkm_wr32(device, GPC_UNIT(gpc, 0x2c90), 0xffffffff); - nvkm_wr32(device, GPC_UNIT(gpc, 0x2c94), 0xffffffff); - } - - for (rop = 0; rop < gr->rop_nr; rop++) { - nvkm_wr32(device, ROP_UNIT(rop, 0x144), 0x40000000); - nvkm_wr32(device, ROP_UNIT(rop, 0x070), 0x40000000); - nvkm_wr32(device, ROP_UNIT(rop, 0x204), 0xffffffff); - nvkm_wr32(device, ROP_UNIT(rop, 0x208), 0xffffffff); - } - - nvkm_wr32(device, 0x400108, 0xffffffff); - nvkm_wr32(device, 0x400138, 0xffffffff); - nvkm_wr32(device, 0x400118, 0xffffffff); - nvkm_wr32(device, 0x400130, 0xffffffff); - nvkm_wr32(device, 0x40011c, 0xffffffff); - nvkm_wr32(device, 0x400134, 0xffffffff); - - gf100_gr_zbc_init(gr); +void +gp100_gr_init_fecs_exceptions(struct gf100_gr *gr) +{ + nvkm_wr32(gr->base.engine.subdev.device, 0x409c24, 0x000f0002); +} - return gf100_gr_init_ctxctl(gr); +void +gp100_gr_init_rop_active_fbps(struct gf100_gr *gr) +{ + struct nvkm_device *device = gr->base.engine.subdev.device; + /*XXX: otherwise identical to gm200 aside from mask.. do everywhere? */ + const u32 fbp_count = nvkm_rd32(device, 0x12006c) & 0x0000000f; + nvkm_mask(device, 0x408850, 0x0000000f, fbp_count); /* zrop */ + nvkm_mask(device, 0x408958, 0x0000000f, fbp_count); /* crop */ } static const struct gf100_gr_func gp100_gr = { - .init = gp100_gr_init, + .oneinit_tiles = gm200_gr_oneinit_tiles, + .oneinit_sm_id = gm200_gr_oneinit_sm_id, + .init = gf100_gr_init, .init_gpc_mmu = gm200_gr_init_gpc_mmu, + .init_vsc_stream_master = gk104_gr_init_vsc_stream_master, + .init_zcull = gf117_gr_init_zcull, + .init_num_active_ltcs = gm200_gr_init_num_active_ltcs, .init_rop_active_fbps = gp100_gr_init_rop_active_fbps, + .init_fecs_exceptions = gp100_gr_init_fecs_exceptions, + .init_ds_hww_esr_2 = gm200_gr_init_ds_hww_esr_2, + .init_sked_hww_esr = gk104_gr_init_sked_hww_esr, + .init_419cc0 = gf100_gr_init_419cc0, + .init_419c9c = gp100_gr_init_419c9c, .init_ppc_exceptions = gk104_gr_init_ppc_exceptions, - .init_num_active_ltcs = gp100_gr_init_num_active_ltcs, + .init_tex_hww_esr = gf100_gr_init_tex_hww_esr, + .init_504430 = gm107_gr_init_504430, + .init_shader_exceptions = gp100_gr_init_shader_exceptions, + .trap_mp = gf100_gr_trap_mp, .rops = gm200_gr_rops, + .gpc_nr = 6, + .tpc_nr = 5, .ppc_nr = 2, .grctx = &gp100_grctx, + .zbc = &gp100_gr_zbc, .sclass = { { -1, -1, FERMI_TWOD_A }, { -1, -1, KEPLER_INLINE_TO_MEMORY_B }, diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gp102.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gp102.c index 61e3a0b08559..37f7d739bf80 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gp102.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gp102.c @@ -26,6 +26,62 @@ #include <nvif/class.h> +static void +gp102_gr_zbc_clear_stencil(struct gf100_gr *gr, int zbc) +{ + struct nvkm_device *device = gr->base.engine.subdev.device; + const int znum = zbc - 1; + const u32 zoff = znum * 4; + + if (gr->zbc_stencil[zbc].format) + nvkm_wr32(device, 0x41815c + zoff, gr->zbc_stencil[zbc].ds); + nvkm_mask(device, 0x418198 + ((znum / 4) * 4), + 0x0000007f << ((znum % 4) * 7), + gr->zbc_stencil[zbc].format << ((znum % 4) * 7)); +} + +static int +gp102_gr_zbc_stencil_get(struct gf100_gr *gr, int format, + const u32 ds, const u32 l2) +{ + struct nvkm_ltc *ltc = gr->base.engine.subdev.device->ltc; + int zbc = -ENOSPC, i; + + for (i = ltc->zbc_min; i <= ltc->zbc_max; i++) { + if (gr->zbc_stencil[i].format) { + if (gr->zbc_stencil[i].format != format) + continue; + if (gr->zbc_stencil[i].ds != ds) + continue; + if (gr->zbc_stencil[i].l2 != l2) { + WARN_ON(1); + return -EINVAL; + } + return i; + } else { + zbc = (zbc < 0) ? i : zbc; + } + } + + if (zbc < 0) + return zbc; + + gr->zbc_stencil[zbc].format = format; + gr->zbc_stencil[zbc].ds = ds; + gr->zbc_stencil[zbc].l2 = l2; + nvkm_ltc_zbc_stencil_get(ltc, zbc, l2); + gr->func->zbc->clear_stencil(gr, zbc); + return zbc; +} + +const struct gf100_gr_func_zbc +gp102_gr_zbc = { + .clear_color = gp100_gr_zbc_clear_color, + .clear_depth = gp100_gr_zbc_clear_depth, + .stencil_get = gp102_gr_zbc_stencil_get, + .clear_stencil = gp102_gr_zbc_clear_stencil, +}; + void gp102_gr_init_swdx_pes_mask(struct gf100_gr *gr) { @@ -42,15 +98,30 @@ gp102_gr_init_swdx_pes_mask(struct gf100_gr *gr) static const struct gf100_gr_func gp102_gr = { - .init = gp100_gr_init, + .oneinit_tiles = gm200_gr_oneinit_tiles, + .oneinit_sm_id = gm200_gr_oneinit_sm_id, + .init = gf100_gr_init, .init_gpc_mmu = gm200_gr_init_gpc_mmu, + .init_vsc_stream_master = gk104_gr_init_vsc_stream_master, + .init_zcull = gf117_gr_init_zcull, + .init_num_active_ltcs = gm200_gr_init_num_active_ltcs, .init_rop_active_fbps = gp100_gr_init_rop_active_fbps, - .init_ppc_exceptions = gk104_gr_init_ppc_exceptions, .init_swdx_pes_mask = gp102_gr_init_swdx_pes_mask, - .init_num_active_ltcs = gp100_gr_init_num_active_ltcs, + .init_fecs_exceptions = gp100_gr_init_fecs_exceptions, + .init_ds_hww_esr_2 = gm200_gr_init_ds_hww_esr_2, + .init_sked_hww_esr = gk104_gr_init_sked_hww_esr, + .init_419cc0 = gf100_gr_init_419cc0, + .init_ppc_exceptions = gk104_gr_init_ppc_exceptions, + .init_tex_hww_esr = gf100_gr_init_tex_hww_esr, + .init_504430 = gm107_gr_init_504430, + .init_shader_exceptions = gp100_gr_init_shader_exceptions, + .trap_mp = gf100_gr_trap_mp, .rops = gm200_gr_rops, + .gpc_nr = 6, + .tpc_nr = 5, .ppc_nr = 3, .grctx = &gp102_grctx, + .zbc = &gp102_gr_zbc, .sclass = { { -1, -1, FERMI_TWOD_A }, { -1, -1, KEPLER_INLINE_TO_MEMORY_B }, diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gp104.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gp104.c new file mode 100644 index 000000000000..4573c914c021 --- /dev/null +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gp104.c @@ -0,0 +1,66 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "gf100.h" +#include "ctxgf100.h" + +#include <nvif/class.h> + +static const struct gf100_gr_func +gp104_gr = { + .oneinit_tiles = gm200_gr_oneinit_tiles, + .oneinit_sm_id = gm200_gr_oneinit_sm_id, + .init = gf100_gr_init, + .init_gpc_mmu = gm200_gr_init_gpc_mmu, + .init_vsc_stream_master = gk104_gr_init_vsc_stream_master, + .init_zcull = gf117_gr_init_zcull, + .init_num_active_ltcs = gm200_gr_init_num_active_ltcs, + .init_rop_active_fbps = gp100_gr_init_rop_active_fbps, + .init_swdx_pes_mask = gp102_gr_init_swdx_pes_mask, + .init_fecs_exceptions = gp100_gr_init_fecs_exceptions, + .init_ds_hww_esr_2 = gm200_gr_init_ds_hww_esr_2, + .init_sked_hww_esr = gk104_gr_init_sked_hww_esr, + .init_419cc0 = gf100_gr_init_419cc0, + .init_ppc_exceptions = gk104_gr_init_ppc_exceptions, + .init_tex_hww_esr = gf100_gr_init_tex_hww_esr, + .init_504430 = gm107_gr_init_504430, + .init_shader_exceptions = gp100_gr_init_shader_exceptions, + .trap_mp = gf100_gr_trap_mp, + .rops = gm200_gr_rops, + .gpc_nr = 6, + .tpc_nr = 5, + .ppc_nr = 3, + .grctx = &gp104_grctx, + .zbc = &gp102_gr_zbc, + .sclass = { + { -1, -1, FERMI_TWOD_A }, + { -1, -1, KEPLER_INLINE_TO_MEMORY_B }, + { -1, -1, PASCAL_B, &gf100_fermi }, + { -1, -1, PASCAL_COMPUTE_B }, + {} + } +}; + +int +gp104_gr_new(struct nvkm_device *device, int index, struct nvkm_gr **pgr) +{ + return gm200_gr_new_(&gp104_gr, device, index, pgr); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gp107.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gp107.c index f7272323f694..812aba91653f 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gp107.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gp107.c @@ -28,15 +28,30 @@ static const struct gf100_gr_func gp107_gr = { - .init = gp100_gr_init, + .oneinit_tiles = gm200_gr_oneinit_tiles, + .oneinit_sm_id = gm200_gr_oneinit_sm_id, + .init = gf100_gr_init, .init_gpc_mmu = gm200_gr_init_gpc_mmu, + .init_vsc_stream_master = gk104_gr_init_vsc_stream_master, + .init_zcull = gf117_gr_init_zcull, + .init_num_active_ltcs = gm200_gr_init_num_active_ltcs, .init_rop_active_fbps = gp100_gr_init_rop_active_fbps, - .init_ppc_exceptions = gk104_gr_init_ppc_exceptions, .init_swdx_pes_mask = gp102_gr_init_swdx_pes_mask, - .init_num_active_ltcs = gp100_gr_init_num_active_ltcs, + .init_fecs_exceptions = gp100_gr_init_fecs_exceptions, + .init_ds_hww_esr_2 = gm200_gr_init_ds_hww_esr_2, + .init_sked_hww_esr = gk104_gr_init_sked_hww_esr, + .init_419cc0 = gf100_gr_init_419cc0, + .init_ppc_exceptions = gk104_gr_init_ppc_exceptions, + .init_tex_hww_esr = gf100_gr_init_tex_hww_esr, + .init_504430 = gm107_gr_init_504430, + .init_shader_exceptions = gp100_gr_init_shader_exceptions, + .trap_mp = gf100_gr_trap_mp, .rops = gm200_gr_rops, + .gpc_nr = 2, + .tpc_nr = 3, .ppc_nr = 1, .grctx = &gp107_grctx, + .zbc = &gp102_gr_zbc, .sclass = { { -1, -1, FERMI_TWOD_A }, { -1, -1, KEPLER_INLINE_TO_MEMORY_B }, diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gp10b.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gp10b.c index 5f3d161a0842..303dceddd4a8 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gp10b.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gp10b.c @@ -25,24 +25,31 @@ #include <nvif/class.h> -static void -gp10b_gr_init_num_active_ltcs(struct gf100_gr *gr) -{ - struct nvkm_device *device = gr->base.engine.subdev.device; - - nvkm_wr32(device, GPC_BCAST(0x08ac), nvkm_rd32(device, 0x100800)); -} - static const struct gf100_gr_func gp10b_gr = { - .init = gp100_gr_init, + .oneinit_tiles = gm200_gr_oneinit_tiles, + .oneinit_sm_id = gm200_gr_oneinit_sm_id, + .init = gf100_gr_init, .init_gpc_mmu = gm200_gr_init_gpc_mmu, + .init_vsc_stream_master = gk104_gr_init_vsc_stream_master, + .init_zcull = gf117_gr_init_zcull, + .init_num_active_ltcs = gf100_gr_init_num_active_ltcs, .init_rop_active_fbps = gp100_gr_init_rop_active_fbps, + .init_fecs_exceptions = gp100_gr_init_fecs_exceptions, + .init_ds_hww_esr_2 = gm200_gr_init_ds_hww_esr_2, + .init_sked_hww_esr = gk104_gr_init_sked_hww_esr, + .init_419cc0 = gf100_gr_init_419cc0, .init_ppc_exceptions = gk104_gr_init_ppc_exceptions, - .init_num_active_ltcs = gp10b_gr_init_num_active_ltcs, + .init_tex_hww_esr = gf100_gr_init_tex_hww_esr, + .init_504430 = gm107_gr_init_504430, + .init_shader_exceptions = gp100_gr_init_shader_exceptions, + .trap_mp = gf100_gr_trap_mp, .rops = gm200_gr_rops, + .gpc_nr = 1, + .tpc_nr = 2, .ppc_nr = 1, .grctx = &gp102_grctx, + .zbc = &gp102_gr_zbc, .sclass = { { -1, -1, FERMI_TWOD_A }, { -1, -1, KEPLER_INLINE_TO_MEMORY_B }, diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gv100.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gv100.c new file mode 100644 index 000000000000..19173ea19096 --- /dev/null +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gv100.c @@ -0,0 +1,120 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "gf100.h" +#include "ctxgf100.h" + +#include <nvif/class.h> + +static void +gv100_gr_trap_mp(struct gf100_gr *gr, int gpc, int tpc) +{ + struct nvkm_subdev *subdev = &gr->base.engine.subdev; + struct nvkm_device *device = subdev->device; + u32 werr = nvkm_rd32(device, TPC_UNIT(gpc, tpc, 0x730)); + u32 gerr = nvkm_rd32(device, TPC_UNIT(gpc, tpc, 0x734)); + const struct nvkm_enum *warp; + char glob[128]; + + nvkm_snprintbf(glob, sizeof(glob), gf100_mp_global_error, gerr); + warp = nvkm_enum_find(gf100_mp_warp_error, werr & 0xffff); + + nvkm_error(subdev, "GPC%i/TPC%i/MP trap: " + "global %08x [%s] warp %04x [%s]\n", + gpc, tpc, gerr, glob, werr, warp ? warp->name : ""); + + nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x730), 0x00000000); + nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x734), gerr); +} + +static void +gv100_gr_init_4188a4(struct gf100_gr *gr) +{ + struct nvkm_device *device = gr->base.engine.subdev.device; + nvkm_mask(device, 0x4188a4, 0x03000000, 0x03000000); +} + +static void +gv100_gr_init_shader_exceptions(struct gf100_gr *gr, int gpc, int tpc) +{ + struct nvkm_device *device = gr->base.engine.subdev.device; + int sm; + for (sm = 0; sm < 0x100; sm += 0x80) { + nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x728 + sm), 0x0085eb64); + nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x610), 0x00000001); + nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x72c + sm), 0x00000004); + } +} + +static void +gv100_gr_init_504430(struct gf100_gr *gr, int gpc, int tpc) +{ + struct nvkm_device *device = gr->base.engine.subdev.device; + nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x430), 0x403f0000); +} + +static void +gv100_gr_init_419bd8(struct gf100_gr *gr) +{ + struct nvkm_device *device = gr->base.engine.subdev.device; + nvkm_mask(device, 0x419bd8, 0x00000700, 0x00000000); +} + +static const struct gf100_gr_func +gv100_gr = { + .oneinit_tiles = gm200_gr_oneinit_tiles, + .oneinit_sm_id = gm200_gr_oneinit_sm_id, + .init = gf100_gr_init, + .init_419bd8 = gv100_gr_init_419bd8, + .init_gpc_mmu = gm200_gr_init_gpc_mmu, + .init_vsc_stream_master = gk104_gr_init_vsc_stream_master, + .init_zcull = gf117_gr_init_zcull, + .init_num_active_ltcs = gm200_gr_init_num_active_ltcs, + .init_rop_active_fbps = gp100_gr_init_rop_active_fbps, + .init_swdx_pes_mask = gp102_gr_init_swdx_pes_mask, + .init_fecs_exceptions = gp100_gr_init_fecs_exceptions, + .init_ds_hww_esr_2 = gm200_gr_init_ds_hww_esr_2, + .init_sked_hww_esr = gk104_gr_init_sked_hww_esr, + .init_ppc_exceptions = gk104_gr_init_ppc_exceptions, + .init_504430 = gv100_gr_init_504430, + .init_shader_exceptions = gv100_gr_init_shader_exceptions, + .init_4188a4 = gv100_gr_init_4188a4, + .trap_mp = gv100_gr_trap_mp, + .rops = gm200_gr_rops, + .gpc_nr = 6, + .tpc_nr = 5, + .ppc_nr = 3, + .grctx = &gv100_grctx, + .zbc = &gp102_gr_zbc, + .sclass = { + { -1, -1, FERMI_TWOD_A }, + { -1, -1, KEPLER_INLINE_TO_MEMORY_B }, + { -1, -1, VOLTA_A, &gf100_fermi }, + { -1, -1, VOLTA_COMPUTE_A }, + {} + } +}; + +int +gv100_gr_new(struct nvkm_device *device, int index, struct nvkm_gr **pgr) +{ + return gm200_gr_new_(&gv100_gr, device, index, pgr); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/falcon/msgqueue.c b/drivers/gpu/drm/nouveau/nvkm/falcon/msgqueue.c index 58a59b7db2e5..771e16a16267 100644 --- a/drivers/gpu/drm/nouveau/nvkm/falcon/msgqueue.c +++ b/drivers/gpu/drm/nouveau/nvkm/falcon/msgqueue.c @@ -506,6 +506,7 @@ nvkm_msgqueue_new(u32 version, struct nvkm_falcon *falcon, break; case 0x0148cdec: case 0x015ccf3e: + case 0x0167d263: ret = msgqueue_0148cdec_new(falcon, sb, queue); break; default: diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/Kbuild b/drivers/gpu/drm/nouveau/nvkm/subdev/Kbuild index 3f5d38d74fba..cfdffef1afb9 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/Kbuild +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/Kbuild @@ -3,6 +3,7 @@ include $(src)/nvkm/subdev/bios/Kbuild include $(src)/nvkm/subdev/bus/Kbuild include $(src)/nvkm/subdev/clk/Kbuild include $(src)/nvkm/subdev/devinit/Kbuild +include $(src)/nvkm/subdev/fault/Kbuild include $(src)/nvkm/subdev/fb/Kbuild include $(src)/nvkm/subdev/fuse/Kbuild include $(src)/nvkm/subdev/gpio/Kbuild diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/bios/dp.c b/drivers/gpu/drm/nouveau/nvkm/subdev/bios/dp.c index 7c7efa4ea0d0..3133b28f849c 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/bios/dp.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/bios/dp.c @@ -25,7 +25,7 @@ #include <subdev/bios/bit.h> #include <subdev/bios/dp.h> -static u16 +u16 nvbios_dp_table(struct nvkm_bios *bios, u8 *ver, u8 *hdr, u8 *cnt, u8 *len) { struct bit_entry d; diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/bios/pll.c b/drivers/gpu/drm/nouveau/nvkm/subdev/bios/pll.c index 2ca23a9157ab..e6e804cee2bc 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/bios/pll.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/bios/pll.c @@ -193,7 +193,10 @@ pll_map_type(struct nvkm_bios *bios, u8 type, u32 *reg, u8 *ver, u8 *len) data += hdr; while (cnt--) { if (nvbios_rd08(bios, data + 0) == type) { - *reg = nvbios_rd32(bios, data + 3); + if (*ver < 0x50) + *reg = nvbios_rd32(bios, data + 3); + else + *reg = 0; return data; } data += *len; @@ -361,6 +364,20 @@ nvbios_pll_parse(struct nvkm_bios *bios, u32 type, struct nvbios_pll *info) info->min_p = nvbios_rd08(bios, data + 12); info->max_p = nvbios_rd08(bios, data + 13); break; + case 0x50: + info->refclk = nvbios_rd16(bios, data + 1) * 1000; + /* info->refclk_alt = nvbios_rd16(bios, data + 3) * 1000; */ + info->vco1.min_freq = nvbios_rd16(bios, data + 5) * 1000; + info->vco1.max_freq = nvbios_rd16(bios, data + 7) * 1000; + info->vco1.min_inputfreq = nvbios_rd16(bios, data + 9) * 1000; + info->vco1.max_inputfreq = nvbios_rd16(bios, data + 11) * 1000; + info->vco1.min_m = nvbios_rd08(bios, data + 13); + info->vco1.max_m = nvbios_rd08(bios, data + 14); + info->vco1.min_n = nvbios_rd08(bios, data + 15); + info->vco1.max_n = nvbios_rd08(bios, data + 16); + info->min_p = nvbios_rd08(bios, data + 17); + info->max_p = nvbios_rd08(bios, data + 18); + break; default: nvkm_error(subdev, "unknown pll limits version 0x%02x\n", ver); return -EINVAL; diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/bios/shadowramin.c b/drivers/gpu/drm/nouveau/nvkm/subdev/bios/shadowramin.c index 0f537c22804c..3634cd0630b8 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/bios/shadowramin.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/bios/shadowramin.c @@ -78,7 +78,10 @@ pramin_init(struct nvkm_bios *bios, const char *name) * important as we don't want to be touching vram on an * uninitialised board */ - addr = nvkm_rd32(device, 0x619f04); + if (device->card_type >= GV100) + addr = nvkm_rd32(device, 0x625f04); + else + addr = nvkm_rd32(device, 0x619f04); if (!(addr & 0x00000008)) { nvkm_debug(subdev, "... not enabled\n"); return ERR_PTR(-ENODEV); diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/clk/base.c b/drivers/gpu/drm/nouveau/nvkm/subdev/clk/base.c index 81c3567d4e67..ba6a868d4c95 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/clk/base.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/clk/base.c @@ -109,18 +109,17 @@ nvkm_cstate_valid(struct nvkm_clk *clk, struct nvkm_cstate *cstate, static struct nvkm_cstate * nvkm_cstate_find_best(struct nvkm_clk *clk, struct nvkm_pstate *pstate, - struct nvkm_cstate *start) + struct nvkm_cstate *cstate) { struct nvkm_device *device = clk->subdev.device; struct nvkm_volt *volt = device->volt; - struct nvkm_cstate *cstate; int max_volt; - if (!pstate || !start) + if (!pstate || !cstate) return NULL; if (!volt) - return start; + return cstate; max_volt = volt->max_uv; if (volt->max0_id != 0xff) @@ -133,8 +132,7 @@ nvkm_cstate_find_best(struct nvkm_clk *clk, struct nvkm_pstate *pstate, max_volt = min(max_volt, nvkm_volt_map(volt, volt->max2_id, clk->temp)); - for (cstate = start; &cstate->head != &pstate->list; - cstate = list_prev_entry(cstate, head)) { + list_for_each_entry_from_reverse(cstate, &pstate->list, head) { if (nvkm_cstate_valid(clk, cstate, max_volt, clk->temp)) break; } diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/Kbuild b/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/Kbuild index eac88e3dc6e5..50a436926484 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/Kbuild +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/Kbuild @@ -12,3 +12,4 @@ nvkm-y += nvkm/subdev/devinit/mcp89.o nvkm-y += nvkm/subdev/devinit/gf100.o nvkm-y += nvkm/subdev/devinit/gm107.o nvkm-y += nvkm/subdev/devinit/gm200.o +nvkm-y += nvkm/subdev/devinit/gv100.o diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/gm200.c b/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/gm200.c index 1730371933df..b80618e35491 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/gm200.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/gm200.c @@ -107,7 +107,7 @@ pmu_load(struct nv50_devinit *init, u8 type, bool post, return pmu_exec(init, pmu.init_addr_pmu), 0; } -static int +int gm200_devinit_post(struct nvkm_devinit *base, bool post) { struct nv50_devinit *init = nv50_devinit(base); diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/gv100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/gv100.c new file mode 100644 index 000000000000..fbde6828bd38 --- /dev/null +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/gv100.c @@ -0,0 +1,79 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "nv50.h" + +#include <subdev/bios.h> +#include <subdev/bios/pll.h> +#include <subdev/clk/pll.h> + +static int +gv100_devinit_pll_set(struct nvkm_devinit *init, u32 type, u32 freq) +{ + struct nvkm_subdev *subdev = &init->subdev; + struct nvkm_device *device = subdev->device; + struct nvbios_pll info; + int head = type - PLL_VPLL0; + int N, fN, M, P; + int ret; + + ret = nvbios_pll_parse(device->bios, type, &info); + if (ret) + return ret; + + ret = gt215_pll_calc(subdev, &info, freq, &N, &fN, &M, &P); + if (ret < 0) + return ret; + + switch (info.type) { + case PLL_VPLL0: + case PLL_VPLL1: + case PLL_VPLL2: + case PLL_VPLL3: + nvkm_wr32(device, 0x00ef10 + (head * 0x40), fN << 16); + nvkm_wr32(device, 0x00ef04 + (head * 0x40), (P << 16) | + (N << 8) | + (M << 0)); + break; + default: + nvkm_warn(subdev, "%08x/%dKhz unimplemented\n", type, freq); + ret = -EINVAL; + break; + } + + return ret; +} + +static const struct nvkm_devinit_func +gv100_devinit = { + .preinit = gf100_devinit_preinit, + .init = nv50_devinit_init, + .post = gm200_devinit_post, + .pll_set = gv100_devinit_pll_set, + .disable = gm107_devinit_disable, +}; + +int +gv100_devinit_new(struct nvkm_device *device, int index, + struct nvkm_devinit **pinit) +{ + return nv50_devinit_new_(&gv100_devinit, device, index, pinit); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/nv50.h b/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/nv50.h index 315ebaff1165..9b9f0dc1e192 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/nv50.h +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/nv50.h @@ -24,4 +24,6 @@ int gf100_devinit_pll_set(struct nvkm_devinit *, u32, u32); void gf100_devinit_preinit(struct nvkm_devinit *); u64 gm107_devinit_disable(struct nvkm_devinit *); + +int gm200_devinit_post(struct nvkm_devinit *, bool); #endif diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fault/Kbuild b/drivers/gpu/drm/nouveau/nvkm/subdev/fault/Kbuild new file mode 100644 index 000000000000..45bb46fb0929 --- /dev/null +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fault/Kbuild @@ -0,0 +1,3 @@ +nvkm-y += nvkm/subdev/fault/base.o +nvkm-y += nvkm/subdev/fault/gp100.o +nvkm-y += nvkm/subdev/fault/gv100.o diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fault/base.c b/drivers/gpu/drm/nouveau/nvkm/subdev/fault/base.c new file mode 100644 index 000000000000..007bf4af33b9 --- /dev/null +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fault/base.c @@ -0,0 +1,179 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "priv.h" + +#include <core/memory.h> +#include <core/notify.h> +#include <subdev/bar.h> +#include <subdev/mmu.h> + +static void +nvkm_fault_ntfy_fini(struct nvkm_event *event, int type, int index) +{ + struct nvkm_fault *fault = container_of(event, typeof(*fault), event); + fault->func->buffer.fini(fault->buffer[index]); +} + +static void +nvkm_fault_ntfy_init(struct nvkm_event *event, int type, int index) +{ + struct nvkm_fault *fault = container_of(event, typeof(*fault), event); + fault->func->buffer.init(fault->buffer[index]); +} + +static int +nvkm_fault_ntfy_ctor(struct nvkm_object *object, void *argv, u32 argc, + struct nvkm_notify *notify) +{ + struct nvkm_fault_buffer *buffer = nvkm_fault_buffer(object); + if (argc == 0) { + notify->size = 0; + notify->types = 1; + notify->index = buffer->id; + return 0; + } + return -ENOSYS; +} + +static const struct nvkm_event_func +nvkm_fault_ntfy = { + .ctor = nvkm_fault_ntfy_ctor, + .init = nvkm_fault_ntfy_init, + .fini = nvkm_fault_ntfy_fini, +}; + +static void +nvkm_fault_intr(struct nvkm_subdev *subdev) +{ + struct nvkm_fault *fault = nvkm_fault(subdev); + return fault->func->intr(fault); +} + +static int +nvkm_fault_fini(struct nvkm_subdev *subdev, bool suspend) +{ + struct nvkm_fault *fault = nvkm_fault(subdev); + if (fault->func->fini) + fault->func->fini(fault); + return 0; +} + +static int +nvkm_fault_init(struct nvkm_subdev *subdev) +{ + struct nvkm_fault *fault = nvkm_fault(subdev); + if (fault->func->init) + fault->func->init(fault); + return 0; +} + +static int +nvkm_fault_oneinit_buffer(struct nvkm_fault *fault, int id) +{ + struct nvkm_subdev *subdev = &fault->subdev; + struct nvkm_device *device = subdev->device; + struct nvkm_vmm *bar2 = nvkm_bar_bar2_vmm(device); + struct nvkm_fault_buffer *buffer; + int ret; + + if (!(buffer = kzalloc(sizeof(*buffer), GFP_KERNEL))) + return -ENOMEM; + buffer->fault = fault; + buffer->id = id; + buffer->entries = fault->func->buffer.entries(buffer); + fault->buffer[id] = buffer; + + nvkm_debug(subdev, "buffer %d: %d entries\n", id, buffer->entries); + + ret = nvkm_memory_new(device, NVKM_MEM_TARGET_INST, buffer->entries * + fault->func->buffer.entry_size, 0x1000, true, + &buffer->mem); + if (ret) + return ret; + + ret = nvkm_vmm_get(bar2, 12, nvkm_memory_size(buffer->mem), + &buffer->vma); + if (ret) + return ret; + + return nvkm_memory_map(buffer->mem, 0, bar2, buffer->vma, NULL, 0); +} + +static int +nvkm_fault_oneinit(struct nvkm_subdev *subdev) +{ + struct nvkm_fault *fault = nvkm_fault(subdev); + int ret, i; + + for (i = 0; i < ARRAY_SIZE(fault->buffer); i++) { + if (i < fault->func->buffer.nr) { + ret = nvkm_fault_oneinit_buffer(fault, i); + if (ret) + return ret; + fault->buffer_nr = i + 1; + } + } + + return nvkm_event_init(&nvkm_fault_ntfy, 1, fault->buffer_nr, + &fault->event); +} + +static void * +nvkm_fault_dtor(struct nvkm_subdev *subdev) +{ + struct nvkm_vmm *bar2 = nvkm_bar_bar2_vmm(subdev->device); + struct nvkm_fault *fault = nvkm_fault(subdev); + int i; + + nvkm_event_fini(&fault->event); + + for (i = 0; i < fault->buffer_nr; i++) { + if (fault->buffer[i]) { + nvkm_vmm_put(bar2, &fault->buffer[i]->vma); + nvkm_memory_unref(&fault->buffer[i]->mem); + kfree(fault->buffer[i]); + } + } + + return fault; +} + +static const struct nvkm_subdev_func +nvkm_fault = { + .dtor = nvkm_fault_dtor, + .oneinit = nvkm_fault_oneinit, + .init = nvkm_fault_init, + .fini = nvkm_fault_fini, + .intr = nvkm_fault_intr, +}; + +int +nvkm_fault_new_(const struct nvkm_fault_func *func, struct nvkm_device *device, + int index, struct nvkm_fault **pfault) +{ + struct nvkm_fault *fault; + if (!(fault = *pfault = kzalloc(sizeof(*fault), GFP_KERNEL))) + return -ENOMEM; + nvkm_subdev_ctor(&nvkm_fault, device, index, &fault->subdev); + fault->func = func; + return 0; +} diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fault/gp100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/fault/gp100.c new file mode 100644 index 000000000000..5e71db2e8d75 --- /dev/null +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fault/gp100.c @@ -0,0 +1,69 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "priv.h" + +#include <subdev/mmu.h> + +static void +gp100_fault_buffer_fini(struct nvkm_fault_buffer *buffer) +{ + struct nvkm_device *device = buffer->fault->subdev.device; + nvkm_mask(device, 0x002a70, 0x00000001, 0x00000000); +} + +static void +gp100_fault_buffer_init(struct nvkm_fault_buffer *buffer) +{ + struct nvkm_device *device = buffer->fault->subdev.device; + nvkm_wr32(device, 0x002a74, upper_32_bits(buffer->vma->addr)); + nvkm_wr32(device, 0x002a70, lower_32_bits(buffer->vma->addr)); + nvkm_mask(device, 0x002a70, 0x00000001, 0x00000001); +} + +static u32 +gp100_fault_buffer_entries(struct nvkm_fault_buffer *buffer) +{ + return nvkm_rd32(buffer->fault->subdev.device, 0x002a78); +} + +static void +gp100_fault_intr(struct nvkm_fault *fault) +{ + nvkm_event_send(&fault->event, 1, 0, NULL, 0); +} + +static const struct nvkm_fault_func +gp100_fault = { + .intr = gp100_fault_intr, + .buffer.nr = 1, + .buffer.entry_size = 32, + .buffer.entries = gp100_fault_buffer_entries, + .buffer.init = gp100_fault_buffer_init, + .buffer.fini = gp100_fault_buffer_fini, +}; + +int +gp100_fault_new(struct nvkm_device *device, int index, + struct nvkm_fault **pfault) +{ + return nvkm_fault_new_(&gp100_fault, device, index, pfault); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fault/gv100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/fault/gv100.c new file mode 100644 index 000000000000..73c7728b5969 --- /dev/null +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fault/gv100.c @@ -0,0 +1,206 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "priv.h" + +#include <core/memory.h> +#include <subdev/mmu.h> +#include <engine/fifo.h> + +static void +gv100_fault_buffer_process(struct nvkm_fault_buffer *buffer) +{ + struct nvkm_device *device = buffer->fault->subdev.device; + struct nvkm_memory *mem = buffer->mem; + const u32 foff = buffer->id * 0x14; + u32 get = nvkm_rd32(device, 0x100e2c + foff); + u32 put = nvkm_rd32(device, 0x100e30 + foff); + if (put == get) + return; + + nvkm_kmap(mem); + while (get != put) { + const u32 base = get * buffer->fault->func->buffer.entry_size; + const u32 instlo = nvkm_ro32(mem, base + 0x00); + const u32 insthi = nvkm_ro32(mem, base + 0x04); + const u32 addrlo = nvkm_ro32(mem, base + 0x08); + const u32 addrhi = nvkm_ro32(mem, base + 0x0c); + const u32 timelo = nvkm_ro32(mem, base + 0x10); + const u32 timehi = nvkm_ro32(mem, base + 0x14); + const u32 info0 = nvkm_ro32(mem, base + 0x18); + const u32 info1 = nvkm_ro32(mem, base + 0x1c); + struct nvkm_fault_data info; + + if (++get == buffer->entries) + get = 0; + nvkm_wr32(device, 0x100e2c + foff, get); + + info.addr = ((u64)addrhi << 32) | addrlo; + info.inst = ((u64)insthi << 32) | instlo; + info.time = ((u64)timehi << 32) | timelo; + info.engine = (info0 & 0x000000ff); + info.valid = (info1 & 0x80000000) >> 31; + info.gpc = (info1 & 0x1f000000) >> 24; + info.hub = (info1 & 0x00100000) >> 20; + info.access = (info1 & 0x000f0000) >> 16; + info.client = (info1 & 0x00007f00) >> 8; + info.reason = (info1 & 0x0000001f); + + nvkm_fifo_fault(device->fifo, &info); + } + nvkm_done(mem); +} + +static void +gv100_fault_buffer_fini(struct nvkm_fault_buffer *buffer) +{ + struct nvkm_device *device = buffer->fault->subdev.device; + const u32 intr = buffer->id ? 0x08000000 : 0x20000000; + const u32 foff = buffer->id * 0x14; + + nvkm_mask(device, 0x100a34, intr, intr); + nvkm_mask(device, 0x100e34 + foff, 0x80000000, 0x00000000); +} + +static void +gv100_fault_buffer_init(struct nvkm_fault_buffer *buffer) +{ + struct nvkm_device *device = buffer->fault->subdev.device; + const u32 intr = buffer->id ? 0x08000000 : 0x20000000; + const u32 foff = buffer->id * 0x14; + + nvkm_mask(device, 0x100e34 + foff, 0xc0000000, 0x40000000); + nvkm_wr32(device, 0x100e28 + foff, upper_32_bits(buffer->vma->addr)); + nvkm_wr32(device, 0x100e24 + foff, lower_32_bits(buffer->vma->addr)); + nvkm_mask(device, 0x100e34 + foff, 0x80000000, 0x80000000); + nvkm_mask(device, 0x100a2c, intr, intr); +} + +static u32 +gv100_fault_buffer_entries(struct nvkm_fault_buffer *buffer) +{ + struct nvkm_device *device = buffer->fault->subdev.device; + const u32 foff = buffer->id * 0x14; + nvkm_mask(device, 0x100e34 + foff, 0x40000000, 0x40000000); + return nvkm_rd32(device, 0x100e34 + foff) & 0x000fffff; +} + +static int +gv100_fault_ntfy_nrpfb(struct nvkm_notify *notify) +{ + struct nvkm_fault *fault = container_of(notify, typeof(*fault), nrpfb); + gv100_fault_buffer_process(fault->buffer[0]); + return NVKM_NOTIFY_KEEP; +} + +static void +gv100_fault_intr_fault(struct nvkm_fault *fault) +{ + struct nvkm_subdev *subdev = &fault->subdev; + struct nvkm_device *device = subdev->device; + struct nvkm_fault_data info; + const u32 addrlo = nvkm_rd32(device, 0x100e4c); + const u32 addrhi = nvkm_rd32(device, 0x100e50); + const u32 info0 = nvkm_rd32(device, 0x100e54); + const u32 insthi = nvkm_rd32(device, 0x100e58); + const u32 info1 = nvkm_rd32(device, 0x100e5c); + + info.addr = ((u64)addrhi << 32) | addrlo; + info.inst = ((u64)insthi << 32) | (info0 & 0xfffff000); + info.time = 0; + info.engine = (info0 & 0x000000ff); + info.valid = (info1 & 0x80000000) >> 31; + info.gpc = (info1 & 0x1f000000) >> 24; + info.hub = (info1 & 0x00100000) >> 20; + info.access = (info1 & 0x000f0000) >> 16; + info.client = (info1 & 0x00007f00) >> 8; + info.reason = (info1 & 0x0000001f); + + nvkm_fifo_fault(device->fifo, &info); +} + +static void +gv100_fault_intr(struct nvkm_fault *fault) +{ + struct nvkm_subdev *subdev = &fault->subdev; + struct nvkm_device *device = subdev->device; + u32 stat = nvkm_rd32(device, 0x100a20); + + if (stat & 0x80000000) { + gv100_fault_intr_fault(fault); + nvkm_wr32(device, 0x100e60, 0x80000000); + stat &= ~0x80000000; + } + + if (stat & 0x20000000) { + if (fault->buffer[0]) { + nvkm_event_send(&fault->event, 1, 0, NULL, 0); + stat &= ~0x20000000; + } + } + + if (stat) { + nvkm_debug(subdev, "intr %08x\n", stat); + } +} + +static void +gv100_fault_fini(struct nvkm_fault *fault) +{ + nvkm_notify_put(&fault->nrpfb); + nvkm_mask(fault->subdev.device, 0x100a34, 0x80000000, 0x80000000); +} + +static void +gv100_fault_init(struct nvkm_fault *fault) +{ + nvkm_mask(fault->subdev.device, 0x100a2c, 0x80000000, 0x80000000); + nvkm_notify_get(&fault->nrpfb); +} + +static const struct nvkm_fault_func +gv100_fault = { + .init = gv100_fault_init, + .fini = gv100_fault_fini, + .intr = gv100_fault_intr, + .buffer.nr = 2, + .buffer.entry_size = 32, + .buffer.entries = gv100_fault_buffer_entries, + .buffer.init = gv100_fault_buffer_init, + .buffer.fini = gv100_fault_buffer_fini, +}; + +int +gv100_fault_new(struct nvkm_device *device, int index, + struct nvkm_fault **pfault) +{ + struct nvkm_fault *fault; + int ret; + + ret = nvkm_fault_new_(&gv100_fault, device, index, &fault); + *pfault = fault; + if (ret) + return ret; + + return nvkm_notify_init(&fault->buffer[0]->object, &fault->event, + gv100_fault_ntfy_nrpfb, false, NULL, 0, 0, + &fault->nrpfb); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fault/priv.h b/drivers/gpu/drm/nouveau/nvkm/subdev/fault/priv.h new file mode 100644 index 000000000000..44843ecf12b0 --- /dev/null +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fault/priv.h @@ -0,0 +1,34 @@ +#ifndef __NVKM_FAULT_PRIV_H__ +#define __NVKM_FAULT_PRIV_H__ +#define nvkm_fault_buffer(p) container_of((p), struct nvkm_fault_buffer, object) +#define nvkm_fault(p) container_of((p), struct nvkm_fault, subdev) +#include <subdev/fault.h> + +#include <core/event.h> +#include <core/object.h> + +struct nvkm_fault_buffer { + struct nvkm_object object; + struct nvkm_fault *fault; + int id; + int entries; + struct nvkm_memory *mem; + struct nvkm_vma *vma; +}; + +int nvkm_fault_new_(const struct nvkm_fault_func *, struct nvkm_device *, + int index, struct nvkm_fault **); + +struct nvkm_fault_func { + void (*init)(struct nvkm_fault *); + void (*fini)(struct nvkm_fault *); + void (*intr)(struct nvkm_fault *); + struct { + int nr; + u32 entry_size; + u32 (*entries)(struct nvkm_fault_buffer *); + void (*init)(struct nvkm_fault_buffer *); + void (*fini)(struct nvkm_fault_buffer *); + } buffer; +}; +#endif diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/Kbuild b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/Kbuild index b4f22cce5d43..969610951263 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/Kbuild +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/Kbuild @@ -30,6 +30,7 @@ nvkm-y += nvkm/subdev/fb/gm20b.o nvkm-y += nvkm/subdev/fb/gp100.o nvkm-y += nvkm/subdev/fb/gp102.o nvkm-y += nvkm/subdev/fb/gp10b.o +nvkm-y += nvkm/subdev/fb/gv100.o nvkm-y += nvkm/subdev/fb/ram.o nvkm-y += nvkm/subdev/fb/ramnv04.o diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/gf100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/gf100.c index cdc4e0a2cc6b..e8dc4e913494 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/gf100.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/gf100.c @@ -46,10 +46,10 @@ gf100_fb_oneinit(struct nvkm_fb *base) { struct gf100_fb *fb = gf100_fb(base); struct nvkm_device *device = fb->base.subdev.device; - int ret, size = 0x1000; + int ret, size = 1 << (fb->base.page ? fb->base.page : 17); size = nvkm_longopt(device->cfgopt, "MmuDebugBufferSize", size); - size = min(size, 0x1000); + size = max(size, 0x1000); ret = nvkm_memory_new(device, NVKM_MEM_TARGET_INST, size, 0x1000, true, &fb->base.mmu_rd); diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/gm200.c b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/gm200.c index 8137e19d3292..d3b8c3367152 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/gm200.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/gm200.c @@ -49,8 +49,6 @@ gm200_fb_init(struct nvkm_fb *base) if (fb->r100c10_page) nvkm_wr32(device, 0x100c10, fb->r100c10 >> 8); - nvkm_mask(device, 0x100c80, 0x00000001, 0x00000000); /* 128KiB lpg */ - nvkm_wr32(device, 0x100cc8, nvkm_memory_addr(fb->base.mmu_wr) >> 8); nvkm_wr32(device, 0x100ccc, nvkm_memory_addr(fb->base.mmu_rd) >> 8); nvkm_mask(device, 0x100cc4, 0x00060000, diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/gp100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/gp100.c index 147f69b30cd8..dffe1f5e1071 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/gp100.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/gp100.c @@ -26,7 +26,7 @@ #include <core/memory.h> -static void +void gp100_fb_init_unkn(struct nvkm_fb *base) { struct nvkm_device *device = gf100_fb(base)->base.subdev.device; @@ -48,7 +48,7 @@ gp100_fb_init(struct nvkm_fb *base) nvkm_wr32(device, 0x100cc8, nvkm_memory_addr(fb->base.mmu_wr) >> 8); nvkm_wr32(device, 0x100ccc, nvkm_memory_addr(fb->base.mmu_rd) >> 8); nvkm_mask(device, 0x100cc4, 0x00060000, - max(nvkm_memory_size(fb->base.mmu_rd) >> 16, (u64)2) << 17); + min(nvkm_memory_size(fb->base.mmu_rd) >> 16, (u64)2) << 17); } static const struct nvkm_fb_func diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/gv100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/gv100.c new file mode 100644 index 000000000000..3c5e02e9794a --- /dev/null +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/gv100.c @@ -0,0 +1,46 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "gf100.h" +#include "ram.h" + +static int +gv100_fb_init_page(struct nvkm_fb *fb) +{ + return (fb->page == 16) ? 0 : -EINVAL; +} + +static const struct nvkm_fb_func +gv100_fb = { + .dtor = gf100_fb_dtor, + .oneinit = gf100_fb_oneinit, + .init = gp100_fb_init, + .init_page = gv100_fb_init_page, + .init_unkn = gp100_fb_init_unkn, + .ram_new = gp100_ram_new, + .default_bigpage = 16, +}; + +int +gv100_fb_new(struct nvkm_device *device, int index, struct nvkm_fb **pfb) +{ + return gf100_fb_new_(&gv100_fb, device, index, pfb); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/priv.h b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/priv.h index 414a423e0e55..2857f31466bf 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/priv.h +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/priv.h @@ -68,4 +68,6 @@ int gf100_fb_oneinit(struct nvkm_fb *); int gf100_fb_init_page(struct nvkm_fb *); int gm200_fb_init_page(struct nvkm_fb *); + +void gp100_fb_init_unkn(struct nvkm_fb *); #endif diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/Kbuild b/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/Kbuild index 12d6f4f102cb..290ff1c425a9 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/Kbuild +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/Kbuild @@ -4,3 +4,4 @@ nvkm-y += nvkm/subdev/ltc/gk104.o nvkm-y += nvkm/subdev/ltc/gm107.o nvkm-y += nvkm/subdev/ltc/gm200.o nvkm-y += nvkm/subdev/ltc/gp100.o +nvkm-y += nvkm/subdev/ltc/gp102.o diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/base.c b/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/base.c index 1f185274d3e6..23242179e600 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/base.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/base.c @@ -55,6 +55,14 @@ nvkm_ltc_zbc_depth_get(struct nvkm_ltc *ltc, int index, const u32 depth) return index; } +int +nvkm_ltc_zbc_stencil_get(struct nvkm_ltc *ltc, int index, const u32 stencil) +{ + ltc->zbc_stencil[index] = stencil; + ltc->func->zbc_clear_stencil(ltc, index, stencil); + return index; +} + void nvkm_ltc_invalidate(struct nvkm_ltc *ltc) { @@ -92,6 +100,8 @@ nvkm_ltc_init(struct nvkm_subdev *subdev) for (i = ltc->zbc_min; i <= ltc->zbc_max; i++) { ltc->func->zbc_clear_color(ltc, i, ltc->zbc_color[i]); ltc->func->zbc_clear_depth(ltc, i, ltc->zbc_depth[i]); + if (ltc->func->zbc_clear_stencil) + ltc->func->zbc_clear_stencil(ltc, i, ltc->zbc_stencil[i]); } ltc->func->init(ltc); diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/gp100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/gp100.c index e34d42108019..e923ed76d37a 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/gp100.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/gp100.c @@ -23,7 +23,7 @@ */ #include "priv.h" -static void +void gp100_ltc_intr(struct nvkm_ltc *ltc) { struct nvkm_device *device = ltc->subdev.device; @@ -38,7 +38,7 @@ gp100_ltc_intr(struct nvkm_ltc *ltc) } } -static int +int gp100_ltc_oneinit(struct nvkm_ltc *ltc) { struct nvkm_device *device = ltc->subdev.device; @@ -48,7 +48,7 @@ gp100_ltc_oneinit(struct nvkm_ltc *ltc) return 0; } -static void +void gp100_ltc_init(struct nvkm_ltc *ltc) { /*XXX: PMU LS call to setup tagram address */ diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/gp102.c b/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/gp102.c new file mode 100644 index 000000000000..601747ada655 --- /dev/null +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/gp102.c @@ -0,0 +1,51 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "priv.h" + +void +gp102_ltc_zbc_clear_stencil(struct nvkm_ltc *ltc, int i, const u32 stencil) +{ + struct nvkm_device *device = ltc->subdev.device; + nvkm_mask(device, 0x17e338, 0x0000000f, i); + nvkm_wr32(device, 0x17e204, stencil); +} + +static const struct nvkm_ltc_func +gp102_ltc = { + .oneinit = gp100_ltc_oneinit, + .init = gp100_ltc_init, + .intr = gp100_ltc_intr, + .cbc_clear = gm107_ltc_cbc_clear, + .cbc_wait = gm107_ltc_cbc_wait, + .zbc = 16, + .zbc_clear_color = gm107_ltc_zbc_clear_color, + .zbc_clear_depth = gm107_ltc_zbc_clear_depth, + .zbc_clear_stencil = gp102_ltc_zbc_clear_stencil, + .invalidate = gf100_ltc_invalidate, + .flush = gf100_ltc_flush, +}; + +int +gp102_ltc_new(struct nvkm_device *device, int index, struct nvkm_ltc **pltc) +{ + return nvkm_ltc_new_(&gp102_ltc, device, index, pltc); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/priv.h b/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/priv.h index e71cc25cc775..9dcde43c0f3c 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/priv.h +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/priv.h @@ -19,6 +19,7 @@ struct nvkm_ltc_func { int zbc; void (*zbc_clear_color)(struct nvkm_ltc *, int, const u32[4]); void (*zbc_clear_depth)(struct nvkm_ltc *, int, const u32); + void (*zbc_clear_stencil)(struct nvkm_ltc *, int, const u32); void (*invalidate)(struct nvkm_ltc *); void (*flush)(struct nvkm_ltc *); @@ -41,4 +42,8 @@ void gm107_ltc_cbc_clear(struct nvkm_ltc *, u32, u32); void gm107_ltc_cbc_wait(struct nvkm_ltc *); void gm107_ltc_zbc_clear_color(struct nvkm_ltc *, int, const u32[4]); void gm107_ltc_zbc_clear_depth(struct nvkm_ltc *, int, const u32); + +int gp100_ltc_oneinit(struct nvkm_ltc *); +void gp100_ltc_init(struct nvkm_ltc *); +void gp100_ltc_intr(struct nvkm_ltc *); #endif diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mc/gp100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/mc/gp100.c index 7321ad3758c3..43db245eec9a 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/mc/gp100.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mc/gp100.c @@ -75,10 +75,28 @@ gp100_mc_intr_mask(struct nvkm_mc *base, u32 mask, u32 intr) spin_unlock_irqrestore(&mc->lock, flags); } +const struct nvkm_mc_map +gp100_mc_intr[] = { + { 0x04000000, NVKM_ENGINE_DISP }, + { 0x00000100, NVKM_ENGINE_FIFO }, + { 0x00000200, NVKM_SUBDEV_FAULT }, + { 0x40000000, NVKM_SUBDEV_IBUS }, + { 0x10000000, NVKM_SUBDEV_BUS }, + { 0x08000000, NVKM_SUBDEV_FB }, + { 0x02000000, NVKM_SUBDEV_LTC }, + { 0x01000000, NVKM_SUBDEV_PMU }, + { 0x00200000, NVKM_SUBDEV_GPIO }, + { 0x00200000, NVKM_SUBDEV_I2C }, + { 0x00100000, NVKM_SUBDEV_TIMER }, + { 0x00040000, NVKM_SUBDEV_THERM }, + { 0x00002000, NVKM_SUBDEV_FB }, + {}, +}; + static const struct nvkm_mc_func gp100_mc = { .init = nv50_mc_init, - .intr = gk104_mc_intr, + .intr = gp100_mc_intr, .intr_unarm = gp100_mc_intr_unarm, .intr_rearm = gp100_mc_intr_rearm, .intr_mask = gp100_mc_intr_mask, diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mc/gp10b.c b/drivers/gpu/drm/nouveau/nvkm/subdev/mc/gp10b.c index 2283e3b74277..ff8629de97d6 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/mc/gp10b.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mc/gp10b.c @@ -34,7 +34,7 @@ gp10b_mc_init(struct nvkm_mc *mc) static const struct nvkm_mc_func gp10b_mc = { .init = gp10b_mc_init, - .intr = gk104_mc_intr, + .intr = gp100_mc_intr, .intr_unarm = gp100_mc_intr_unarm, .intr_rearm = gp100_mc_intr_rearm, .intr_mask = gp100_mc_intr_mask, diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mc/priv.h b/drivers/gpu/drm/nouveau/nvkm/subdev/mc/priv.h index 8869d79c2b59..d9e3691d45b7 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/mc/priv.h +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mc/priv.h @@ -57,4 +57,6 @@ int gp100_mc_new_(const struct nvkm_mc_func *, struct nvkm_device *, int, extern const struct nvkm_mc_map gk104_mc_intr[]; extern const struct nvkm_mc_map gk104_mc_reset[]; + +extern const struct nvkm_mc_map gp100_mc_intr[]; #endif diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/Kbuild b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/Kbuild index 67ee983bb026..58a24e3a0598 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/Kbuild +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/Kbuild @@ -12,6 +12,7 @@ nvkm-y += nvkm/subdev/mmu/gm200.o nvkm-y += nvkm/subdev/mmu/gm20b.o nvkm-y += nvkm/subdev/mmu/gp100.o nvkm-y += nvkm/subdev/mmu/gp10b.o +nvkm-y += nvkm/subdev/mmu/gv100.o nvkm-y += nvkm/subdev/mmu/mem.o nvkm-y += nvkm/subdev/mmu/memnv04.o @@ -31,6 +32,7 @@ nvkm-y += nvkm/subdev/mmu/vmmgm200.o nvkm-y += nvkm/subdev/mmu/vmmgm20b.o nvkm-y += nvkm/subdev/mmu/vmmgp100.o nvkm-y += nvkm/subdev/mmu/vmmgp10b.o +nvkm-y += nvkm/subdev/mmu/vmmgv100.o nvkm-y += nvkm/subdev/mmu/umem.o nvkm-y += nvkm/subdev/mmu/ummu.o diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/gv100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/gv100.c new file mode 100644 index 000000000000..f666cb57f69e --- /dev/null +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/gv100.c @@ -0,0 +1,43 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "mem.h" +#include "vmm.h" + +#include <core/option.h> + +#include <nvif/class.h> + +static const struct nvkm_mmu_func +gv100_mmu = { + .dma_bits = 47, + .mmu = {{ -1, -1, NVIF_CLASS_MMU_GF100}}, + .mem = {{ -1, 0, NVIF_CLASS_MEM_GF100}, gf100_mem_new, gf100_mem_map }, + .vmm = {{ -1, -1, NVIF_CLASS_VMM_GP100}, gv100_vmm_new }, + .kind = gm200_mmu_kind, + .kind_sys = true, +}; + +int +gv100_mmu_new(struct nvkm_device *device, int index, struct nvkm_mmu **pmmu) +{ + return nvkm_mmu_new_(&gv100_mmu, device, index, pmmu); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmm.h b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmm.h index da06e64d8a7d..1a3b0a3724ca 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmm.h +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmm.h @@ -236,6 +236,9 @@ int gp100_vmm_new(struct nvkm_mmu *, u64, u64, void *, u32, int gp10b_vmm_new(struct nvkm_mmu *, u64, u64, void *, u32, struct lock_class_key *, const char *, struct nvkm_vmm **); +int gv100_vmm_new(struct nvkm_mmu *, u64, u64, void *, u32, + struct lock_class_key *, const char *, + struct nvkm_vmm **); #define VMM_PRINT(l,v,p,f,a...) do { \ struct nvkm_vmm *_vmm = (v); \ diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmgv100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmgv100.c new file mode 100644 index 000000000000..2fa40c16e6d2 --- /dev/null +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmgv100.c @@ -0,0 +1,87 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "vmm.h" + +#include <subdev/fb.h> +#include <subdev/ltc.h> + +#include <nvif/ifc00d.h> +#include <nvif/unpack.h> + +int +gv100_vmm_join(struct nvkm_vmm *vmm, struct nvkm_memory *inst) +{ + u64 data[2], mask; + int ret = gp100_vmm_join(vmm, inst), i; + if (ret) + return ret; + + nvkm_kmap(inst); + data[0] = nvkm_ro32(inst, 0x200); + data[1] = nvkm_ro32(inst, 0x204); + mask = BIT_ULL(0); + + nvkm_wo32(inst, 0x21c, 0x00000000); + + for (i = 0; i < 64; i++) { + if (mask & BIT_ULL(i)) { + nvkm_wo32(inst, 0x2a4 + (i * 0x10), data[1]); + nvkm_wo32(inst, 0x2a0 + (i * 0x10), data[0]); + } else { + nvkm_wo32(inst, 0x2a4 + (i * 0x10), 0x00000001); + nvkm_wo32(inst, 0x2a0 + (i * 0x10), 0x00000001); + } + nvkm_wo32(inst, 0x2a8 + (i * 0x10), 0x00000000); + } + + nvkm_wo32(inst, 0x298, lower_32_bits(mask)); + nvkm_wo32(inst, 0x29c, upper_32_bits(mask)); + nvkm_done(inst); + return 0; +} + +static const struct nvkm_vmm_func +gv100_vmm = { + .join = gv100_vmm_join, + .part = gf100_vmm_part, + .aper = gf100_vmm_aper, + .valid = gp100_vmm_valid, + .flush = gp100_vmm_flush, + .page = { + { 47, &gp100_vmm_desc_16[4], NVKM_VMM_PAGE_Sxxx }, + { 38, &gp100_vmm_desc_16[3], NVKM_VMM_PAGE_Sxxx }, + { 29, &gp100_vmm_desc_16[2], NVKM_VMM_PAGE_Sxxx }, + { 21, &gp100_vmm_desc_16[1], NVKM_VMM_PAGE_SVxC }, + { 16, &gp100_vmm_desc_16[0], NVKM_VMM_PAGE_SVxC }, + { 12, &gp100_vmm_desc_12[0], NVKM_VMM_PAGE_SVHx }, + {} + } +}; + +int +gv100_vmm_new(struct nvkm_mmu *mmu, u64 addr, u64 size, void *argv, u32 argc, + struct lock_class_key *key, const char *name, + struct nvkm_vmm **pvmm) +{ + return nv04_vmm_new_(&gv100_vmm, mmu, 0, addr, size, + argv, argc, key, name, pvmm); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/secboot/gp108.c b/drivers/gpu/drm/nouveau/nvkm/subdev/secboot/gp108.c index e8c27ec700de..737a8d50a1f2 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/secboot/gp108.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/secboot/gp108.c @@ -65,3 +65,24 @@ MODULE_FIRMWARE("nvidia/gp108/nvdec/scrubber.bin"); MODULE_FIRMWARE("nvidia/gp108/sec2/desc.bin"); MODULE_FIRMWARE("nvidia/gp108/sec2/image.bin"); MODULE_FIRMWARE("nvidia/gp108/sec2/sig.bin"); + +MODULE_FIRMWARE("nvidia/gv100/acr/bl.bin"); +MODULE_FIRMWARE("nvidia/gv100/acr/unload_bl.bin"); +MODULE_FIRMWARE("nvidia/gv100/acr/ucode_load.bin"); +MODULE_FIRMWARE("nvidia/gv100/acr/ucode_unload.bin"); +MODULE_FIRMWARE("nvidia/gv100/gr/fecs_bl.bin"); +MODULE_FIRMWARE("nvidia/gv100/gr/fecs_inst.bin"); +MODULE_FIRMWARE("nvidia/gv100/gr/fecs_data.bin"); +MODULE_FIRMWARE("nvidia/gv100/gr/fecs_sig.bin"); +MODULE_FIRMWARE("nvidia/gv100/gr/gpccs_bl.bin"); +MODULE_FIRMWARE("nvidia/gv100/gr/gpccs_inst.bin"); +MODULE_FIRMWARE("nvidia/gv100/gr/gpccs_data.bin"); +MODULE_FIRMWARE("nvidia/gv100/gr/gpccs_sig.bin"); +MODULE_FIRMWARE("nvidia/gv100/gr/sw_ctx.bin"); +MODULE_FIRMWARE("nvidia/gv100/gr/sw_nonctx.bin"); +MODULE_FIRMWARE("nvidia/gv100/gr/sw_bundle_init.bin"); +MODULE_FIRMWARE("nvidia/gv100/gr/sw_method_init.bin"); +MODULE_FIRMWARE("nvidia/gv100/nvdec/scrubber.bin"); +MODULE_FIRMWARE("nvidia/gv100/sec2/desc.bin"); +MODULE_FIRMWARE("nvidia/gv100/sec2/image.bin"); +MODULE_FIRMWARE("nvidia/gv100/sec2/sig.bin"); diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/secboot/ls_ucode_msgqueue.c b/drivers/gpu/drm/nouveau/nvkm/subdev/secboot/ls_ucode_msgqueue.c index 6f10b098676c..1e1f1c635cab 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/secboot/ls_ucode_msgqueue.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/secboot/ls_ucode_msgqueue.c @@ -80,12 +80,11 @@ acr_ls_msgqueue_post_run(struct nvkm_msgqueue *queue, struct nvkm_falcon *falcon, u32 addr_args) { struct nvkm_device *device = falcon->owner->device; - u32 cmdline_size = NVKM_MSGQUEUE_CMDLINE_SIZE; - u8 buf[cmdline_size]; + u8 buf[NVKM_MSGQUEUE_CMDLINE_SIZE]; - memset(buf, 0, cmdline_size); + memset(buf, 0, sizeof(buf)); nvkm_msgqueue_write_cmdline(queue, buf); - nvkm_falcon_load_dmem(falcon, buf, addr_args, cmdline_size, 0); + nvkm_falcon_load_dmem(falcon, buf, addr_args, sizeof(buf), 0); /* rearm the queue so it will wait for the init message */ nvkm_msgqueue_reinit(queue); diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/top/gk104.c b/drivers/gpu/drm/nouveau/nvkm/subdev/top/gk104.c index fea4957291da..4f1f3e890650 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/top/gk104.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/top/gk104.c @@ -48,7 +48,8 @@ gk104_top_oneinit(struct nvkm_top *top) case 0x00000001: /* DATA */ inst = (data & 0x3c000000) >> 26; info->addr = (data & 0x00fff000); - info->fault = (data & 0x000000f8) >> 3; + if (data & 0x00000004) + info->fault = (data & 0x000003f8) >> 3; break; case 0x00000002: /* ENUM */ if (data & 0x00000020) diff --git a/drivers/gpu/drm/omapdrm/dss/dispc.c b/drivers/gpu/drm/omapdrm/dss/dispc.c index 5e2e65e88847..7f3ac6b13b56 100644 --- a/drivers/gpu/drm/omapdrm/dss/dispc.c +++ b/drivers/gpu/drm/omapdrm/dss/dispc.c @@ -828,6 +828,12 @@ static void dispc_ovl_set_scale_coef(struct dispc_device *dispc, h_coef = dispc_ovl_get_scale_coef(fir_hinc, true); v_coef = dispc_ovl_get_scale_coef(fir_vinc, five_taps); + if (!h_coef || !v_coef) { + dev_err(&dispc->pdev->dev, "%s: failed to find scale coefs\n", + __func__); + return; + } + for (i = 0; i < 8; i++) { u32 h, hv; @@ -2342,7 +2348,7 @@ static int dispc_ovl_calc_scaling_24xx(struct dispc_device *dispc, } if (in_width > maxsinglelinewidth) { - DSSERR("Cannot scale max input width exceeded"); + DSSERR("Cannot scale max input width exceeded\n"); return -EINVAL; } return 0; @@ -2424,13 +2430,13 @@ again: } if (in_width > (maxsinglelinewidth * 2)) { - DSSERR("Cannot setup scaling"); - DSSERR("width exceeds maximum width possible"); + DSSERR("Cannot setup scaling\n"); + DSSERR("width exceeds maximum width possible\n"); return -EINVAL; } if (in_width > maxsinglelinewidth && *five_taps) { - DSSERR("cannot setup scaling with five taps"); + DSSERR("cannot setup scaling with five taps\n"); return -EINVAL; } return 0; @@ -2472,7 +2478,7 @@ static int dispc_ovl_calc_scaling_44xx(struct dispc_device *dispc, in_width > maxsinglelinewidth && ++*decim_x); if (in_width > maxsinglelinewidth) { - DSSERR("Cannot scale width exceeds max line width"); + DSSERR("Cannot scale width exceeds max line width\n"); return -EINVAL; } @@ -2490,7 +2496,7 @@ static int dispc_ovl_calc_scaling_44xx(struct dispc_device *dispc, * bandwidth. Despite what theory says this appears to * be true also for 16-bit color formats. */ - DSSERR("Not enough bandwidth, too much downscaling (x-decimation factor %d > 4)", *decim_x); + DSSERR("Not enough bandwidth, too much downscaling (x-decimation factor %d > 4)\n", *decim_x); return -EINVAL; } @@ -4633,7 +4639,7 @@ static int dispc_errata_i734_wa_init(struct dispc_device *dispc) i734_buf.size, &i734_buf.paddr, GFP_KERNEL); if (!i734_buf.vaddr) { - dev_err(&dispc->pdev->dev, "%s: dma_alloc_writecombine failed", + dev_err(&dispc->pdev->dev, "%s: dma_alloc_writecombine failed\n", __func__); return -ENOMEM; } diff --git a/drivers/gpu/drm/omapdrm/dss/hdmi4.c b/drivers/gpu/drm/omapdrm/dss/hdmi4.c index 97c88861d67a..5879f45f6fc9 100644 --- a/drivers/gpu/drm/omapdrm/dss/hdmi4.c +++ b/drivers/gpu/drm/omapdrm/dss/hdmi4.c @@ -679,7 +679,7 @@ static int hdmi_audio_config(struct device *dev, struct omap_dss_audio *dss_audio) { struct omap_hdmi *hd = dev_get_drvdata(dev); - int ret; + int ret = 0; mutex_lock(&hd->lock); diff --git a/drivers/gpu/drm/omapdrm/dss/hdmi4_core.c b/drivers/gpu/drm/omapdrm/dss/hdmi4_core.c index 35ed2add6189..813ba42f2753 100644 --- a/drivers/gpu/drm/omapdrm/dss/hdmi4_core.c +++ b/drivers/gpu/drm/omapdrm/dss/hdmi4_core.c @@ -922,8 +922,13 @@ int hdmi4_core_init(struct platform_device *pdev, struct hdmi_core_data *core) { const struct hdmi4_features *features; struct resource *res; + const struct soc_device_attribute *soc; - features = soc_device_match(hdmi4_soc_devices)->data; + soc = soc_device_match(hdmi4_soc_devices); + if (!soc) + return -ENODEV; + + features = soc->data; core->cts_swmode = features->cts_swmode; core->audio_use_mclk = features->audio_use_mclk; diff --git a/drivers/gpu/drm/omapdrm/dss/hdmi5.c b/drivers/gpu/drm/omapdrm/dss/hdmi5.c index d28da9ac3e90..ae1a001d1b83 100644 --- a/drivers/gpu/drm/omapdrm/dss/hdmi5.c +++ b/drivers/gpu/drm/omapdrm/dss/hdmi5.c @@ -671,7 +671,7 @@ static int hdmi_audio_config(struct device *dev, struct omap_dss_audio *dss_audio) { struct omap_hdmi *hd = dev_get_drvdata(dev); - int ret; + int ret = 0; mutex_lock(&hd->lock); diff --git a/drivers/gpu/drm/omapdrm/omap_connector.c b/drivers/gpu/drm/omapdrm/omap_connector.c index a0d7b1d905e8..5cde26ac937b 100644 --- a/drivers/gpu/drm/omapdrm/omap_connector.c +++ b/drivers/gpu/drm/omapdrm/omap_connector.c @@ -121,6 +121,9 @@ static int omap_connector_get_modes(struct drm_connector *connector) if (dssdrv->read_edid) { void *edid = kzalloc(MAX_EDID, GFP_KERNEL); + if (!edid) + return 0; + if ((dssdrv->read_edid(dssdev, edid, MAX_EDID) > 0) && drm_edid_is_valid(edid)) { drm_mode_connector_update_edid_property( @@ -139,6 +142,9 @@ static int omap_connector_get_modes(struct drm_connector *connector) struct drm_display_mode *mode = drm_mode_create(dev); struct videomode vm = {0}; + if (!mode) + return 0; + dssdrv->get_timings(dssdev, &vm); drm_display_mode_from_videomode(&vm, mode); @@ -200,6 +206,10 @@ static int omap_connector_mode_valid(struct drm_connector *connector, if (!r) { /* check if vrefresh is still valid */ new_mode = drm_mode_duplicate(dev, mode); + + if (!new_mode) + return MODE_BAD; + new_mode->clock = vm.pixelclock / 1000; new_mode->vrefresh = 0; if (mode->vrefresh == drm_mode_vrefresh(new_mode)) diff --git a/drivers/gpu/drm/omapdrm/omap_dmm_tiler.c b/drivers/gpu/drm/omapdrm/omap_dmm_tiler.c index f9fa1c90b35c..401c02e9e6b2 100644 --- a/drivers/gpu/drm/omapdrm/omap_dmm_tiler.c +++ b/drivers/gpu/drm/omapdrm/omap_dmm_tiler.c @@ -401,12 +401,16 @@ int tiler_unpin(struct tiler_block *block) struct tiler_block *tiler_reserve_2d(enum tiler_fmt fmt, u16 w, u16 h, u16 align) { - struct tiler_block *block = kzalloc(sizeof(*block), GFP_KERNEL); + struct tiler_block *block; u32 min_align = 128; int ret; unsigned long flags; u32 slot_bytes; + block = kzalloc(sizeof(*block), GFP_KERNEL); + if (!block) + return ERR_PTR(-ENOMEM); + BUG_ON(!validfmt(fmt)); /* convert width/height to slots */ diff --git a/drivers/gpu/drm/omapdrm/tcm-sita.c b/drivers/gpu/drm/omapdrm/tcm-sita.c index d7f7bc9f061a..817be3c41863 100644 --- a/drivers/gpu/drm/omapdrm/tcm-sita.c +++ b/drivers/gpu/drm/omapdrm/tcm-sita.c @@ -90,7 +90,7 @@ static int l2r_t2b(u16 w, u16 h, u16 a, s16 offset, { int i; unsigned long index; - bool area_free; + bool area_free = false; unsigned long slots_per_band = PAGE_SIZE / slot_bytes; unsigned long bit_offset = (offset > 0) ? offset / slot_bytes : 0; unsigned long curr_bit = bit_offset; diff --git a/drivers/gpu/drm/scheduler/gpu_scheduler.c b/drivers/gpu/drm/scheduler/gpu_scheduler.c index a364fc0b38c3..df1578d6f42e 100644 --- a/drivers/gpu/drm/scheduler/gpu_scheduler.c +++ b/drivers/gpu/drm/scheduler/gpu_scheduler.c @@ -139,7 +139,6 @@ int drm_sched_entity_init(struct drm_gpu_scheduler *sched, entity->last_scheduled = NULL; spin_lock_init(&entity->rq_lock); - spin_lock_init(&entity->queue_lock); spsc_queue_init(&entity->job_queue); atomic_set(&entity->fence_seq, 0); @@ -413,6 +412,10 @@ drm_sched_entity_pop_job(struct drm_sched_entity *entity) * * @sched_job The pointer to job required to submit * + * Note: To guarantee that the order of insertion to queue matches + * the job's fence sequence number this function should be + * called with drm_sched_job_init under common lock. + * * Returns 0 for success, negative error code otherwise. */ void drm_sched_entity_push_job(struct drm_sched_job *sched_job, @@ -423,11 +426,8 @@ void drm_sched_entity_push_job(struct drm_sched_job *sched_job, trace_drm_sched_job(sched_job, entity); - spin_lock(&entity->queue_lock); first = spsc_queue_push(&entity->job_queue, &sched_job->queue_node); - spin_unlock(&entity->queue_lock); - /* first job wakes up scheduler */ if (first) { /* Add the entity to the run queue */ @@ -593,7 +593,12 @@ void drm_sched_job_recovery(struct drm_gpu_scheduler *sched) } EXPORT_SYMBOL(drm_sched_job_recovery); -/* init a sched_job with basic field */ +/** + * Init a sched_job with basic field + * + * Note: Refer to drm_sched_entity_push_job documentation + * for locking considerations. + */ int drm_sched_job_init(struct drm_sched_job *job, struct drm_gpu_scheduler *sched, struct drm_sched_entity *entity, diff --git a/drivers/gpu/drm/scheduler/sched_fence.c b/drivers/gpu/drm/scheduler/sched_fence.c index 69aab086b913..df4461648e3f 100644 --- a/drivers/gpu/drm/scheduler/sched_fence.c +++ b/drivers/gpu/drm/scheduler/sched_fence.c @@ -87,7 +87,7 @@ static bool drm_sched_fence_enable_signaling(struct dma_fence *f) } /** - * amd_sched_fence_free - free up the fence memory + * drm_sched_fence_free - free up the fence memory * * @rcu: RCU callback head * @@ -98,12 +98,11 @@ static void drm_sched_fence_free(struct rcu_head *rcu) struct dma_fence *f = container_of(rcu, struct dma_fence, rcu); struct drm_sched_fence *fence = to_drm_sched_fence(f); - dma_fence_put(fence->parent); kmem_cache_free(sched_fence_slab, fence); } /** - * amd_sched_fence_release_scheduled - callback that fence can be freed + * drm_sched_fence_release_scheduled - callback that fence can be freed * * @fence: fence * @@ -114,11 +113,12 @@ static void drm_sched_fence_release_scheduled(struct dma_fence *f) { struct drm_sched_fence *fence = to_drm_sched_fence(f); + dma_fence_put(fence->parent); call_rcu(&fence->finished.rcu, drm_sched_fence_free); } /** - * amd_sched_fence_release_finished - drop extra reference + * drm_sched_fence_release_finished - drop extra reference * * @f: fence * diff --git a/drivers/gpu/drm/tegra/dc.c b/drivers/gpu/drm/tegra/dc.c index 9f83a65b5ea9..c3afe7b2237e 100644 --- a/drivers/gpu/drm/tegra/dc.c +++ b/drivers/gpu/drm/tegra/dc.c @@ -163,28 +163,89 @@ static void tegra_plane_setup_blending_legacy(struct tegra_plane *plane) BLEND_COLOR_KEY_NONE; u32 blendnokey = BLEND_WEIGHT1(255) | BLEND_WEIGHT0(255); struct tegra_plane_state *state; + u32 blending[2]; unsigned int i; + /* disable blending for non-overlapping case */ + tegra_plane_writel(plane, blendnokey, DC_WIN_BLEND_NOKEY); + tegra_plane_writel(plane, foreground, DC_WIN_BLEND_1WIN); + state = to_tegra_plane_state(plane->base.state); - /* alpha contribution is 1 minus sum of overlapping windows */ - for (i = 0; i < 3; i++) { - if (state->dependent[i]) - background[i] |= BLEND_CONTROL_DEPENDENT; - } + if (state->opaque) { + /* + * Since custom fix-weight blending isn't utilized and weight + * of top window is set to max, we can enforce dependent + * blending which in this case results in transparent bottom + * window if top window is opaque and if top window enables + * alpha blending, then bottom window is getting alpha value + * of 1 minus the sum of alpha components of the overlapping + * plane. + */ + background[0] |= BLEND_CONTROL_DEPENDENT; + background[1] |= BLEND_CONTROL_DEPENDENT; - /* enable alpha blending if pixel format has an alpha component */ - if (!state->opaque) + /* + * The region where three windows overlap is the intersection + * of the two regions where two windows overlap. It contributes + * to the area if all of the windows on top of it have an alpha + * component. + */ + switch (state->base.normalized_zpos) { + case 0: + if (state->blending[0].alpha && + state->blending[1].alpha) + background[2] |= BLEND_CONTROL_DEPENDENT; + break; + + case 1: + background[2] |= BLEND_CONTROL_DEPENDENT; + break; + } + } else { + /* + * Enable alpha blending if pixel format has an alpha + * component. + */ foreground |= BLEND_CONTROL_ALPHA; - /* - * Disable blending and assume Window A is the bottom-most window, - * Window C is the top-most window and Window B is in the middle. - */ - tegra_plane_writel(plane, blendnokey, DC_WIN_BLEND_NOKEY); - tegra_plane_writel(plane, foreground, DC_WIN_BLEND_1WIN); + /* + * If any of the windows on top of this window is opaque, it + * will completely conceal this window within that area. If + * top window has an alpha component, it is blended over the + * bottom window. + */ + for (i = 0; i < 2; i++) { + if (state->blending[i].alpha && + state->blending[i].top) + background[i] |= BLEND_CONTROL_DEPENDENT; + } + + switch (state->base.normalized_zpos) { + case 0: + if (state->blending[0].alpha && + state->blending[1].alpha) + background[2] |= BLEND_CONTROL_DEPENDENT; + break; - switch (plane->index) { + case 1: + /* + * When both middle and topmost windows have an alpha, + * these windows a mixed together and then the result + * is blended over the bottom window. + */ + if (state->blending[0].alpha && + state->blending[0].top) + background[2] |= BLEND_CONTROL_ALPHA; + + if (state->blending[1].alpha && + state->blending[1].top) + background[2] |= BLEND_CONTROL_ALPHA; + break; + } + } + + switch (state->base.normalized_zpos) { case 0: tegra_plane_writel(plane, background[0], DC_WIN_BLEND_2WIN_X); tegra_plane_writel(plane, background[1], DC_WIN_BLEND_2WIN_Y); @@ -192,8 +253,21 @@ static void tegra_plane_setup_blending_legacy(struct tegra_plane *plane) break; case 1: - tegra_plane_writel(plane, foreground, DC_WIN_BLEND_2WIN_X); - tegra_plane_writel(plane, background[1], DC_WIN_BLEND_2WIN_Y); + /* + * If window B / C is topmost, then X / Y registers are + * matching the order of blending[...] state indices, + * otherwise a swap is required. + */ + if (!state->blending[0].top && state->blending[1].top) { + blending[0] = foreground; + blending[1] = background[1]; + } else { + blending[0] = background[0]; + blending[1] = foreground; + } + + tegra_plane_writel(plane, blending[0], DC_WIN_BLEND_2WIN_X); + tegra_plane_writel(plane, blending[1], DC_WIN_BLEND_2WIN_Y); tegra_plane_writel(plane, background[2], DC_WIN_BLEND_3WIN_XY); break; @@ -224,6 +298,39 @@ static void tegra_plane_setup_blending(struct tegra_plane *plane, tegra_plane_writel(plane, value, DC_WIN_BLEND_LAYER_CONTROL); } +static bool +tegra_plane_use_horizontal_filtering(struct tegra_plane *plane, + const struct tegra_dc_window *window) +{ + struct tegra_dc *dc = plane->dc; + + if (window->src.w == window->dst.w) + return false; + + if (plane->index == 0 && dc->soc->has_win_a_without_filters) + return false; + + return true; +} + +static bool +tegra_plane_use_vertical_filtering(struct tegra_plane *plane, + const struct tegra_dc_window *window) +{ + struct tegra_dc *dc = plane->dc; + + if (window->src.h == window->dst.h) + return false; + + if (plane->index == 0 && dc->soc->has_win_a_without_filters) + return false; + + if (plane->index == 2 && dc->soc->has_win_c_without_vert_filter) + return false; + + return true; +} + static void tegra_dc_setup_window(struct tegra_plane *plane, const struct tegra_dc_window *window) { @@ -361,12 +468,50 @@ static void tegra_dc_setup_window(struct tegra_plane *plane, if (window->bottom_up) value |= V_DIRECTION; + if (tegra_plane_use_horizontal_filtering(plane, window)) { + /* + * Enable horizontal 6-tap filter and set filtering + * coefficients to the default values defined in TRM. + */ + tegra_plane_writel(plane, 0x00008000, DC_WIN_H_FILTER_P(0)); + tegra_plane_writel(plane, 0x3e087ce1, DC_WIN_H_FILTER_P(1)); + tegra_plane_writel(plane, 0x3b117ac1, DC_WIN_H_FILTER_P(2)); + tegra_plane_writel(plane, 0x591b73aa, DC_WIN_H_FILTER_P(3)); + tegra_plane_writel(plane, 0x57256d9a, DC_WIN_H_FILTER_P(4)); + tegra_plane_writel(plane, 0x552f668b, DC_WIN_H_FILTER_P(5)); + tegra_plane_writel(plane, 0x73385e8b, DC_WIN_H_FILTER_P(6)); + tegra_plane_writel(plane, 0x72435583, DC_WIN_H_FILTER_P(7)); + tegra_plane_writel(plane, 0x714c4c8b, DC_WIN_H_FILTER_P(8)); + tegra_plane_writel(plane, 0x70554393, DC_WIN_H_FILTER_P(9)); + tegra_plane_writel(plane, 0x715e389b, DC_WIN_H_FILTER_P(10)); + tegra_plane_writel(plane, 0x71662faa, DC_WIN_H_FILTER_P(11)); + tegra_plane_writel(plane, 0x536d25ba, DC_WIN_H_FILTER_P(12)); + tegra_plane_writel(plane, 0x55731bca, DC_WIN_H_FILTER_P(13)); + tegra_plane_writel(plane, 0x387a11d9, DC_WIN_H_FILTER_P(14)); + tegra_plane_writel(plane, 0x3c7c08f1, DC_WIN_H_FILTER_P(15)); + + value |= H_FILTER; + } + + if (tegra_plane_use_vertical_filtering(plane, window)) { + unsigned int i, k; + + /* + * Enable vertical 2-tap filter and set filtering + * coefficients to the default values defined in TRM. + */ + for (i = 0, k = 128; i < 16; i++, k -= 8) + tegra_plane_writel(plane, k, DC_WIN_V_FILTER_P(i)); + + value |= V_FILTER; + } + tegra_plane_writel(plane, value, DC_WIN_WIN_OPTIONS); - if (dc->soc->supports_blending) - tegra_plane_setup_blending(plane, window); - else + if (dc->soc->has_legacy_blending) tegra_plane_setup_blending_legacy(plane); + else + tegra_plane_setup_blending(plane, window); } static const u32 tegra20_primary_formats[] = { @@ -451,17 +596,18 @@ static int tegra_plane_atomic_check(struct drm_plane *plane, struct drm_plane_state *state) { struct tegra_plane_state *plane_state = to_tegra_plane_state(state); + unsigned int rotation = DRM_MODE_ROTATE_0 | DRM_MODE_REFLECT_Y; struct tegra_bo_tiling *tiling = &plane_state->tiling; struct tegra_plane *tegra = to_tegra_plane(plane); struct tegra_dc *dc = to_tegra_dc(state->crtc); - unsigned int format; int err; /* no need for further checks if the plane is being disabled */ if (!state->crtc) return 0; - err = tegra_plane_format(state->fb->format->format, &format, + err = tegra_plane_format(state->fb->format->format, + &plane_state->format, &plane_state->swap); if (err < 0) return err; @@ -472,22 +618,12 @@ static int tegra_plane_atomic_check(struct drm_plane *plane, * the corresponding opaque formats. However, the opaque formats can * be emulated by disabling alpha blending for the plane. */ - if (!dc->soc->supports_blending) { - if (!tegra_plane_format_has_alpha(format)) { - err = tegra_plane_format_get_alpha(format, &format); - if (err < 0) - return err; - - plane_state->opaque = true; - } else { - plane_state->opaque = false; - } - - tegra_plane_check_dependent(tegra, plane_state); + if (dc->soc->has_legacy_blending) { + err = tegra_plane_setup_legacy_state(tegra, plane_state); + if (err < 0) + return err; } - plane_state->format = format; - err = tegra_fb_get_tiling(state->fb, tiling); if (err < 0) return err; @@ -498,6 +634,13 @@ static int tegra_plane_atomic_check(struct drm_plane *plane, return -EINVAL; } + rotation = drm_rotation_simplify(state->rotation, rotation); + + if (rotation & DRM_MODE_REFLECT_Y) + plane_state->bottom_up = true; + else + plane_state->bottom_up = false; + /* * Tegra doesn't support different strides for U and V planes so we * error out if the user tries to display a framebuffer with such a @@ -558,7 +701,7 @@ static void tegra_plane_atomic_update(struct drm_plane *plane, window.dst.w = drm_rect_width(&plane->state->dst); window.dst.h = drm_rect_height(&plane->state->dst); window.bits_per_pixel = fb->format->cpp[0] * 8; - window.bottom_up = tegra_fb_is_bottom_up(fb); + window.bottom_up = tegra_fb_is_bottom_up(fb) || state->bottom_up; /* copy from state */ window.zpos = plane->state->normalized_zpos; @@ -639,9 +782,15 @@ static struct drm_plane *tegra_primary_plane_create(struct drm_device *drm, } drm_plane_helper_add(&plane->base, &tegra_plane_helper_funcs); + drm_plane_create_zpos_property(&plane->base, plane->index, 0, 255); - if (dc->soc->supports_blending) - drm_plane_create_zpos_property(&plane->base, 0, 0, 255); + err = drm_plane_create_rotation_property(&plane->base, + DRM_MODE_ROTATE_0, + DRM_MODE_ROTATE_0 | + DRM_MODE_REFLECT_Y); + if (err < 0) + dev_err(dc->dev, "failed to create rotation property: %d\n", + err); return &plane->base; } @@ -918,9 +1067,15 @@ static struct drm_plane *tegra_dc_overlay_plane_create(struct drm_device *drm, } drm_plane_helper_add(&plane->base, &tegra_plane_helper_funcs); + drm_plane_create_zpos_property(&plane->base, plane->index, 0, 255); - if (dc->soc->supports_blending) - drm_plane_create_zpos_property(&plane->base, 0, 0, 255); + err = drm_plane_create_rotation_property(&plane->base, + DRM_MODE_ROTATE_0, + DRM_MODE_ROTATE_0 | + DRM_MODE_REFLECT_Y); + if (err < 0) + dev_err(dc->dev, "failed to create rotation property: %d\n", + err); return &plane->base; } @@ -1826,7 +1981,6 @@ static irqreturn_t tegra_dc_irq(int irq, void *data) static int tegra_dc_init(struct host1x_client *client) { struct drm_device *drm = dev_get_drvdata(client->parent); - struct iommu_group *group = iommu_group_get(client->dev); unsigned long flags = HOST1X_SYNCPT_CLIENT_MANAGED; struct tegra_dc *dc = host1x_client_to_dc(client); struct tegra_drm *tegra = drm->dev_private; @@ -1838,20 +1992,11 @@ static int tegra_dc_init(struct host1x_client *client) if (!dc->syncpt) dev_warn(dc->dev, "failed to allocate syncpoint\n"); - if (group && tegra->domain) { - if (group != tegra->group) { - err = iommu_attach_group(tegra->domain, group); - if (err < 0) { - dev_err(dc->dev, - "failed to attach to domain: %d\n", - err); - return err; - } - - tegra->group = group; - } - - dc->domain = tegra->domain; + dc->group = host1x_client_iommu_attach(client, true); + if (IS_ERR(dc->group)) { + err = PTR_ERR(dc->group); + dev_err(client->dev, "failed to attach to domain: %d\n", err); + return err; } if (dc->soc->wgrps) @@ -1916,24 +2061,15 @@ cleanup: if (!IS_ERR(primary)) drm_plane_cleanup(primary); - if (group && dc->domain) { - if (group == tegra->group) { - iommu_detach_group(dc->domain, group); - tegra->group = NULL; - } - - dc->domain = NULL; - } + host1x_client_iommu_detach(client, dc->group); + host1x_syncpt_free(dc->syncpt); return err; } static int tegra_dc_exit(struct host1x_client *client) { - struct drm_device *drm = dev_get_drvdata(client->parent); - struct iommu_group *group = iommu_group_get(client->dev); struct tegra_dc *dc = host1x_client_to_dc(client); - struct tegra_drm *tegra = drm->dev_private; int err; devm_free_irq(dc->dev, dc->irq, dc); @@ -1944,15 +2080,7 @@ static int tegra_dc_exit(struct host1x_client *client) return err; } - if (group && dc->domain) { - if (group == tegra->group) { - iommu_detach_group(dc->domain, group); - tegra->group = NULL; - } - - dc->domain = NULL; - } - + host1x_client_iommu_detach(client, dc->group); host1x_syncpt_free(dc->syncpt); return 0; @@ -1968,7 +2096,7 @@ static const struct tegra_dc_soc_info tegra20_dc_soc_info = { .supports_interlacing = false, .supports_cursor = false, .supports_block_linear = false, - .supports_blending = false, + .has_legacy_blending = true, .pitch_align = 8, .has_powergate = false, .coupled_pm = true, @@ -1978,6 +2106,8 @@ static const struct tegra_dc_soc_info tegra20_dc_soc_info = { .num_overlay_formats = ARRAY_SIZE(tegra20_overlay_formats), .overlay_formats = tegra20_overlay_formats, .modifiers = tegra20_modifiers, + .has_win_a_without_filters = true, + .has_win_c_without_vert_filter = true, }; static const struct tegra_dc_soc_info tegra30_dc_soc_info = { @@ -1985,7 +2115,7 @@ static const struct tegra_dc_soc_info tegra30_dc_soc_info = { .supports_interlacing = false, .supports_cursor = false, .supports_block_linear = false, - .supports_blending = false, + .has_legacy_blending = true, .pitch_align = 8, .has_powergate = false, .coupled_pm = false, @@ -1995,6 +2125,8 @@ static const struct tegra_dc_soc_info tegra30_dc_soc_info = { .num_overlay_formats = ARRAY_SIZE(tegra20_overlay_formats), .overlay_formats = tegra20_overlay_formats, .modifiers = tegra20_modifiers, + .has_win_a_without_filters = false, + .has_win_c_without_vert_filter = false, }; static const struct tegra_dc_soc_info tegra114_dc_soc_info = { @@ -2002,7 +2134,7 @@ static const struct tegra_dc_soc_info tegra114_dc_soc_info = { .supports_interlacing = false, .supports_cursor = false, .supports_block_linear = false, - .supports_blending = false, + .has_legacy_blending = true, .pitch_align = 64, .has_powergate = true, .coupled_pm = false, @@ -2012,6 +2144,8 @@ static const struct tegra_dc_soc_info tegra114_dc_soc_info = { .num_overlay_formats = ARRAY_SIZE(tegra114_overlay_formats), .overlay_formats = tegra114_overlay_formats, .modifiers = tegra20_modifiers, + .has_win_a_without_filters = false, + .has_win_c_without_vert_filter = false, }; static const struct tegra_dc_soc_info tegra124_dc_soc_info = { @@ -2019,7 +2153,7 @@ static const struct tegra_dc_soc_info tegra124_dc_soc_info = { .supports_interlacing = true, .supports_cursor = true, .supports_block_linear = true, - .supports_blending = true, + .has_legacy_blending = false, .pitch_align = 64, .has_powergate = true, .coupled_pm = false, @@ -2029,6 +2163,8 @@ static const struct tegra_dc_soc_info tegra124_dc_soc_info = { .num_overlay_formats = ARRAY_SIZE(tegra124_overlay_formats), .overlay_formats = tegra124_overlay_formats, .modifiers = tegra124_modifiers, + .has_win_a_without_filters = false, + .has_win_c_without_vert_filter = false, }; static const struct tegra_dc_soc_info tegra210_dc_soc_info = { @@ -2036,7 +2172,7 @@ static const struct tegra_dc_soc_info tegra210_dc_soc_info = { .supports_interlacing = true, .supports_cursor = true, .supports_block_linear = true, - .supports_blending = true, + .has_legacy_blending = false, .pitch_align = 64, .has_powergate = true, .coupled_pm = false, @@ -2046,6 +2182,8 @@ static const struct tegra_dc_soc_info tegra210_dc_soc_info = { .num_overlay_formats = ARRAY_SIZE(tegra114_overlay_formats), .overlay_formats = tegra114_overlay_formats, .modifiers = tegra124_modifiers, + .has_win_a_without_filters = false, + .has_win_c_without_vert_filter = false, }; static const struct tegra_windowgroup_soc tegra186_dc_wgrps[] = { @@ -2087,7 +2225,7 @@ static const struct tegra_dc_soc_info tegra186_dc_soc_info = { .supports_interlacing = true, .supports_cursor = true, .supports_block_linear = true, - .supports_blending = true, + .has_legacy_blending = false, .pitch_align = 64, .has_powergate = false, .coupled_pm = false, diff --git a/drivers/gpu/drm/tegra/dc.h b/drivers/gpu/drm/tegra/dc.h index d2b50d32de4d..e96f582ca692 100644 --- a/drivers/gpu/drm/tegra/dc.h +++ b/drivers/gpu/drm/tegra/dc.h @@ -55,7 +55,7 @@ struct tegra_dc_soc_info { bool supports_interlacing; bool supports_cursor; bool supports_block_linear; - bool supports_blending; + bool has_legacy_blending; unsigned int pitch_align; bool has_powergate; bool coupled_pm; @@ -67,6 +67,8 @@ struct tegra_dc_soc_info { const u32 *overlay_formats; unsigned int num_overlay_formats; const u64 *modifiers; + bool has_win_a_without_filters; + bool has_win_c_without_vert_filter; }; struct tegra_dc { @@ -92,7 +94,7 @@ struct tegra_dc { const struct tegra_dc_soc_info *soc; - struct iommu_domain *domain; + struct iommu_group *group; }; static inline struct tegra_dc * @@ -553,6 +555,9 @@ int tegra_dc_rgb_exit(struct tegra_dc *dc); #define THREAD_NUM(x) (((x) & 0x1f) << 1) #define THREAD_GROUP_ENABLE (1 << 0) +#define DC_WIN_H_FILTER_P(p) (0x601 + (p)) +#define DC_WIN_V_FILTER_P(p) (0x619 + (p)) + #define DC_WIN_CSC_YOF 0x611 #define DC_WIN_CSC_KYRGB 0x612 #define DC_WIN_CSC_KUR 0x613 @@ -566,6 +571,8 @@ int tegra_dc_rgb_exit(struct tegra_dc *dc); #define H_DIRECTION (1 << 0) #define V_DIRECTION (1 << 2) #define COLOR_EXPAND (1 << 6) +#define H_FILTER (1 << 8) +#define V_FILTER (1 << 10) #define CSC_ENABLE (1 << 18) #define WIN_ENABLE (1 << 30) diff --git a/drivers/gpu/drm/tegra/drm.c b/drivers/gpu/drm/tegra/drm.c index a0519612ae2c..776c1513e582 100644 --- a/drivers/gpu/drm/tegra/drm.c +++ b/drivers/gpu/drm/tegra/drm.c @@ -98,6 +98,10 @@ static int tegra_drm_load(struct drm_device *drm, unsigned long flags) goto free; } + err = iova_cache_get(); + if (err < 0) + goto domain; + geometry = &tegra->domain->geometry; gem_start = geometry->aperture_start; gem_end = geometry->aperture_end - CARVEOUT_SZ; @@ -191,11 +195,14 @@ config: drm_mode_config_cleanup(drm); if (tegra->domain) { - iommu_domain_free(tegra->domain); - drm_mm_takedown(&tegra->mm); mutex_destroy(&tegra->mm_lock); + drm_mm_takedown(&tegra->mm); put_iova_domain(&tegra->carveout.domain); + iova_cache_put(); } +domain: + if (tegra->domain) + iommu_domain_free(tegra->domain); free: kfree(tegra); return err; @@ -217,10 +224,11 @@ static void tegra_drm_unload(struct drm_device *drm) return; if (tegra->domain) { - iommu_domain_free(tegra->domain); - drm_mm_takedown(&tegra->mm); mutex_destroy(&tegra->mm_lock); + drm_mm_takedown(&tegra->mm); put_iova_domain(&tegra->carveout.domain); + iova_cache_put(); + iommu_domain_free(tegra->domain); } kfree(tegra); @@ -300,46 +308,15 @@ static int host1x_reloc_copy_from_user(struct host1x_reloc *dest, return 0; } -static int host1x_waitchk_copy_from_user(struct host1x_waitchk *dest, - struct drm_tegra_waitchk __user *src, - struct drm_file *file) -{ - u32 cmdbuf; - int err; - - err = get_user(cmdbuf, &src->handle); - if (err < 0) - return err; - - err = get_user(dest->offset, &src->offset); - if (err < 0) - return err; - - err = get_user(dest->syncpt_id, &src->syncpt); - if (err < 0) - return err; - - err = get_user(dest->thresh, &src->thresh); - if (err < 0) - return err; - - dest->bo = host1x_bo_lookup(file, cmdbuf); - if (!dest->bo) - return -ENOENT; - - return 0; -} - int tegra_drm_submit(struct tegra_drm_context *context, struct drm_tegra_submit *args, struct drm_device *drm, struct drm_file *file) { + struct host1x_client *client = &context->client->base; unsigned int num_cmdbufs = args->num_cmdbufs; unsigned int num_relocs = args->num_relocs; - unsigned int num_waitchks = args->num_waitchks; struct drm_tegra_cmdbuf __user *user_cmdbufs; struct drm_tegra_reloc __user *user_relocs; - struct drm_tegra_waitchk __user *user_waitchks; struct drm_tegra_syncpt __user *user_syncpt; struct drm_tegra_syncpt syncpt; struct host1x *host1x = dev_get_drvdata(drm->dev->parent); @@ -351,7 +328,6 @@ int tegra_drm_submit(struct tegra_drm_context *context, user_cmdbufs = u64_to_user_ptr(args->cmdbufs); user_relocs = u64_to_user_ptr(args->relocs); - user_waitchks = u64_to_user_ptr(args->waitchks); user_syncpt = u64_to_user_ptr(args->syncpts); /* We don't yet support other than one syncpt_incr struct per submit */ @@ -363,21 +339,20 @@ int tegra_drm_submit(struct tegra_drm_context *context, return -EINVAL; job = host1x_job_alloc(context->channel, args->num_cmdbufs, - args->num_relocs, args->num_waitchks); + args->num_relocs); if (!job) return -ENOMEM; job->num_relocs = args->num_relocs; - job->num_waitchk = args->num_waitchks; - job->client = (u32)args->context; - job->class = context->client->base.class; + job->client = client; + job->class = client->class; job->serialize = true; /* * Track referenced BOs so that they can be unreferenced after the * submission is complete. */ - num_refs = num_cmdbufs + num_relocs * 2 + num_waitchks; + num_refs = num_cmdbufs + num_relocs * 2; refs = kmalloc_array(num_refs, sizeof(*refs), GFP_KERNEL); if (!refs) { @@ -438,13 +413,13 @@ int tegra_drm_submit(struct tegra_drm_context *context, struct host1x_reloc *reloc; struct tegra_bo *obj; - err = host1x_reloc_copy_from_user(&job->relocarray[num_relocs], + err = host1x_reloc_copy_from_user(&job->relocs[num_relocs], &user_relocs[num_relocs], drm, file); if (err < 0) goto fail; - reloc = &job->relocarray[num_relocs]; + reloc = &job->relocs[num_relocs]; obj = host1x_to_tegra_bo(reloc->cmdbuf.bo); refs[num_refs++] = &obj->gem; @@ -468,30 +443,6 @@ int tegra_drm_submit(struct tegra_drm_context *context, } } - /* copy and resolve waitchks from submit */ - while (num_waitchks--) { - struct host1x_waitchk *wait = &job->waitchk[num_waitchks]; - struct tegra_bo *obj; - - err = host1x_waitchk_copy_from_user( - wait, &user_waitchks[num_waitchks], file); - if (err < 0) - goto fail; - - obj = host1x_to_tegra_bo(wait->bo); - refs[num_refs++] = &obj->gem; - - /* - * The unaligned offset will cause an unaligned write during - * of the waitchks patching, corrupting the commands stream. - */ - if (wait->offset & 3 || - wait->offset >= obj->gem.size) { - err = -EINVAL; - goto fail; - } - } - if (copy_from_user(&syncpt, user_syncpt, sizeof(syncpt))) { err = -EFAULT; goto fail; @@ -1101,6 +1052,52 @@ int tegra_drm_unregister_client(struct tegra_drm *tegra, return 0; } +struct iommu_group *host1x_client_iommu_attach(struct host1x_client *client, + bool shared) +{ + struct drm_device *drm = dev_get_drvdata(client->parent); + struct tegra_drm *tegra = drm->dev_private; + struct iommu_group *group = NULL; + int err; + + if (tegra->domain) { + group = iommu_group_get(client->dev); + if (!group) { + dev_err(client->dev, "failed to get IOMMU group\n"); + return ERR_PTR(-ENODEV); + } + + if (!shared || (shared && (group != tegra->group))) { + err = iommu_attach_group(tegra->domain, group); + if (err < 0) { + iommu_group_put(group); + return ERR_PTR(err); + } + + if (shared && !tegra->group) + tegra->group = group; + } + } + + return group; +} + +void host1x_client_iommu_detach(struct host1x_client *client, + struct iommu_group *group) +{ + struct drm_device *drm = dev_get_drvdata(client->parent); + struct tegra_drm *tegra = drm->dev_private; + + if (group) { + if (group == tegra->group) { + iommu_detach_group(tegra->domain, group); + tegra->group = NULL; + } + + iommu_group_put(group); + } +} + void *tegra_drm_alloc(struct tegra_drm *tegra, size_t size, dma_addr_t *dma) { struct iova *alloc; diff --git a/drivers/gpu/drm/tegra/drm.h b/drivers/gpu/drm/tegra/drm.h index 4f41aaec8530..92d248784396 100644 --- a/drivers/gpu/drm/tegra/drm.h +++ b/drivers/gpu/drm/tegra/drm.h @@ -29,16 +29,10 @@ struct reset_control; -struct tegra_fb { - struct drm_framebuffer base; - struct tegra_bo **planes; - unsigned int num_planes; -}; - #ifdef CONFIG_DRM_FBDEV_EMULATION struct tegra_fbdev { struct drm_fb_helper base; - struct tegra_fb *fb; + struct drm_framebuffer *fb; }; #endif @@ -97,6 +91,7 @@ struct tegra_drm_client { struct host1x_client base; struct list_head list; + unsigned int version; const struct tegra_drm_client_ops *ops; }; @@ -110,6 +105,10 @@ int tegra_drm_register_client(struct tegra_drm *tegra, struct tegra_drm_client *client); int tegra_drm_unregister_client(struct tegra_drm *tegra, struct tegra_drm_client *client); +struct iommu_group *host1x_client_iommu_attach(struct host1x_client *client, + bool shared); +void host1x_client_iommu_detach(struct host1x_client *client, + struct iommu_group *group); int tegra_drm_init(struct tegra_drm *tegra, struct drm_device *drm); int tegra_drm_exit(struct tegra_drm *tegra); diff --git a/drivers/gpu/drm/tegra/fb.c b/drivers/gpu/drm/tegra/fb.c index e69434909a42..4c22cdded3c2 100644 --- a/drivers/gpu/drm/tegra/fb.c +++ b/drivers/gpu/drm/tegra/fb.c @@ -14,11 +14,7 @@ #include "drm.h" #include "gem.h" - -static inline struct tegra_fb *to_tegra_fb(struct drm_framebuffer *fb) -{ - return container_of(fb, struct tegra_fb, base); -} +#include <drm/drm_gem_framebuffer_helper.h> #ifdef CONFIG_DRM_FBDEV_EMULATION static inline struct tegra_fbdev *to_tegra_fbdev(struct drm_fb_helper *helper) @@ -30,19 +26,14 @@ static inline struct tegra_fbdev *to_tegra_fbdev(struct drm_fb_helper *helper) struct tegra_bo *tegra_fb_get_plane(struct drm_framebuffer *framebuffer, unsigned int index) { - struct tegra_fb *fb = to_tegra_fb(framebuffer); - - if (index >= framebuffer->format->num_planes) - return NULL; - - return fb->planes[index]; + return to_tegra_bo(drm_gem_fb_get_obj(framebuffer, index)); } bool tegra_fb_is_bottom_up(struct drm_framebuffer *framebuffer) { - struct tegra_fb *fb = to_tegra_fb(framebuffer); + struct tegra_bo *bo = tegra_fb_get_plane(framebuffer, 0); - if (fb->planes[0]->flags & TEGRA_BO_BOTTOM_UP) + if (bo->flags & TEGRA_BO_BOTTOM_UP) return true; return false; @@ -51,8 +42,7 @@ bool tegra_fb_is_bottom_up(struct drm_framebuffer *framebuffer) int tegra_fb_get_tiling(struct drm_framebuffer *framebuffer, struct tegra_bo_tiling *tiling) { - struct tegra_fb *fb = to_tegra_fb(framebuffer); - uint64_t modifier = fb->base.modifier; + uint64_t modifier = framebuffer->modifier; switch (modifier) { case DRM_FORMAT_MOD_LINEAR: @@ -102,46 +92,17 @@ int tegra_fb_get_tiling(struct drm_framebuffer *framebuffer, return 0; } -static void tegra_fb_destroy(struct drm_framebuffer *framebuffer) -{ - struct tegra_fb *fb = to_tegra_fb(framebuffer); - unsigned int i; - - for (i = 0; i < fb->num_planes; i++) { - struct tegra_bo *bo = fb->planes[i]; - - if (bo) { - if (bo->pages) - vunmap(bo->vaddr); - - drm_gem_object_put_unlocked(&bo->gem); - } - } - - drm_framebuffer_cleanup(framebuffer); - kfree(fb->planes); - kfree(fb); -} - -static int tegra_fb_create_handle(struct drm_framebuffer *framebuffer, - struct drm_file *file, unsigned int *handle) -{ - struct tegra_fb *fb = to_tegra_fb(framebuffer); - - return drm_gem_handle_create(file, &fb->planes[0]->gem, handle); -} - static const struct drm_framebuffer_funcs tegra_fb_funcs = { - .destroy = tegra_fb_destroy, - .create_handle = tegra_fb_create_handle, + .destroy = drm_gem_fb_destroy, + .create_handle = drm_gem_fb_create_handle, }; -static struct tegra_fb *tegra_fb_alloc(struct drm_device *drm, - const struct drm_mode_fb_cmd2 *mode_cmd, - struct tegra_bo **planes, - unsigned int num_planes) +static struct drm_framebuffer *tegra_fb_alloc(struct drm_device *drm, + const struct drm_mode_fb_cmd2 *mode_cmd, + struct tegra_bo **planes, + unsigned int num_planes) { - struct tegra_fb *fb; + struct drm_framebuffer *fb; unsigned int i; int err; @@ -149,24 +110,15 @@ static struct tegra_fb *tegra_fb_alloc(struct drm_device *drm, if (!fb) return ERR_PTR(-ENOMEM); - fb->planes = kzalloc(num_planes * sizeof(*planes), GFP_KERNEL); - if (!fb->planes) { - kfree(fb); - return ERR_PTR(-ENOMEM); - } - - fb->num_planes = num_planes; + drm_helper_mode_fill_fb_struct(drm, fb, mode_cmd); - drm_helper_mode_fill_fb_struct(drm, &fb->base, mode_cmd); + for (i = 0; i < fb->format->num_planes; i++) + fb->obj[i] = &planes[i]->gem; - for (i = 0; i < fb->num_planes; i++) - fb->planes[i] = planes[i]; - - err = drm_framebuffer_init(drm, &fb->base, &tegra_fb_funcs); + err = drm_framebuffer_init(drm, fb, &tegra_fb_funcs); if (err < 0) { dev_err(drm->dev, "failed to initialize framebuffer: %d\n", err); - kfree(fb->planes); kfree(fb); return ERR_PTR(err); } @@ -181,7 +133,7 @@ struct drm_framebuffer *tegra_fb_create(struct drm_device *drm, unsigned int hsub, vsub, i; struct tegra_bo *planes[4]; struct drm_gem_object *gem; - struct tegra_fb *fb; + struct drm_framebuffer *fb; int err; hsub = drm_format_horz_chroma_subsampling(cmd->pixel_format); @@ -217,7 +169,7 @@ struct drm_framebuffer *tegra_fb_create(struct drm_device *drm, goto unreference; } - return &fb->base; + return fb; unreference: while (i--) @@ -298,7 +250,7 @@ static int tegra_fbdev_probe(struct drm_fb_helper *helper, return PTR_ERR(fbdev->fb); } - fb = &fbdev->fb->base; + fb = fbdev->fb; helper->fb = fb; helper->fbdev = info; @@ -398,8 +350,17 @@ static void tegra_fbdev_exit(struct tegra_fbdev *fbdev) { drm_fb_helper_unregister_fbi(&fbdev->base); - if (fbdev->fb) - drm_framebuffer_remove(&fbdev->fb->base); + if (fbdev->fb) { + struct tegra_bo *bo = tegra_fb_get_plane(fbdev->fb, 0); + + /* Undo the special mapping we made in fbdev probe. */ + if (bo && bo->pages) { + vunmap(bo->vaddr); + bo->vaddr = 0; + } + + drm_framebuffer_remove(fbdev->fb); + } drm_fb_helper_fini(&fbdev->base); tegra_fbdev_free(fbdev); diff --git a/drivers/gpu/drm/tegra/gem.c b/drivers/gpu/drm/tegra/gem.c index 8b0b4ff64bb4..00a5c9f32254 100644 --- a/drivers/gpu/drm/tegra/gem.c +++ b/drivers/gpu/drm/tegra/gem.c @@ -422,14 +422,13 @@ int tegra_bo_dumb_create(struct drm_file *file, struct drm_device *drm, return 0; } -static int tegra_bo_fault(struct vm_fault *vmf) +static vm_fault_t tegra_bo_fault(struct vm_fault *vmf) { struct vm_area_struct *vma = vmf->vma; struct drm_gem_object *gem = vma->vm_private_data; struct tegra_bo *bo = to_tegra_bo(gem); struct page *page; pgoff_t offset; - int err; if (!bo->pages) return VM_FAULT_SIGBUS; @@ -437,20 +436,7 @@ static int tegra_bo_fault(struct vm_fault *vmf) offset = (vmf->address - vma->vm_start) >> PAGE_SHIFT; page = bo->pages[offset]; - err = vm_insert_page(vma, vmf->address, page); - switch (err) { - case -EAGAIN: - case 0: - case -ERESTARTSYS: - case -EINTR: - case -EBUSY: - return VM_FAULT_NOPAGE; - - case -ENOMEM: - return VM_FAULT_OOM; - } - - return VM_FAULT_SIGBUS; + return vmf_insert_page(vma, vmf->address, page); } const struct vm_operations_struct tegra_bo_vm_ops = { @@ -663,6 +649,8 @@ struct dma_buf *tegra_gem_prime_export(struct drm_device *drm, { DEFINE_DMA_BUF_EXPORT_INFO(exp_info); + exp_info.exp_name = KBUILD_MODNAME; + exp_info.owner = drm->driver->fops->owner; exp_info.ops = &tegra_gem_prime_dmabuf_ops; exp_info.size = gem->size; exp_info.flags = flags; diff --git a/drivers/gpu/drm/tegra/gr2d.c b/drivers/gpu/drm/tegra/gr2d.c index 9a8ea93016a9..673059fd2fcb 100644 --- a/drivers/gpu/drm/tegra/gr2d.c +++ b/drivers/gpu/drm/tegra/gr2d.c @@ -7,16 +7,25 @@ */ #include <linux/clk.h> +#include <linux/iommu.h> +#include <linux/of_device.h> #include "drm.h" #include "gem.h" #include "gr2d.h" +struct gr2d_soc { + unsigned int version; +}; + struct gr2d { + struct iommu_group *group; struct tegra_drm_client client; struct host1x_channel *channel; struct clk *clk; + const struct gr2d_soc *soc; + DECLARE_BITMAP(addr_regs, GR2D_NUM_REGS); }; @@ -31,6 +40,7 @@ static int gr2d_init(struct host1x_client *client) struct drm_device *dev = dev_get_drvdata(client->parent); unsigned long flags = HOST1X_SYNCPT_HAS_BASE; struct gr2d *gr2d = to_gr2d(drm); + int err; gr2d->channel = host1x_channel_request(client->dev); if (!gr2d->channel) @@ -38,24 +48,48 @@ static int gr2d_init(struct host1x_client *client) client->syncpts[0] = host1x_syncpt_request(client, flags); if (!client->syncpts[0]) { - host1x_channel_put(gr2d->channel); - return -ENOMEM; + err = -ENOMEM; + dev_err(client->dev, "failed to request syncpoint: %d\n", err); + goto put; + } + + gr2d->group = host1x_client_iommu_attach(client, false); + if (IS_ERR(gr2d->group)) { + err = PTR_ERR(gr2d->group); + dev_err(client->dev, "failed to attach to domain: %d\n", err); + goto free; + } + + err = tegra_drm_register_client(dev->dev_private, drm); + if (err < 0) { + dev_err(client->dev, "failed to register client: %d\n", err); + goto detach; } - return tegra_drm_register_client(dev->dev_private, drm); + return 0; + +detach: + host1x_client_iommu_detach(client, gr2d->group); +free: + host1x_syncpt_free(client->syncpts[0]); +put: + host1x_channel_put(gr2d->channel); + return err; } static int gr2d_exit(struct host1x_client *client) { struct tegra_drm_client *drm = host1x_to_drm_client(client); struct drm_device *dev = dev_get_drvdata(client->parent); + struct tegra_drm *tegra = dev->dev_private; struct gr2d *gr2d = to_gr2d(drm); int err; - err = tegra_drm_unregister_client(dev->dev_private, drm); + err = tegra_drm_unregister_client(tegra, drm); if (err < 0) return err; + host1x_client_iommu_detach(client, gr2d->group); host1x_syncpt_free(client->syncpts[0]); host1x_channel_put(gr2d->channel); @@ -123,9 +157,17 @@ static const struct tegra_drm_client_ops gr2d_ops = { .submit = tegra_drm_submit, }; +static const struct gr2d_soc tegra20_gr2d_soc = { + .version = 0x20, +}; + +static const struct gr2d_soc tegra30_gr2d_soc = { + .version = 0x30, +}; + static const struct of_device_id gr2d_match[] = { - { .compatible = "nvidia,tegra30-gr2d" }, - { .compatible = "nvidia,tegra20-gr2d" }, + { .compatible = "nvidia,tegra30-gr2d", .data = &tegra20_gr2d_soc }, + { .compatible = "nvidia,tegra20-gr2d", .data = &tegra30_gr2d_soc }, { }, }; MODULE_DEVICE_TABLE(of, gr2d_match); @@ -158,6 +200,8 @@ static int gr2d_probe(struct platform_device *pdev) if (!gr2d) return -ENOMEM; + gr2d->soc = of_device_get_match_data(dev); + syncpts = devm_kzalloc(dev, sizeof(*syncpts), GFP_KERNEL); if (!syncpts) return -ENOMEM; @@ -182,6 +226,7 @@ static int gr2d_probe(struct platform_device *pdev) gr2d->client.base.num_syncpts = 1; INIT_LIST_HEAD(&gr2d->client.list); + gr2d->client.version = gr2d->soc->version; gr2d->client.ops = &gr2d_ops; err = host1x_client_register(&gr2d->client.base); diff --git a/drivers/gpu/drm/tegra/gr3d.c b/drivers/gpu/drm/tegra/gr3d.c index 28c4ef63065b..4778ae999668 100644 --- a/drivers/gpu/drm/tegra/gr3d.c +++ b/drivers/gpu/drm/tegra/gr3d.c @@ -9,7 +9,9 @@ #include <linux/clk.h> #include <linux/host1x.h> +#include <linux/iommu.h> #include <linux/module.h> +#include <linux/of_device.h> #include <linux/platform_device.h> #include <linux/reset.h> @@ -19,7 +21,12 @@ #include "gem.h" #include "gr3d.h" +struct gr3d_soc { + unsigned int version; +}; + struct gr3d { + struct iommu_group *group; struct tegra_drm_client client; struct host1x_channel *channel; struct clk *clk_secondary; @@ -27,6 +34,8 @@ struct gr3d { struct reset_control *rst_secondary; struct reset_control *rst; + const struct gr3d_soc *soc; + DECLARE_BITMAP(addr_regs, GR3D_NUM_REGS); }; @@ -41,6 +50,7 @@ static int gr3d_init(struct host1x_client *client) struct drm_device *dev = dev_get_drvdata(client->parent); unsigned long flags = HOST1X_SYNCPT_HAS_BASE; struct gr3d *gr3d = to_gr3d(drm); + int err; gr3d->channel = host1x_channel_request(client->dev); if (!gr3d->channel) @@ -48,11 +58,33 @@ static int gr3d_init(struct host1x_client *client) client->syncpts[0] = host1x_syncpt_request(client, flags); if (!client->syncpts[0]) { - host1x_channel_put(gr3d->channel); - return -ENOMEM; + err = -ENOMEM; + dev_err(client->dev, "failed to request syncpoint: %d\n", err); + goto put; } - return tegra_drm_register_client(dev->dev_private, drm); + gr3d->group = host1x_client_iommu_attach(client, false); + if (IS_ERR(gr3d->group)) { + err = PTR_ERR(gr3d->group); + dev_err(client->dev, "failed to attach to domain: %d\n", err); + goto free; + } + + err = tegra_drm_register_client(dev->dev_private, drm); + if (err < 0) { + dev_err(client->dev, "failed to register client: %d\n", err); + goto detach; + } + + return 0; + +detach: + host1x_client_iommu_detach(client, gr3d->group); +free: + host1x_syncpt_free(client->syncpts[0]); +put: + host1x_channel_put(gr3d->channel); + return err; } static int gr3d_exit(struct host1x_client *client) @@ -66,6 +98,7 @@ static int gr3d_exit(struct host1x_client *client) if (err < 0) return err; + host1x_client_iommu_detach(client, gr3d->group); host1x_syncpt_free(client->syncpts[0]); host1x_channel_put(gr3d->channel); @@ -125,10 +158,22 @@ static const struct tegra_drm_client_ops gr3d_ops = { .submit = tegra_drm_submit, }; +static const struct gr3d_soc tegra20_gr3d_soc = { + .version = 0x20, +}; + +static const struct gr3d_soc tegra30_gr3d_soc = { + .version = 0x30, +}; + +static const struct gr3d_soc tegra114_gr3d_soc = { + .version = 0x35, +}; + static const struct of_device_id tegra_gr3d_match[] = { - { .compatible = "nvidia,tegra114-gr3d" }, - { .compatible = "nvidia,tegra30-gr3d" }, - { .compatible = "nvidia,tegra20-gr3d" }, + { .compatible = "nvidia,tegra114-gr3d", .data = &tegra114_gr3d_soc }, + { .compatible = "nvidia,tegra30-gr3d", .data = &tegra30_gr3d_soc }, + { .compatible = "nvidia,tegra20-gr3d", .data = &tegra20_gr3d_soc }, { } }; MODULE_DEVICE_TABLE(of, tegra_gr3d_match); @@ -250,6 +295,8 @@ static int gr3d_probe(struct platform_device *pdev) if (!gr3d) return -ENOMEM; + gr3d->soc = of_device_get_match_data(&pdev->dev); + syncpts = devm_kzalloc(&pdev->dev, sizeof(*syncpts), GFP_KERNEL); if (!syncpts) return -ENOMEM; @@ -307,6 +354,7 @@ static int gr3d_probe(struct platform_device *pdev) gr3d->client.base.num_syncpts = 1; INIT_LIST_HEAD(&gr3d->client.list); + gr3d->client.version = gr3d->soc->version; gr3d->client.ops = &gr3d_ops; err = host1x_client_register(&gr3d->client.base); diff --git a/drivers/gpu/drm/tegra/hub.c b/drivers/gpu/drm/tegra/hub.c index 9a3f23d4780f..8f4fcbb515fb 100644 --- a/drivers/gpu/drm/tegra/hub.c +++ b/drivers/gpu/drm/tegra/hub.c @@ -687,7 +687,7 @@ void tegra_display_hub_atomic_commit(struct drm_device *drm, struct device *dev = hub->client.dev; int err; - hub_state = tegra_display_hub_get_state(hub, state); + hub_state = to_tegra_display_hub_state(hub->base.state); if (hub_state->clk) { err = clk_set_rate(hub_state->clk, hub_state->rate); diff --git a/drivers/gpu/drm/tegra/plane.c b/drivers/gpu/drm/tegra/plane.c index 176ef46c615c..d068e8aa3553 100644 --- a/drivers/gpu/drm/tegra/plane.c +++ b/drivers/gpu/drm/tegra/plane.c @@ -23,6 +23,7 @@ static void tegra_plane_destroy(struct drm_plane *plane) static void tegra_plane_reset(struct drm_plane *plane) { + struct tegra_plane *p = to_tegra_plane(plane); struct tegra_plane_state *state; if (plane->state) @@ -35,6 +36,8 @@ static void tegra_plane_reset(struct drm_plane *plane) if (state) { plane->state = &state->base; plane->state->plane = plane; + plane->state->zpos = p->index; + plane->state->normalized_zpos = p->index; } } @@ -53,10 +56,11 @@ tegra_plane_atomic_duplicate_state(struct drm_plane *plane) copy->tiling = state->tiling; copy->format = state->format; copy->swap = state->swap; + copy->bottom_up = state->bottom_up; copy->opaque = state->opaque; - for (i = 0; i < 3; i++) - copy->dependent[i] = state->dependent[i]; + for (i = 0; i < 2; i++) + copy->blending[i] = state->blending[i]; return ©->base; } @@ -267,24 +271,8 @@ static bool __drm_format_has_alpha(u32 format) return false; } -/* - * This is applicable to Tegra20 and Tegra30 only where the opaque formats can - * be emulated using the alpha formats and alpha blending disabled. - */ -bool tegra_plane_format_has_alpha(unsigned int format) -{ - switch (format) { - case WIN_COLOR_DEPTH_B5G5R5A1: - case WIN_COLOR_DEPTH_A1B5G5R5: - case WIN_COLOR_DEPTH_R8G8B8A8: - case WIN_COLOR_DEPTH_B8G8R8A8: - return true; - } - - return false; -} - -int tegra_plane_format_get_alpha(unsigned int opaque, unsigned int *alpha) +static int tegra_plane_format_get_alpha(unsigned int opaque, + unsigned int *alpha) { if (tegra_plane_format_is_yuv(opaque, NULL)) { *alpha = opaque; @@ -316,6 +304,67 @@ int tegra_plane_format_get_alpha(unsigned int opaque, unsigned int *alpha) return -EINVAL; } +/* + * This is applicable to Tegra20 and Tegra30 only where the opaque formats can + * be emulated using the alpha formats and alpha blending disabled. + */ +static int tegra_plane_setup_opacity(struct tegra_plane *tegra, + struct tegra_plane_state *state) +{ + unsigned int format; + int err; + + switch (state->format) { + case WIN_COLOR_DEPTH_B5G5R5A1: + case WIN_COLOR_DEPTH_A1B5G5R5: + case WIN_COLOR_DEPTH_R8G8B8A8: + case WIN_COLOR_DEPTH_B8G8R8A8: + state->opaque = false; + break; + + default: + err = tegra_plane_format_get_alpha(state->format, &format); + if (err < 0) + return err; + + state->format = format; + state->opaque = true; + break; + } + + return 0; +} + +static int tegra_plane_check_transparency(struct tegra_plane *tegra, + struct tegra_plane_state *state) +{ + struct drm_plane_state *old, *plane_state; + struct drm_plane *plane; + + old = drm_atomic_get_old_plane_state(state->base.state, &tegra->base); + + /* check if zpos / transparency changed */ + if (old->normalized_zpos == state->base.normalized_zpos && + to_tegra_plane_state(old)->opaque == state->opaque) + return 0; + + /* include all sibling planes into this commit */ + drm_for_each_plane(plane, tegra->base.dev) { + struct tegra_plane *p = to_tegra_plane(plane); + + /* skip this plane and planes on different CRTCs */ + if (p == tegra || p->dc != tegra->dc) + continue; + + plane_state = drm_atomic_get_plane_state(state->base.state, + plane); + if (IS_ERR(plane_state)) + return PTR_ERR(plane_state); + } + + return 1; +} + static unsigned int tegra_plane_get_overlap_index(struct tegra_plane *plane, struct tegra_plane *other) { @@ -336,61 +385,98 @@ static unsigned int tegra_plane_get_overlap_index(struct tegra_plane *plane, return index; } -void tegra_plane_check_dependent(struct tegra_plane *tegra, - struct tegra_plane_state *state) +static void tegra_plane_update_transparency(struct tegra_plane *tegra, + struct tegra_plane_state *state) { - struct drm_plane_state *old, *new; + struct drm_plane_state *new; struct drm_plane *plane; - unsigned int zpos[2]; unsigned int i; - for (i = 0; i < 2; i++) - zpos[i] = 0; - - for_each_oldnew_plane_in_state(state->base.state, plane, old, new, i) { + for_each_new_plane_in_state(state->base.state, plane, new, i) { struct tegra_plane *p = to_tegra_plane(plane); unsigned index; /* skip this plane and planes on different CRTCs */ - if (p == tegra || new->crtc != state->base.crtc) + if (p == tegra || p->dc != tegra->dc) continue; index = tegra_plane_get_overlap_index(tegra, p); - state->dependent[index] = false; + if (new->fb && __drm_format_has_alpha(new->fb->format->format)) + state->blending[index].alpha = true; + else + state->blending[index].alpha = false; + + if (new->normalized_zpos > state->base.normalized_zpos) + state->blending[index].top = true; + else + state->blending[index].top = false; /* - * If any of the other planes is on top of this plane and uses - * a format with an alpha component, mark this plane as being - * dependent, meaning it's alpha value will be 1 minus the sum - * of alpha components of the overlapping planes. + * Missing framebuffer means that plane is disabled, in this + * case mark B / C window as top to be able to differentiate + * windows indices order in regards to zPos for the middle + * window X / Y registers programming. */ - if (p->index > tegra->index) { - if (__drm_format_has_alpha(new->fb->format->format)) - state->dependent[index] = true; - - /* keep track of the Z position */ - zpos[index] = p->index; - } + if (!new->fb) + state->blending[index].top = (index == 1); } +} + +static int tegra_plane_setup_transparency(struct tegra_plane *tegra, + struct tegra_plane_state *state) +{ + struct tegra_plane_state *tegra_state; + struct drm_plane_state *new; + struct drm_plane *plane; + int err; /* - * The region where three windows overlap is the intersection of the - * two regions where two windows overlap. It contributes to the area - * if any of the windows on top of it have an alpha component. + * If planes zpos / transparency changed, sibling planes blending + * state may require adjustment and in this case they will be included + * into this atom commit, otherwise blending state is unchanged. */ - for (i = 0; i < 2; i++) - state->dependent[2] = state->dependent[2] || - state->dependent[i]; + err = tegra_plane_check_transparency(tegra, state); + if (err <= 0) + return err; /* - * However, if any of the windows on top of this window is opaque, it - * will completely conceal this window within that area, so avoid the - * window from contributing to the area. + * All planes are now in the atomic state, walk them up and update + * transparency state for each plane. */ - for (i = 0; i < 2; i++) { - if (zpos[i] > tegra->index) - state->dependent[2] = state->dependent[2] && - state->dependent[i]; + drm_for_each_plane(plane, tegra->base.dev) { + struct tegra_plane *p = to_tegra_plane(plane); + + /* skip planes on different CRTCs */ + if (p->dc != tegra->dc) + continue; + + new = drm_atomic_get_new_plane_state(state->base.state, plane); + tegra_state = to_tegra_plane_state(new); + + /* + * There is no need to update blending state for the disabled + * plane. + */ + if (new->fb) + tegra_plane_update_transparency(p, tegra_state); } + + return 0; +} + +int tegra_plane_setup_legacy_state(struct tegra_plane *tegra, + struct tegra_plane_state *state) +{ + int err; + + err = tegra_plane_setup_opacity(tegra, state); + if (err < 0) + return err; + + err = tegra_plane_setup_transparency(tegra, state); + if (err < 0) + return err; + + return 0; } diff --git a/drivers/gpu/drm/tegra/plane.h b/drivers/gpu/drm/tegra/plane.h index 6938719e7e5d..e79e6b4a8e0a 100644 --- a/drivers/gpu/drm/tegra/plane.h +++ b/drivers/gpu/drm/tegra/plane.h @@ -34,6 +34,11 @@ static inline struct tegra_plane *to_tegra_plane(struct drm_plane *plane) return container_of(plane, struct tegra_plane, base); } +struct tegra_plane_legacy_blending_state { + bool alpha; + bool top; +}; + struct tegra_plane_state { struct drm_plane_state base; @@ -41,9 +46,11 @@ struct tegra_plane_state { u32 format; u32 swap; + bool bottom_up; + /* used for legacy blending support only */ + struct tegra_plane_legacy_blending_state blending[2]; bool opaque; - bool dependent[3]; }; static inline struct tegra_plane_state * @@ -62,9 +69,7 @@ int tegra_plane_state_add(struct tegra_plane *plane, int tegra_plane_format(u32 fourcc, u32 *format, u32 *swap); bool tegra_plane_format_is_yuv(unsigned int format, bool *planar); -bool tegra_plane_format_has_alpha(unsigned int format); -int tegra_plane_format_get_alpha(unsigned int opaque, unsigned int *alpha); -void tegra_plane_check_dependent(struct tegra_plane *tegra, - struct tegra_plane_state *state); +int tegra_plane_setup_legacy_state(struct tegra_plane *tegra, + struct tegra_plane_state *state); #endif /* TEGRA_PLANE_H */ diff --git a/drivers/gpu/drm/tegra/vic.c b/drivers/gpu/drm/tegra/vic.c index f5794dd49f3b..9f657a63b0bb 100644 --- a/drivers/gpu/drm/tegra/vic.c +++ b/drivers/gpu/drm/tegra/vic.c @@ -25,6 +25,7 @@ struct vic_config { const char *firmware; + unsigned int version; }; struct vic { @@ -264,18 +265,21 @@ static const struct tegra_drm_client_ops vic_ops = { static const struct vic_config vic_t124_config = { .firmware = NVIDIA_TEGRA_124_VIC_FIRMWARE, + .version = 0x40, }; #define NVIDIA_TEGRA_210_VIC_FIRMWARE "nvidia/tegra210/vic04_ucode.bin" static const struct vic_config vic_t210_config = { .firmware = NVIDIA_TEGRA_210_VIC_FIRMWARE, + .version = 0x21, }; #define NVIDIA_TEGRA_186_VIC_FIRMWARE "nvidia/tegra186/vic04_ucode.bin" static const struct vic_config vic_t186_config = { .firmware = NVIDIA_TEGRA_186_VIC_FIRMWARE, + .version = 0x18, }; static const struct of_device_id vic_match[] = { @@ -342,6 +346,7 @@ static int vic_probe(struct platform_device *pdev) vic->dev = dev; INIT_LIST_HEAD(&vic->client.list); + vic->client.version = vic->config->version; vic->client.ops = &vic_ops; err = host1x_client_register(&vic->client.base); diff --git a/drivers/gpu/drm/v3d/Kconfig b/drivers/gpu/drm/v3d/Kconfig index a0c0259355bd..1552bf552c94 100644 --- a/drivers/gpu/drm/v3d/Kconfig +++ b/drivers/gpu/drm/v3d/Kconfig @@ -3,6 +3,7 @@ config DRM_V3D depends on ARCH_BCM || ARCH_BCMSTB || COMPILE_TEST depends on DRM depends on COMMON_CLK + depends on MMU select DRM_SCHED help Choose this option if you have a system that has a Broadcom diff --git a/drivers/gpu/drm/vc4/vc4_dpi.c b/drivers/gpu/drm/vc4/vc4_dpi.c index 72c9dbd81d7f..f185812970da 100644 --- a/drivers/gpu/drm/vc4/vc4_dpi.c +++ b/drivers/gpu/drm/vc4/vc4_dpi.c @@ -96,7 +96,6 @@ struct vc4_dpi { struct platform_device *pdev; struct drm_encoder *encoder; - struct drm_connector *connector; void __iomem *regs; @@ -164,14 +163,31 @@ static void vc4_dpi_encoder_disable(struct drm_encoder *encoder) static void vc4_dpi_encoder_enable(struct drm_encoder *encoder) { + struct drm_device *dev = encoder->dev; struct drm_display_mode *mode = &encoder->crtc->mode; struct vc4_dpi_encoder *vc4_encoder = to_vc4_dpi_encoder(encoder); struct vc4_dpi *dpi = vc4_encoder->dpi; + struct drm_connector_list_iter conn_iter; + struct drm_connector *connector = NULL, *connector_scan; u32 dpi_c = DPI_ENABLE | DPI_OUTPUT_ENABLE_MODE; int ret; - if (dpi->connector->display_info.num_bus_formats) { - u32 bus_format = dpi->connector->display_info.bus_formats[0]; + /* Look up the connector attached to DPI so we can get the + * bus_format. Ideally the bridge would tell us the + * bus_format we want, but it doesn't yet, so assume that it's + * uniform throughout the bridge chain. + */ + drm_connector_list_iter_begin(dev, &conn_iter); + drm_for_each_connector_iter(connector_scan, &conn_iter) { + if (connector_scan->encoder == encoder) { + connector = connector_scan; + break; + } + } + drm_connector_list_iter_end(&conn_iter); + + if (connector && connector->display_info.num_bus_formats) { + u32 bus_format = connector->display_info.bus_formats[0]; switch (bus_format) { case MEDIA_BUS_FMT_RGB888_1X24: @@ -199,6 +215,9 @@ static void vc4_dpi_encoder_enable(struct drm_encoder *encoder) DRM_ERROR("Unknown media bus format %d\n", bus_format); break; } + } else { + /* Default to 24bit if no connector found. */ + dpi_c |= VC4_SET_FIELD(DPI_FORMAT_24BIT_888_RGB, DPI_FORMAT); } if (mode->flags & DRM_MODE_FLAG_NHSYNC) diff --git a/drivers/gpu/drm/vc4/vc4_drv.c b/drivers/gpu/drm/vc4/vc4_drv.c index d9b8b701d2ce..466d0a27b415 100644 --- a/drivers/gpu/drm/vc4/vc4_drv.c +++ b/drivers/gpu/drm/vc4/vc4_drv.c @@ -130,6 +130,7 @@ static void vc4_close(struct drm_device *dev, struct drm_file *file) struct vc4_file *vc4file = file->driver_priv; vc4_perfmon_close_file(vc4file); + kfree(vc4file); } static const struct vm_operations_struct vc4_vm_ops = { diff --git a/drivers/gpu/drm/vc4/vc4_plane.c b/drivers/gpu/drm/vc4/vc4_plane.c index 3483c05cc3d6..71d44c357d35 100644 --- a/drivers/gpu/drm/vc4/vc4_plane.c +++ b/drivers/gpu/drm/vc4/vc4_plane.c @@ -505,7 +505,7 @@ static int vc4_plane_mode_set(struct drm_plane *plane, * the scl fields here. */ if (num_planes == 1) { - scl0 = vc4_get_scl_field(state, 1); + scl0 = vc4_get_scl_field(state, 0); scl1 = scl0; } else { scl0 = vc4_get_scl_field(state, 1); diff --git a/drivers/gpu/host1x/cdma.c b/drivers/gpu/host1x/cdma.c index 28541b280739..91df51e631b2 100644 --- a/drivers/gpu/host1x/cdma.c +++ b/drivers/gpu/host1x/cdma.c @@ -51,7 +51,7 @@ static void host1x_pushbuffer_destroy(struct push_buffer *pb) struct host1x_cdma *cdma = pb_to_cdma(pb); struct host1x *host1x = cdma_to_host1x(cdma); - if (!pb->phys) + if (!pb->mapped) return; if (host1x->domain) { @@ -127,7 +127,7 @@ static int host1x_pushbuffer_init(struct push_buffer *pb) iommu_free_iova: __free_iova(&host1x->iova, alloc); iommu_free_mem: - dma_free_wc(host1x->dev, pb->alloc_size, pb->mapped, pb->phys); + dma_free_wc(host1x->dev, size, pb->mapped, pb->phys); return err; } @@ -247,7 +247,7 @@ static void cdma_start_timer_locked(struct host1x_cdma *cdma, static void stop_cdma_timer_locked(struct host1x_cdma *cdma) { cancel_delayed_work(&cdma->timeout.wq); - cdma->timeout.client = 0; + cdma->timeout.client = NULL; } /* diff --git a/drivers/gpu/host1x/cdma.h b/drivers/gpu/host1x/cdma.h index 286d49386be9..e97e17b82370 100644 --- a/drivers/gpu/host1x/cdma.h +++ b/drivers/gpu/host1x/cdma.h @@ -44,7 +44,7 @@ struct host1x_job; struct push_buffer { void *mapped; /* mapped pushbuffer memory */ dma_addr_t dma; /* device address of pushbuffer */ - phys_addr_t phys; /* physical address of pushbuffer */ + dma_addr_t phys; /* physical address of pushbuffer */ u32 fence; /* index we've written */ u32 pos; /* index to write to */ u32 size; @@ -58,7 +58,7 @@ struct buffer_timeout { u32 syncpt_val; /* syncpt value when completed */ ktime_t start_ktime; /* starting time */ /* context timeout information */ - int client; + struct host1x_client *client; }; enum cdma_event { diff --git a/drivers/gpu/host1x/debug.c b/drivers/gpu/host1x/debug.c index dc77ec452ffc..329e4a3d8ae7 100644 --- a/drivers/gpu/host1x/debug.c +++ b/drivers/gpu/host1x/debug.c @@ -103,7 +103,7 @@ static void show_syncpts(struct host1x *m, struct output *o) static void show_all(struct host1x *m, struct output *o, bool show_fifo) { - int i; + unsigned int i; host1x_hw_show_mlocks(m, o); show_syncpts(m, o); diff --git a/drivers/gpu/host1x/dev.c b/drivers/gpu/host1x/dev.c index 03db71173f5d..f1d5f76e9c33 100644 --- a/drivers/gpu/host1x/dev.c +++ b/drivers/gpu/host1x/dev.c @@ -223,10 +223,14 @@ static int host1x_probe(struct platform_device *pdev) struct iommu_domain_geometry *geometry; unsigned long order; + err = iova_cache_get(); + if (err < 0) + goto put_group; + host->domain = iommu_domain_alloc(&platform_bus_type); if (!host->domain) { err = -ENOMEM; - goto put_group; + goto put_cache; } err = iommu_attach_group(host->domain, host->group); @@ -234,6 +238,7 @@ static int host1x_probe(struct platform_device *pdev) if (err == -ENODEV) { iommu_domain_free(host->domain); host->domain = NULL; + iova_cache_put(); iommu_group_put(host->group); host->group = NULL; goto skip_iommu; @@ -308,6 +313,9 @@ fail_detach_device: fail_free_domain: if (host->domain) iommu_domain_free(host->domain); +put_cache: + if (host->group) + iova_cache_put(); put_group: iommu_group_put(host->group); @@ -328,6 +336,7 @@ static int host1x_remove(struct platform_device *pdev) put_iova_domain(&host->iova); iommu_detach_group(host->domain, host->group); iommu_domain_free(host->domain); + iova_cache_put(); iommu_group_put(host->group); } diff --git a/drivers/gpu/host1x/dev.h b/drivers/gpu/host1x/dev.h index 43e9fabb43a1..36f44ffebe73 100644 --- a/drivers/gpu/host1x/dev.h +++ b/drivers/gpu/host1x/dev.h @@ -78,7 +78,6 @@ struct host1x_syncpt_ops { void (*load_wait_base)(struct host1x_syncpt *syncpt); u32 (*load)(struct host1x_syncpt *syncpt); int (*cpu_incr)(struct host1x_syncpt *syncpt); - int (*patch_wait)(struct host1x_syncpt *syncpt, void *patch_addr); void (*assign_to_channel)(struct host1x_syncpt *syncpt, struct host1x_channel *channel); void (*enable_protection)(struct host1x *host); @@ -183,13 +182,6 @@ static inline int host1x_hw_syncpt_cpu_incr(struct host1x *host, return host->syncpt_op->cpu_incr(sp); } -static inline int host1x_hw_syncpt_patch_wait(struct host1x *host, - struct host1x_syncpt *sp, - void *patch_addr) -{ - return host->syncpt_op->patch_wait(sp, patch_addr); -} - static inline void host1x_hw_syncpt_assign_to_channel( struct host1x *host, struct host1x_syncpt *sp, struct host1x_channel *ch) diff --git a/drivers/gpu/host1x/hw/channel_hw.c b/drivers/gpu/host1x/hw/channel_hw.c index 9af758785a11..d188f9068b91 100644 --- a/drivers/gpu/host1x/hw/channel_hw.c +++ b/drivers/gpu/host1x/hw/channel_hw.c @@ -104,8 +104,7 @@ static int channel_submit(struct host1x_job *job) sp = host->syncpt + job->syncpt_id; trace_host1x_channel_submit(dev_name(ch->dev), job->num_gathers, job->num_relocs, - job->num_waitchk, job->syncpt_id, - job->syncpt_incrs); + job->syncpt_id, job->syncpt_incrs); /* before error checks, return current max */ prev_max = job->syncpt_end = host1x_syncpt_read_max(sp); @@ -165,7 +164,7 @@ static int channel_submit(struct host1x_job *job) trace_host1x_channel_submitted(dev_name(ch->dev), prev_max, syncval); /* schedule a submit complete interrupt */ - err = host1x_intr_add_action(host, job->syncpt_id, syncval, + err = host1x_intr_add_action(host, sp, syncval, HOST1X_INTR_ACTION_SUBMIT_COMPLETE, ch, completed_waiter, NULL); completed_waiter = NULL; diff --git a/drivers/gpu/host1x/hw/syncpt_hw.c b/drivers/gpu/host1x/hw/syncpt_hw.c index 7dfd47d74f89..a23bb3352d02 100644 --- a/drivers/gpu/host1x/hw/syncpt_hw.c +++ b/drivers/gpu/host1x/hw/syncpt_hw.c @@ -96,16 +96,6 @@ static int syncpt_cpu_incr(struct host1x_syncpt *sp) return 0; } -/* remove a wait pointed to by patch_addr */ -static int syncpt_patch_wait(struct host1x_syncpt *sp, void *patch_addr) -{ - u32 override = host1x_class_host_wait_syncpt(HOST1X_SYNCPT_RESERVED, 0); - - *((u32 *)patch_addr) = override; - - return 0; -} - /** * syncpt_assign_to_channel() - Assign syncpoint to channel * @sp: syncpoint @@ -156,7 +146,6 @@ static const struct host1x_syncpt_ops host1x_syncpt_ops = { .load_wait_base = syncpt_read_wait_base, .load = syncpt_load, .cpu_incr = syncpt_cpu_incr, - .patch_wait = syncpt_patch_wait, .assign_to_channel = syncpt_assign_to_channel, .enable_protection = syncpt_enable_protection, }; diff --git a/drivers/gpu/host1x/intr.c b/drivers/gpu/host1x/intr.c index 8b4fad0ab35d..9629c009d10f 100644 --- a/drivers/gpu/host1x/intr.c +++ b/drivers/gpu/host1x/intr.c @@ -144,7 +144,7 @@ static const action_handler action_handlers[HOST1X_INTR_ACTION_COUNT] = { static void run_handlers(struct list_head completed[HOST1X_INTR_ACTION_COUNT]) { struct list_head *head = completed; - int i; + unsigned int i; for (i = 0; i < HOST1X_INTR_ACTION_COUNT; ++i, ++head) { action_handler handler = action_handlers[i]; @@ -211,11 +211,11 @@ static void syncpt_thresh_work(struct work_struct *work) host1x_syncpt_load(host->syncpt + id)); } -int host1x_intr_add_action(struct host1x *host, unsigned int id, u32 thresh, - enum host1x_intr_action action, void *data, - struct host1x_waitlist *waiter, void **ref) +int host1x_intr_add_action(struct host1x *host, struct host1x_syncpt *syncpt, + u32 thresh, enum host1x_intr_action action, + void *data, struct host1x_waitlist *waiter, + void **ref) { - struct host1x_syncpt *syncpt; int queue_was_empty; if (waiter == NULL) { @@ -234,19 +234,17 @@ int host1x_intr_add_action(struct host1x *host, unsigned int id, u32 thresh, waiter->data = data; waiter->count = 1; - syncpt = host->syncpt + id; - spin_lock(&syncpt->intr.lock); queue_was_empty = list_empty(&syncpt->intr.wait_head); if (add_waiter_to_queue(waiter, &syncpt->intr.wait_head)) { /* added at head of list - new threshold value */ - host1x_hw_intr_set_syncpt_threshold(host, id, thresh); + host1x_hw_intr_set_syncpt_threshold(host, syncpt->id, thresh); /* added as first waiter - enable interrupt */ if (queue_was_empty) - host1x_hw_intr_enable_syncpt_intr(host, id); + host1x_hw_intr_enable_syncpt_intr(host, syncpt->id); } spin_unlock(&syncpt->intr.lock); diff --git a/drivers/gpu/host1x/intr.h b/drivers/gpu/host1x/intr.h index 1370c2bb75b8..6db96af484fe 100644 --- a/drivers/gpu/host1x/intr.h +++ b/drivers/gpu/host1x/intr.h @@ -22,6 +22,7 @@ #include <linux/interrupt.h> #include <linux/workqueue.h> +struct host1x_syncpt; struct host1x; enum host1x_intr_action { @@ -75,9 +76,10 @@ struct host1x_waitlist { * * This is a non-blocking api. */ -int host1x_intr_add_action(struct host1x *host, unsigned int id, u32 thresh, - enum host1x_intr_action action, void *data, - struct host1x_waitlist *waiter, void **ref); +int host1x_intr_add_action(struct host1x *host, struct host1x_syncpt *syncpt, + u32 thresh, enum host1x_intr_action action, + void *data, struct host1x_waitlist *waiter, + void **ref); /* * Unreference an action submitted to host1x_intr_add_action(). diff --git a/drivers/gpu/host1x/job.c b/drivers/gpu/host1x/job.c index db509ab8874e..e2f4a4d93d20 100644 --- a/drivers/gpu/host1x/job.c +++ b/drivers/gpu/host1x/job.c @@ -34,8 +34,7 @@ #define HOST1X_WAIT_SYNCPT_OFFSET 0x8 struct host1x_job *host1x_job_alloc(struct host1x_channel *ch, - u32 num_cmdbufs, u32 num_relocs, - u32 num_waitchks) + u32 num_cmdbufs, u32 num_relocs) { struct host1x_job *job = NULL; unsigned int num_unpins = num_cmdbufs + num_relocs; @@ -46,7 +45,6 @@ struct host1x_job *host1x_job_alloc(struct host1x_channel *ch, total = sizeof(struct host1x_job) + (u64)num_relocs * sizeof(struct host1x_reloc) + (u64)num_unpins * sizeof(struct host1x_job_unpin_data) + - (u64)num_waitchks * sizeof(struct host1x_waitchk) + (u64)num_cmdbufs * sizeof(struct host1x_job_gather) + (u64)num_unpins * sizeof(dma_addr_t) + (u64)num_unpins * sizeof(u32 *); @@ -62,12 +60,10 @@ struct host1x_job *host1x_job_alloc(struct host1x_channel *ch, /* Redistribute memory to the structs */ mem += sizeof(struct host1x_job); - job->relocarray = num_relocs ? mem : NULL; + job->relocs = num_relocs ? mem : NULL; mem += num_relocs * sizeof(struct host1x_reloc); job->unpins = num_unpins ? mem : NULL; mem += num_unpins * sizeof(struct host1x_job_unpin_data); - job->waitchk = num_waitchks ? mem : NULL; - mem += num_waitchks * sizeof(struct host1x_waitchk); job->gathers = num_cmdbufs ? mem : NULL; mem += num_cmdbufs * sizeof(struct host1x_job_gather); job->addr_phys = num_unpins ? mem : NULL; @@ -100,84 +96,18 @@ void host1x_job_put(struct host1x_job *job) EXPORT_SYMBOL(host1x_job_put); void host1x_job_add_gather(struct host1x_job *job, struct host1x_bo *bo, - u32 words, u32 offset) + unsigned int words, unsigned int offset) { - struct host1x_job_gather *cur_gather = &job->gathers[job->num_gathers]; + struct host1x_job_gather *gather = &job->gathers[job->num_gathers]; + + gather->words = words; + gather->bo = bo; + gather->offset = offset; - cur_gather->words = words; - cur_gather->bo = bo; - cur_gather->offset = offset; job->num_gathers++; } EXPORT_SYMBOL(host1x_job_add_gather); -/* - * NULL an already satisfied WAIT_SYNCPT host method, by patching its - * args in the command stream. The method data is changed to reference - * a reserved (never given out or incr) HOST1X_SYNCPT_RESERVED syncpt - * with a matching threshold value of 0, so is guaranteed to be popped - * by the host HW. - */ -static void host1x_syncpt_patch_offset(struct host1x_syncpt *sp, - struct host1x_bo *h, u32 offset) -{ - void *patch_addr = NULL; - - /* patch the wait */ - patch_addr = host1x_bo_kmap(h, offset >> PAGE_SHIFT); - if (patch_addr) { - host1x_syncpt_patch_wait(sp, - patch_addr + (offset & ~PAGE_MASK)); - host1x_bo_kunmap(h, offset >> PAGE_SHIFT, patch_addr); - } else - pr_err("Could not map cmdbuf for wait check\n"); -} - -/* - * Check driver supplied waitchk structs for syncpt thresholds - * that have already been satisfied and NULL the comparison (to - * avoid a wrap condition in the HW). - */ -static int do_waitchks(struct host1x_job *job, struct host1x *host, - struct host1x_job_gather *g) -{ - struct host1x_bo *patch = g->bo; - int i; - - /* compare syncpt vs wait threshold */ - for (i = 0; i < job->num_waitchk; i++) { - struct host1x_waitchk *wait = &job->waitchk[i]; - struct host1x_syncpt *sp = - host1x_syncpt_get(host, wait->syncpt_id); - - /* validate syncpt id */ - if (wait->syncpt_id > host1x_syncpt_nb_pts(host)) - continue; - - /* skip all other gathers */ - if (patch != wait->bo) - continue; - - trace_host1x_syncpt_wait_check(wait->bo, wait->offset, - wait->syncpt_id, wait->thresh, - host1x_syncpt_read_min(sp)); - - if (host1x_syncpt_is_expired(sp, wait->thresh)) { - dev_dbg(host->dev, - "drop WAIT id %u (%s) thresh 0x%x, min 0x%x\n", - wait->syncpt_id, sp->name, wait->thresh, - host1x_syncpt_read_min(sp)); - - host1x_syncpt_patch_offset(sp, patch, - g->offset + wait->offset); - } - - wait->bo = NULL; - } - - return 0; -} - static unsigned int pin_job(struct host1x *host, struct host1x_job *job) { unsigned int i; @@ -186,7 +116,7 @@ static unsigned int pin_job(struct host1x *host, struct host1x_job *job) job->num_unpins = 0; for (i = 0; i < job->num_relocs; i++) { - struct host1x_reloc *reloc = &job->relocarray[i]; + struct host1x_reloc *reloc = &job->relocs[i]; struct sg_table *sgt; dma_addr_t phys_addr; @@ -267,14 +197,14 @@ unpin: static int do_relocs(struct host1x_job *job, struct host1x_job_gather *g) { - int i = 0; u32 last_page = ~0; void *cmdbuf_page_addr = NULL; struct host1x_bo *cmdbuf = g->bo; + unsigned int i; /* pin & patch the relocs for one gather */ for (i = 0; i < job->num_relocs; i++) { - struct host1x_reloc *reloc = &job->relocarray[i]; + struct host1x_reloc *reloc = &job->relocs[i]; u32 reloc_addr = (job->reloc_addr_phys[i] + reloc->target.offset) >> reloc->shift; u32 *target; @@ -331,17 +261,6 @@ static bool check_reloc(struct host1x_reloc *reloc, struct host1x_bo *cmdbuf, return true; } -static bool check_wait(struct host1x_waitchk *wait, struct host1x_bo *cmdbuf, - unsigned int offset) -{ - offset *= sizeof(u32); - - if (wait->bo != cmdbuf || wait->offset != offset) - return false; - - return true; -} - struct host1x_firewall { struct host1x_job *job; struct device *dev; @@ -349,9 +268,6 @@ struct host1x_firewall { unsigned int num_relocs; struct host1x_reloc *reloc; - unsigned int num_waitchks; - struct host1x_waitchk *waitchk; - struct host1x_bo *cmdbuf; unsigned int offset; @@ -378,20 +294,6 @@ static int check_register(struct host1x_firewall *fw, unsigned long offset) fw->reloc++; } - if (offset == HOST1X_WAIT_SYNCPT_OFFSET) { - if (fw->class != HOST1X_CLASS_HOST1X) - return -EINVAL; - - if (!fw->num_waitchks) - return -EINVAL; - - if (!check_wait(fw->waitchk, fw->cmdbuf, fw->offset)) - return -EINVAL; - - fw->num_waitchks--; - fw->waitchk++; - } - return 0; } @@ -550,14 +452,12 @@ static inline int copy_gathers(struct host1x_job *job, struct device *dev) struct host1x_firewall fw; size_t size = 0; size_t offset = 0; - int i; + unsigned int i; fw.job = job; fw.dev = dev; - fw.reloc = job->relocarray; + fw.reloc = job->relocs; fw.num_relocs = job->num_relocs; - fw.waitchk = job->waitchk; - fw.num_waitchks = job->num_waitchk; fw.class = job->class; for (i = 0; i < job->num_gathers; i++) { @@ -604,8 +504,8 @@ static inline int copy_gathers(struct host1x_job *job, struct device *dev) offset += g->words * sizeof(u32); } - /* No relocs and waitchks should remain at this point */ - if (fw.num_relocs || fw.num_waitchks) + /* No relocs should remain at this point */ + if (fw.num_relocs) return -EINVAL; return 0; @@ -616,19 +516,6 @@ int host1x_job_pin(struct host1x_job *job, struct device *dev) int err; unsigned int i, j; struct host1x *host = dev_get_drvdata(dev->parent); - DECLARE_BITMAP(waitchk_mask, host1x_syncpt_nb_pts(host)); - - bitmap_zero(waitchk_mask, host1x_syncpt_nb_pts(host)); - for (i = 0; i < job->num_waitchk; i++) { - u32 syncpt_id = job->waitchk[i].syncpt_id; - - if (syncpt_id < host1x_syncpt_nb_pts(host)) - set_bit(syncpt_id, waitchk_mask); - } - - /* get current syncpt values for waitchk */ - for_each_set_bit(i, waitchk_mask, host1x_syncpt_nb_pts(host)) - host1x_syncpt_load(host->syncpt + i); /* pin memory */ err = pin_job(host, job); @@ -663,10 +550,6 @@ int host1x_job_pin(struct host1x_job *job, struct device *dev) err = do_relocs(job, g); if (err) break; - - err = do_waitchks(job, host, g); - if (err) - break; } out: diff --git a/drivers/gpu/host1x/job.h b/drivers/gpu/host1x/job.h index 4bda51d503ec..188400e00192 100644 --- a/drivers/gpu/host1x/job.h +++ b/drivers/gpu/host1x/job.h @@ -20,10 +20,10 @@ #define __HOST1X_JOB_H struct host1x_job_gather { - u32 words; + unsigned int words; dma_addr_t base; struct host1x_bo *bo; - u32 offset; + unsigned int offset; bool handled; }; diff --git a/drivers/gpu/host1x/syncpt.c b/drivers/gpu/host1x/syncpt.c index a2a952adc136..a5dbf1ba4645 100644 --- a/drivers/gpu/host1x/syncpt.c +++ b/drivers/gpu/host1x/syncpt.c @@ -57,8 +57,8 @@ static struct host1x_syncpt *host1x_syncpt_alloc(struct host1x *host, struct host1x_client *client, unsigned long flags) { - int i; struct host1x_syncpt *sp = host->syncpt; + unsigned int i; char *name; mutex_lock(&host->syncpt_mutex); @@ -255,7 +255,7 @@ int host1x_syncpt_wait(struct host1x_syncpt *sp, u32 thresh, long timeout, } /* schedule a wakeup when the syncpoint value is reached */ - err = host1x_intr_add_action(sp->host, sp->id, thresh, + err = host1x_intr_add_action(sp->host, sp, thresh, HOST1X_INTR_ACTION_WAKEUP_INTERRUPTIBLE, &wq, waiter, &ref); if (err) @@ -373,12 +373,6 @@ bool host1x_syncpt_is_expired(struct host1x_syncpt *sp, u32 thresh) return (s32)(current_val - thresh) >= 0; } -/* remove a wait pointed to by patch_addr */ -int host1x_syncpt_patch_wait(struct host1x_syncpt *sp, void *patch_addr) -{ - return host1x_hw_syncpt_patch_wait(sp->host, sp, patch_addr); -} - int host1x_syncpt_init(struct host1x *host) { struct host1x_syncpt_base *bases; diff --git a/drivers/gpu/host1x/syncpt.h b/drivers/gpu/host1x/syncpt.h index 9d88d37c2397..d98e22325e9d 100644 --- a/drivers/gpu/host1x/syncpt.h +++ b/drivers/gpu/host1x/syncpt.h @@ -124,7 +124,4 @@ static inline int host1x_syncpt_is_valid(struct host1x_syncpt *sp) return sp->id < host1x_syncpt_nb_pts(sp->host); } -/* Patch a wait by replacing it with a wait for syncpt 0 value 0 */ -int host1x_syncpt_patch_wait(struct host1x_syncpt *sp, void *patch_addr); - #endif diff --git a/drivers/hid/Kconfig b/drivers/hid/Kconfig index 60252fd796f6..0000434a1fbd 100644 --- a/drivers/hid/Kconfig +++ b/drivers/hid/Kconfig @@ -462,10 +462,11 @@ config HID_LENOVO select NEW_LEDS select LEDS_CLASS ---help--- - Support for Lenovo devices that are not fully compliant with HID standard. + Support for IBM/Lenovo devices that are not fully compliant with HID standard. - Say Y if you want support for the non-compliant features of the Lenovo - Thinkpad standalone keyboards, e.g: + Say Y if you want support for horizontal scrolling of the IBM/Lenovo + Scrollpoint mice or the non-compliant features of the Lenovo Thinkpad + standalone keyboards, e.g: - ThinkPad USB Keyboard with TrackPoint (supports extra LEDs and trackpoint configuration) - ThinkPad Compact Bluetooth Keyboard with TrackPoint (supports Fn keys) diff --git a/drivers/hid/hid-ids.h b/drivers/hid/hid-ids.h index 0b5cc910f62e..46f5ecd11bf7 100644 --- a/drivers/hid/hid-ids.h +++ b/drivers/hid/hid-ids.h @@ -552,6 +552,13 @@ #define USB_VENDOR_ID_HUION 0x256c #define USB_DEVICE_ID_HUION_TABLET 0x006e +#define USB_VENDOR_ID_IBM 0x04b3 +#define USB_DEVICE_ID_IBM_SCROLLPOINT_III 0x3100 +#define USB_DEVICE_ID_IBM_SCROLLPOINT_PRO 0x3103 +#define USB_DEVICE_ID_IBM_SCROLLPOINT_OPTICAL 0x3105 +#define USB_DEVICE_ID_IBM_SCROLLPOINT_800DPI_OPTICAL 0x3108 +#define USB_DEVICE_ID_IBM_SCROLLPOINT_800DPI_OPTICAL_PRO 0x3109 + #define USB_VENDOR_ID_IDEACOM 0x1cb6 #define USB_DEVICE_ID_IDEACOM_IDC6650 0x6650 #define USB_DEVICE_ID_IDEACOM_IDC6651 0x6651 @@ -684,6 +691,7 @@ #define USB_DEVICE_ID_LENOVO_TPKBD 0x6009 #define USB_DEVICE_ID_LENOVO_CUSBKBD 0x6047 #define USB_DEVICE_ID_LENOVO_CBTKBD 0x6048 +#define USB_DEVICE_ID_LENOVO_SCROLLPOINT_OPTICAL 0x6049 #define USB_DEVICE_ID_LENOVO_TPPRODOCK 0x6067 #define USB_DEVICE_ID_LENOVO_X1_COVER 0x6085 #define USB_DEVICE_ID_LENOVO_X1_TAB 0x60a3 @@ -964,6 +972,7 @@ #define USB_DEVICE_ID_SIS817_TOUCH 0x0817 #define USB_DEVICE_ID_SIS_TS 0x1013 #define USB_DEVICE_ID_SIS1030_TOUCH 0x1030 +#define USB_DEVICE_ID_SIS10FB_TOUCH 0x10fb #define USB_VENDOR_ID_SKYCABLE 0x1223 #define USB_DEVICE_ID_SKYCABLE_WIRELESS_PRESENTER 0x3F07 diff --git a/drivers/hid/hid-lenovo.c b/drivers/hid/hid-lenovo.c index 1ac4ff4d57a6..643b6eb54442 100644 --- a/drivers/hid/hid-lenovo.c +++ b/drivers/hid/hid-lenovo.c @@ -6,6 +6,17 @@ * * Copyright (c) 2012 Bernhard Seibold * Copyright (c) 2014 Jamie Lentin <jm@lentin.co.uk> + * + * Linux IBM/Lenovo Scrollpoint mouse driver: + * - IBM Scrollpoint III + * - IBM Scrollpoint Pro + * - IBM Scrollpoint Optical + * - IBM Scrollpoint Optical 800dpi + * - IBM Scrollpoint Optical 800dpi Pro + * - Lenovo Scrollpoint Optical + * + * Copyright (c) 2012 Peter De Wachter <pdewacht@gmail.com> + * Copyright (c) 2018 Peter Ganzhorn <peter.ganzhorn@gmail.com> */ /* @@ -160,6 +171,17 @@ static int lenovo_input_mapping_cptkbd(struct hid_device *hdev, return 0; } +static int lenovo_input_mapping_scrollpoint(struct hid_device *hdev, + struct hid_input *hi, struct hid_field *field, + struct hid_usage *usage, unsigned long **bit, int *max) +{ + if (usage->hid == HID_GD_Z) { + hid_map_usage(hi, usage, bit, max, EV_REL, REL_HWHEEL); + return 1; + } + return 0; +} + static int lenovo_input_mapping(struct hid_device *hdev, struct hid_input *hi, struct hid_field *field, struct hid_usage *usage, unsigned long **bit, int *max) @@ -172,6 +194,14 @@ static int lenovo_input_mapping(struct hid_device *hdev, case USB_DEVICE_ID_LENOVO_CBTKBD: return lenovo_input_mapping_cptkbd(hdev, hi, field, usage, bit, max); + case USB_DEVICE_ID_IBM_SCROLLPOINT_III: + case USB_DEVICE_ID_IBM_SCROLLPOINT_PRO: + case USB_DEVICE_ID_IBM_SCROLLPOINT_OPTICAL: + case USB_DEVICE_ID_IBM_SCROLLPOINT_800DPI_OPTICAL: + case USB_DEVICE_ID_IBM_SCROLLPOINT_800DPI_OPTICAL_PRO: + case USB_DEVICE_ID_LENOVO_SCROLLPOINT_OPTICAL: + return lenovo_input_mapping_scrollpoint(hdev, hi, field, + usage, bit, max); default: return 0; } @@ -883,6 +913,12 @@ static const struct hid_device_id lenovo_devices[] = { { HID_USB_DEVICE(USB_VENDOR_ID_LENOVO, USB_DEVICE_ID_LENOVO_CUSBKBD) }, { HID_BLUETOOTH_DEVICE(USB_VENDOR_ID_LENOVO, USB_DEVICE_ID_LENOVO_CBTKBD) }, { HID_USB_DEVICE(USB_VENDOR_ID_LENOVO, USB_DEVICE_ID_LENOVO_TPPRODOCK) }, + { HID_USB_DEVICE(USB_VENDOR_ID_IBM, USB_DEVICE_ID_IBM_SCROLLPOINT_III) }, + { HID_USB_DEVICE(USB_VENDOR_ID_IBM, USB_DEVICE_ID_IBM_SCROLLPOINT_PRO) }, + { HID_USB_DEVICE(USB_VENDOR_ID_IBM, USB_DEVICE_ID_IBM_SCROLLPOINT_OPTICAL) }, + { HID_USB_DEVICE(USB_VENDOR_ID_IBM, USB_DEVICE_ID_IBM_SCROLLPOINT_800DPI_OPTICAL) }, + { HID_USB_DEVICE(USB_VENDOR_ID_IBM, USB_DEVICE_ID_IBM_SCROLLPOINT_800DPI_OPTICAL_PRO) }, + { HID_USB_DEVICE(USB_VENDOR_ID_LENOVO, USB_DEVICE_ID_LENOVO_SCROLLPOINT_OPTICAL) }, { } }; diff --git a/drivers/hid/i2c-hid/i2c-hid.c b/drivers/hid/i2c-hid/i2c-hid.c index 963328674e93..cc33622253aa 100644 --- a/drivers/hid/i2c-hid/i2c-hid.c +++ b/drivers/hid/i2c-hid/i2c-hid.c @@ -174,6 +174,8 @@ static const struct i2c_hid_quirks { I2C_HID_QUIRK_NO_IRQ_AFTER_RESET }, { I2C_VENDOR_ID_RAYD, I2C_PRODUCT_ID_RAYD_3118, I2C_HID_QUIRK_RESEND_REPORT_DESCR }, + { USB_VENDOR_ID_SIS_TOUCH, USB_DEVICE_ID_SIS10FB_TOUCH, + I2C_HID_QUIRK_RESEND_REPORT_DESCR }, { 0, 0 } }; diff --git a/drivers/hid/intel-ish-hid/ishtp-hid-client.c b/drivers/hid/intel-ish-hid/ishtp-hid-client.c index 157b44aacdff..acc2536c8094 100644 --- a/drivers/hid/intel-ish-hid/ishtp-hid-client.c +++ b/drivers/hid/intel-ish-hid/ishtp-hid-client.c @@ -77,21 +77,21 @@ static void process_recv(struct ishtp_cl *hid_ishtp_cl, void *recv_buf, struct ishtp_cl_data *client_data = hid_ishtp_cl->client_data; int curr_hid_dev = client_data->cur_hid_dev; - if (data_len < sizeof(struct hostif_msg_hdr)) { - dev_err(&client_data->cl_device->dev, - "[hid-ish]: error, received %u which is less than data header %u\n", - (unsigned int)data_len, - (unsigned int)sizeof(struct hostif_msg_hdr)); - ++client_data->bad_recv_cnt; - ish_hw_reset(hid_ishtp_cl->dev); - return; - } - payload = recv_buf + sizeof(struct hostif_msg_hdr); total_len = data_len; cur_pos = 0; do { + if (cur_pos + sizeof(struct hostif_msg) > total_len) { + dev_err(&client_data->cl_device->dev, + "[hid-ish]: error, received %u which is less than data header %u\n", + (unsigned int)data_len, + (unsigned int)sizeof(struct hostif_msg_hdr)); + ++client_data->bad_recv_cnt; + ish_hw_reset(hid_ishtp_cl->dev); + break; + } + recv_msg = (struct hostif_msg *)(recv_buf + cur_pos); payload_len = recv_msg->hdr.size; @@ -412,9 +412,7 @@ void hid_ishtp_get_report(struct hid_device *hid, int report_id, { struct ishtp_hid_data *hid_data = hid->driver_data; struct ishtp_cl_data *client_data = hid_data->client_data; - static unsigned char buf[10]; - unsigned int len; - struct hostif_msg_to_sensor *msg = (struct hostif_msg_to_sensor *)buf; + struct hostif_msg_to_sensor msg = {}; int rv; int i; @@ -426,14 +424,11 @@ void hid_ishtp_get_report(struct hid_device *hid, int report_id, return; } - len = sizeof(struct hostif_msg_to_sensor); - - memset(msg, 0, sizeof(struct hostif_msg_to_sensor)); - msg->hdr.command = (report_type == HID_FEATURE_REPORT) ? + msg.hdr.command = (report_type == HID_FEATURE_REPORT) ? HOSTIF_GET_FEATURE_REPORT : HOSTIF_GET_INPUT_REPORT; for (i = 0; i < client_data->num_hid_devices; ++i) { if (hid == client_data->hid_sensor_hubs[i]) { - msg->hdr.device_id = + msg.hdr.device_id = client_data->hid_devices[i].dev_id; break; } @@ -442,8 +437,9 @@ void hid_ishtp_get_report(struct hid_device *hid, int report_id, if (i == client_data->num_hid_devices) return; - msg->report_id = report_id; - rv = ishtp_cl_send(client_data->hid_ishtp_cl, buf, len); + msg.report_id = report_id; + rv = ishtp_cl_send(client_data->hid_ishtp_cl, (uint8_t *)&msg, + sizeof(msg)); if (rv) hid_ishtp_trace(client_data, "%s hid %p send failed\n", __func__, hid); diff --git a/drivers/hid/intel-ish-hid/ishtp/bus.c b/drivers/hid/intel-ish-hid/ishtp/bus.c index f272cdd9bd55..2623a567ffba 100644 --- a/drivers/hid/intel-ish-hid/ishtp/bus.c +++ b/drivers/hid/intel-ish-hid/ishtp/bus.c @@ -418,7 +418,7 @@ static struct ishtp_cl_device *ishtp_bus_add_device(struct ishtp_device *dev, list_del(&device->device_link); spin_unlock_irqrestore(&dev->device_list_lock, flags); dev_err(dev->devc, "Failed to register ISHTP client device\n"); - kfree(device); + put_device(&device->dev); return NULL; } diff --git a/drivers/hid/wacom_sys.c b/drivers/hid/wacom_sys.c index b54ef1ffcbec..ee7a37eb159a 100644 --- a/drivers/hid/wacom_sys.c +++ b/drivers/hid/wacom_sys.c @@ -1213,8 +1213,10 @@ static int __wacom_devm_sysfs_create_group(struct wacom *wacom, devres->root = root; error = sysfs_create_group(devres->root, group); - if (error) + if (error) { + devres_free(devres); return error; + } devres_add(&wacom->hdev->dev, devres); diff --git a/drivers/infiniband/Kconfig b/drivers/infiniband/Kconfig index ee270e065ba9..2a972ed6851b 100644 --- a/drivers/infiniband/Kconfig +++ b/drivers/infiniband/Kconfig @@ -61,9 +61,12 @@ config INFINIBAND_ON_DEMAND_PAGING pages on demand instead. config INFINIBAND_ADDR_TRANS - bool + bool "RDMA/CM" depends on INFINIBAND default y + ---help--- + Support for RDMA communication manager (CM). + This allows for a generic connection abstraction over RDMA. config INFINIBAND_ADDR_TRANS_CONFIGFS bool diff --git a/drivers/infiniband/core/cache.c b/drivers/infiniband/core/cache.c index e337b08de2ff..fb2d347f760f 100644 --- a/drivers/infiniband/core/cache.c +++ b/drivers/infiniband/core/cache.c @@ -291,14 +291,18 @@ static int find_gid(struct ib_gid_table *table, const union ib_gid *gid, * so lookup free slot only if requested. */ if (pempty && empty < 0) { - if (data->props & GID_TABLE_ENTRY_INVALID) { - /* Found an invalid (free) entry; allocate it */ - if (data->props & GID_TABLE_ENTRY_DEFAULT) { - if (default_gid) - empty = curr_index; - } else { - empty = curr_index; - } + if (data->props & GID_TABLE_ENTRY_INVALID && + (default_gid == + !!(data->props & GID_TABLE_ENTRY_DEFAULT))) { + /* + * Found an invalid (free) entry; allocate it. + * If default GID is requested, then our + * found slot must be one of the DEFAULT + * reserved slots or we fail. + * This ensures that only DEFAULT reserved + * slots are used for default property GIDs. + */ + empty = curr_index; } } @@ -420,8 +424,10 @@ int ib_cache_gid_add(struct ib_device *ib_dev, u8 port, return ret; } -int ib_cache_gid_del(struct ib_device *ib_dev, u8 port, - union ib_gid *gid, struct ib_gid_attr *attr) +static int +_ib_cache_gid_del(struct ib_device *ib_dev, u8 port, + union ib_gid *gid, struct ib_gid_attr *attr, + unsigned long mask, bool default_gid) { struct ib_gid_table *table; int ret = 0; @@ -431,11 +437,7 @@ int ib_cache_gid_del(struct ib_device *ib_dev, u8 port, mutex_lock(&table->lock); - ix = find_gid(table, gid, attr, false, - GID_ATTR_FIND_MASK_GID | - GID_ATTR_FIND_MASK_GID_TYPE | - GID_ATTR_FIND_MASK_NETDEV, - NULL); + ix = find_gid(table, gid, attr, default_gid, mask, NULL); if (ix < 0) { ret = -EINVAL; goto out_unlock; @@ -452,6 +454,17 @@ out_unlock: return ret; } +int ib_cache_gid_del(struct ib_device *ib_dev, u8 port, + union ib_gid *gid, struct ib_gid_attr *attr) +{ + unsigned long mask = GID_ATTR_FIND_MASK_GID | + GID_ATTR_FIND_MASK_GID_TYPE | + GID_ATTR_FIND_MASK_DEFAULT | + GID_ATTR_FIND_MASK_NETDEV; + + return _ib_cache_gid_del(ib_dev, port, gid, attr, mask, false); +} + int ib_cache_gid_del_all_netdev_gids(struct ib_device *ib_dev, u8 port, struct net_device *ndev) { @@ -728,7 +741,7 @@ void ib_cache_gid_set_default_gid(struct ib_device *ib_dev, u8 port, unsigned long gid_type_mask, enum ib_cache_gid_default_mode mode) { - union ib_gid gid; + union ib_gid gid = { }; struct ib_gid_attr gid_attr; struct ib_gid_table *table; unsigned int gid_type; @@ -736,7 +749,9 @@ void ib_cache_gid_set_default_gid(struct ib_device *ib_dev, u8 port, table = ib_dev->cache.ports[port - rdma_start_port(ib_dev)].gid; - make_default_gid(ndev, &gid); + mask = GID_ATTR_FIND_MASK_GID_TYPE | + GID_ATTR_FIND_MASK_DEFAULT | + GID_ATTR_FIND_MASK_NETDEV; memset(&gid_attr, 0, sizeof(gid_attr)); gid_attr.ndev = ndev; @@ -747,12 +762,12 @@ void ib_cache_gid_set_default_gid(struct ib_device *ib_dev, u8 port, gid_attr.gid_type = gid_type; if (mode == IB_CACHE_GID_DEFAULT_MODE_SET) { - mask = GID_ATTR_FIND_MASK_GID_TYPE | - GID_ATTR_FIND_MASK_DEFAULT; + make_default_gid(ndev, &gid); __ib_cache_gid_add(ib_dev, port, &gid, &gid_attr, mask, true); } else if (mode == IB_CACHE_GID_DEFAULT_MODE_DELETE) { - ib_cache_gid_del(ib_dev, port, &gid, &gid_attr); + _ib_cache_gid_del(ib_dev, port, &gid, + &gid_attr, mask, true); } } } diff --git a/drivers/infiniband/core/cma.c b/drivers/infiniband/core/cma.c index 51a641002e10..a693fcd4c513 100644 --- a/drivers/infiniband/core/cma.c +++ b/drivers/infiniband/core/cma.c @@ -382,6 +382,8 @@ struct cma_hdr { #define CMA_VERSION 0x00 struct cma_req_info { + struct sockaddr_storage listen_addr_storage; + struct sockaddr_storage src_addr_storage; struct ib_device *device; int port; union ib_gid local_gid; @@ -866,7 +868,6 @@ static int cma_modify_qp_rtr(struct rdma_id_private *id_priv, { struct ib_qp_attr qp_attr; int qp_attr_mask, ret; - union ib_gid sgid; mutex_lock(&id_priv->qp_mutex); if (!id_priv->id.qp) { @@ -889,12 +890,6 @@ static int cma_modify_qp_rtr(struct rdma_id_private *id_priv, if (ret) goto out; - ret = ib_query_gid(id_priv->id.device, id_priv->id.port_num, - rdma_ah_read_grh(&qp_attr.ah_attr)->sgid_index, - &sgid, NULL); - if (ret) - goto out; - BUG_ON(id_priv->cma_dev->device != id_priv->id.device); if (conn_param) @@ -1340,11 +1335,11 @@ static bool validate_net_dev(struct net_device *net_dev, } static struct net_device *cma_get_net_dev(struct ib_cm_event *ib_event, - const struct cma_req_info *req) + struct cma_req_info *req) { - struct sockaddr_storage listen_addr_storage, src_addr_storage; - struct sockaddr *listen_addr = (struct sockaddr *)&listen_addr_storage, - *src_addr = (struct sockaddr *)&src_addr_storage; + struct sockaddr *listen_addr = + (struct sockaddr *)&req->listen_addr_storage; + struct sockaddr *src_addr = (struct sockaddr *)&req->src_addr_storage; struct net_device *net_dev; const union ib_gid *gid = req->has_gid ? &req->local_gid : NULL; int err; @@ -1359,11 +1354,6 @@ static struct net_device *cma_get_net_dev(struct ib_cm_event *ib_event, if (!net_dev) return ERR_PTR(-ENODEV); - if (!validate_net_dev(net_dev, listen_addr, src_addr)) { - dev_put(net_dev); - return ERR_PTR(-EHOSTUNREACH); - } - return net_dev; } @@ -1490,15 +1480,51 @@ static struct rdma_id_private *cma_id_from_event(struct ib_cm_id *cm_id, } } + /* + * Net namespace might be getting deleted while route lookup, + * cm_id lookup is in progress. Therefore, perform netdevice + * validation, cm_id lookup under rcu lock. + * RCU lock along with netdevice state check, synchronizes with + * netdevice migrating to different net namespace and also avoids + * case where net namespace doesn't get deleted while lookup is in + * progress. + * If the device state is not IFF_UP, its properties such as ifindex + * and nd_net cannot be trusted to remain valid without rcu lock. + * net/core/dev.c change_net_namespace() ensures to synchronize with + * ongoing operations on net device after device is closed using + * synchronize_net(). + */ + rcu_read_lock(); + if (*net_dev) { + /* + * If netdevice is down, it is likely that it is administratively + * down or it might be migrating to different namespace. + * In that case avoid further processing, as the net namespace + * or ifindex may change. + */ + if (((*net_dev)->flags & IFF_UP) == 0) { + id_priv = ERR_PTR(-EHOSTUNREACH); + goto err; + } + + if (!validate_net_dev(*net_dev, + (struct sockaddr *)&req.listen_addr_storage, + (struct sockaddr *)&req.src_addr_storage)) { + id_priv = ERR_PTR(-EHOSTUNREACH); + goto err; + } + } + bind_list = cma_ps_find(*net_dev ? dev_net(*net_dev) : &init_net, rdma_ps_from_service_id(req.service_id), cma_port_from_service_id(req.service_id)); id_priv = cma_find_listener(bind_list, cm_id, ib_event, &req, *net_dev); +err: + rcu_read_unlock(); if (IS_ERR(id_priv) && *net_dev) { dev_put(*net_dev); *net_dev = NULL; } - return id_priv; } diff --git a/drivers/infiniband/core/iwpm_util.c b/drivers/infiniband/core/iwpm_util.c index 9821ae900f6d..da12da1c36f6 100644 --- a/drivers/infiniband/core/iwpm_util.c +++ b/drivers/infiniband/core/iwpm_util.c @@ -114,7 +114,7 @@ int iwpm_create_mapinfo(struct sockaddr_storage *local_sockaddr, struct sockaddr_storage *mapped_sockaddr, u8 nl_client) { - struct hlist_head *hash_bucket_head; + struct hlist_head *hash_bucket_head = NULL; struct iwpm_mapping_info *map_info; unsigned long flags; int ret = -EINVAL; @@ -142,6 +142,9 @@ int iwpm_create_mapinfo(struct sockaddr_storage *local_sockaddr, } } spin_unlock_irqrestore(&iwpm_mapinfo_lock, flags); + + if (!hash_bucket_head) + kfree(map_info); return ret; } diff --git a/drivers/infiniband/core/mad.c b/drivers/infiniband/core/mad.c index c50596f7f98a..b28452a55a08 100644 --- a/drivers/infiniband/core/mad.c +++ b/drivers/infiniband/core/mad.c @@ -59,7 +59,7 @@ module_param_named(recv_queue_size, mad_recvq_size, int, 0444); MODULE_PARM_DESC(recv_queue_size, "Size of receive queue in number of work requests"); static struct list_head ib_mad_port_list; -static u32 ib_mad_client_id = 0; +static atomic_t ib_mad_client_id = ATOMIC_INIT(0); /* Port list lock */ static DEFINE_SPINLOCK(ib_mad_port_list_lock); @@ -377,7 +377,7 @@ struct ib_mad_agent *ib_register_mad_agent(struct ib_device *device, } spin_lock_irqsave(&port_priv->reg_lock, flags); - mad_agent_priv->agent.hi_tid = ++ib_mad_client_id; + mad_agent_priv->agent.hi_tid = atomic_inc_return(&ib_mad_client_id); /* * Make sure MAD registration (if supplied) diff --git a/drivers/infiniband/core/roce_gid_mgmt.c b/drivers/infiniband/core/roce_gid_mgmt.c index cc2966380c0c..c0e4fd55e2cc 100644 --- a/drivers/infiniband/core/roce_gid_mgmt.c +++ b/drivers/infiniband/core/roce_gid_mgmt.c @@ -255,6 +255,7 @@ static void bond_delete_netdev_default_gids(struct ib_device *ib_dev, struct net_device *rdma_ndev) { struct net_device *real_dev = rdma_vlan_dev_real_dev(event_ndev); + unsigned long gid_type_mask; if (!rdma_ndev) return; @@ -264,21 +265,22 @@ static void bond_delete_netdev_default_gids(struct ib_device *ib_dev, rcu_read_lock(); - if (rdma_is_upper_dev_rcu(rdma_ndev, event_ndev) && - is_eth_active_slave_of_bonding_rcu(rdma_ndev, real_dev) == - BONDING_SLAVE_STATE_INACTIVE) { - unsigned long gid_type_mask; - + if (((rdma_ndev != event_ndev && + !rdma_is_upper_dev_rcu(rdma_ndev, event_ndev)) || + is_eth_active_slave_of_bonding_rcu(rdma_ndev, real_dev) + == + BONDING_SLAVE_STATE_INACTIVE)) { rcu_read_unlock(); + return; + } - gid_type_mask = roce_gid_type_mask_support(ib_dev, port); + rcu_read_unlock(); - ib_cache_gid_set_default_gid(ib_dev, port, rdma_ndev, - gid_type_mask, - IB_CACHE_GID_DEFAULT_MODE_DELETE); - } else { - rcu_read_unlock(); - } + gid_type_mask = roce_gid_type_mask_support(ib_dev, port); + + ib_cache_gid_set_default_gid(ib_dev, port, rdma_ndev, + gid_type_mask, + IB_CACHE_GID_DEFAULT_MODE_DELETE); } static void enum_netdev_ipv4_ips(struct ib_device *ib_dev, diff --git a/drivers/infiniband/core/ucma.c b/drivers/infiniband/core/ucma.c index 74329483af6d..eab43b17e9cf 100644 --- a/drivers/infiniband/core/ucma.c +++ b/drivers/infiniband/core/ucma.c @@ -159,6 +159,23 @@ static void ucma_put_ctx(struct ucma_context *ctx) complete(&ctx->comp); } +/* + * Same as ucm_get_ctx but requires that ->cm_id->device is valid, eg that the + * CM_ID is bound. + */ +static struct ucma_context *ucma_get_ctx_dev(struct ucma_file *file, int id) +{ + struct ucma_context *ctx = ucma_get_ctx(file, id); + + if (IS_ERR(ctx)) + return ctx; + if (!ctx->cm_id->device) { + ucma_put_ctx(ctx); + return ERR_PTR(-EINVAL); + } + return ctx; +} + static void ucma_close_event_id(struct work_struct *work) { struct ucma_event *uevent_close = container_of(work, struct ucma_event, close_work); @@ -683,7 +700,7 @@ static ssize_t ucma_resolve_ip(struct ucma_file *file, if (copy_from_user(&cmd, inbuf, sizeof(cmd))) return -EFAULT; - if (!rdma_addr_size_in6(&cmd.src_addr) || + if ((cmd.src_addr.sin6_family && !rdma_addr_size_in6(&cmd.src_addr)) || !rdma_addr_size_in6(&cmd.dst_addr)) return -EINVAL; @@ -734,7 +751,7 @@ static ssize_t ucma_resolve_route(struct ucma_file *file, if (copy_from_user(&cmd, inbuf, sizeof(cmd))) return -EFAULT; - ctx = ucma_get_ctx(file, cmd.id); + ctx = ucma_get_ctx_dev(file, cmd.id); if (IS_ERR(ctx)) return PTR_ERR(ctx); @@ -1050,7 +1067,7 @@ static ssize_t ucma_connect(struct ucma_file *file, const char __user *inbuf, if (!cmd.conn_param.valid) return -EINVAL; - ctx = ucma_get_ctx(file, cmd.id); + ctx = ucma_get_ctx_dev(file, cmd.id); if (IS_ERR(ctx)) return PTR_ERR(ctx); @@ -1092,7 +1109,7 @@ static ssize_t ucma_accept(struct ucma_file *file, const char __user *inbuf, if (copy_from_user(&cmd, inbuf, sizeof(cmd))) return -EFAULT; - ctx = ucma_get_ctx(file, cmd.id); + ctx = ucma_get_ctx_dev(file, cmd.id); if (IS_ERR(ctx)) return PTR_ERR(ctx); @@ -1120,7 +1137,7 @@ static ssize_t ucma_reject(struct ucma_file *file, const char __user *inbuf, if (copy_from_user(&cmd, inbuf, sizeof(cmd))) return -EFAULT; - ctx = ucma_get_ctx(file, cmd.id); + ctx = ucma_get_ctx_dev(file, cmd.id); if (IS_ERR(ctx)) return PTR_ERR(ctx); @@ -1139,7 +1156,7 @@ static ssize_t ucma_disconnect(struct ucma_file *file, const char __user *inbuf, if (copy_from_user(&cmd, inbuf, sizeof(cmd))) return -EFAULT; - ctx = ucma_get_ctx(file, cmd.id); + ctx = ucma_get_ctx_dev(file, cmd.id); if (IS_ERR(ctx)) return PTR_ERR(ctx); @@ -1167,15 +1184,10 @@ static ssize_t ucma_init_qp_attr(struct ucma_file *file, if (cmd.qp_state > IB_QPS_ERR) return -EINVAL; - ctx = ucma_get_ctx(file, cmd.id); + ctx = ucma_get_ctx_dev(file, cmd.id); if (IS_ERR(ctx)) return PTR_ERR(ctx); - if (!ctx->cm_id->device) { - ret = -EINVAL; - goto out; - } - resp.qp_attr_mask = 0; memset(&qp_attr, 0, sizeof qp_attr); qp_attr.qp_state = cmd.qp_state; @@ -1316,13 +1328,13 @@ static ssize_t ucma_set_option(struct ucma_file *file, const char __user *inbuf, if (copy_from_user(&cmd, inbuf, sizeof(cmd))) return -EFAULT; + if (unlikely(cmd.optlen > KMALLOC_MAX_SIZE)) + return -EINVAL; + ctx = ucma_get_ctx(file, cmd.id); if (IS_ERR(ctx)) return PTR_ERR(ctx); - if (unlikely(cmd.optlen > KMALLOC_MAX_SIZE)) - return -EINVAL; - optval = memdup_user(u64_to_user_ptr(cmd.optval), cmd.optlen); if (IS_ERR(optval)) { @@ -1384,7 +1396,7 @@ static ssize_t ucma_process_join(struct ucma_file *file, else return -EINVAL; - ctx = ucma_get_ctx(file, cmd->id); + ctx = ucma_get_ctx_dev(file, cmd->id); if (IS_ERR(ctx)) return PTR_ERR(ctx); diff --git a/drivers/infiniband/core/uverbs_cmd.c b/drivers/infiniband/core/uverbs_cmd.c index 13cb5e4deb86..21a887c9523b 100644 --- a/drivers/infiniband/core/uverbs_cmd.c +++ b/drivers/infiniband/core/uverbs_cmd.c @@ -691,6 +691,7 @@ ssize_t ib_uverbs_reg_mr(struct ib_uverbs_file *file, mr->device = pd->device; mr->pd = pd; + mr->dm = NULL; mr->uobject = uobj; atomic_inc(&pd->usecnt); mr->res.type = RDMA_RESTRACK_MR; @@ -765,6 +766,11 @@ ssize_t ib_uverbs_rereg_mr(struct ib_uverbs_file *file, mr = uobj->object; + if (mr->dm) { + ret = -EINVAL; + goto put_uobjs; + } + if (cmd.flags & IB_MR_REREG_ACCESS) { ret = ib_check_mr_access(cmd.access_flags); if (ret) diff --git a/drivers/infiniband/core/uverbs_ioctl.c b/drivers/infiniband/core/uverbs_ioctl.c index 8c93970dc8f1..8d32c4ae368c 100644 --- a/drivers/infiniband/core/uverbs_ioctl.c +++ b/drivers/infiniband/core/uverbs_ioctl.c @@ -234,6 +234,15 @@ static int uverbs_validate_kernel_mandatory(const struct uverbs_method_spec *met return -EINVAL; } + for (; i < method_spec->num_buckets; i++) { + struct uverbs_attr_spec_hash *attr_spec_bucket = + method_spec->attr_buckets[i]; + + if (!bitmap_empty(attr_spec_bucket->mandatory_attrs_bitmask, + attr_spec_bucket->num_attrs)) + return -EINVAL; + } + return 0; } diff --git a/drivers/infiniband/core/uverbs_std_types_flow_action.c b/drivers/infiniband/core/uverbs_std_types_flow_action.c index cbcec3da12f6..b4f016dfa23d 100644 --- a/drivers/infiniband/core/uverbs_std_types_flow_action.c +++ b/drivers/infiniband/core/uverbs_std_types_flow_action.c @@ -363,28 +363,28 @@ static int UVERBS_HANDLER(UVERBS_METHOD_FLOW_ACTION_ESP_MODIFY)(struct ib_device static const struct uverbs_attr_spec uverbs_flow_action_esp_keymat[] = { [IB_UVERBS_FLOW_ACTION_ESP_KEYMAT_AES_GCM] = { - .ptr = { + { .ptr = { .type = UVERBS_ATTR_TYPE_PTR_IN, UVERBS_ATTR_TYPE(struct ib_uverbs_flow_action_esp_keymat_aes_gcm), .flags = UVERBS_ATTR_SPEC_F_MIN_SZ_OR_ZERO, - }, + } }, }, }; static const struct uverbs_attr_spec uverbs_flow_action_esp_replay[] = { [IB_UVERBS_FLOW_ACTION_ESP_REPLAY_NONE] = { - .ptr = { + { .ptr = { .type = UVERBS_ATTR_TYPE_PTR_IN, /* No need to specify any data */ .len = 0, - } + } } }, [IB_UVERBS_FLOW_ACTION_ESP_REPLAY_BMP] = { - .ptr = { + { .ptr = { .type = UVERBS_ATTR_TYPE_PTR_IN, UVERBS_ATTR_STRUCT(struct ib_uverbs_flow_action_esp_replay_bmp, size), .flags = UVERBS_ATTR_SPEC_F_MIN_SZ_OR_ZERO, - } + } } }, }; diff --git a/drivers/infiniband/core/verbs.c b/drivers/infiniband/core/verbs.c index 7eff3aeffe01..6ddfb1fade79 100644 --- a/drivers/infiniband/core/verbs.c +++ b/drivers/infiniband/core/verbs.c @@ -1656,6 +1656,7 @@ struct ib_mr *ib_alloc_mr(struct ib_pd *pd, if (!IS_ERR(mr)) { mr->device = pd->device; mr->pd = pd; + mr->dm = NULL; mr->uobject = NULL; atomic_inc(&pd->usecnt); mr->need_inval = false; diff --git a/drivers/infiniband/hw/cxgb4/cq.c b/drivers/infiniband/hw/cxgb4/cq.c index 6f2b26126c64..2be2e1ac1b5f 100644 --- a/drivers/infiniband/hw/cxgb4/cq.c +++ b/drivers/infiniband/hw/cxgb4/cq.c @@ -315,7 +315,7 @@ static void advance_oldest_read(struct t4_wq *wq) * Deal with out-of-order and/or completions that complete * prior unsignalled WRs. */ -void c4iw_flush_hw_cq(struct c4iw_cq *chp) +void c4iw_flush_hw_cq(struct c4iw_cq *chp, struct c4iw_qp *flush_qhp) { struct t4_cqe *hw_cqe, *swcqe, read_cqe; struct c4iw_qp *qhp; @@ -339,6 +339,13 @@ void c4iw_flush_hw_cq(struct c4iw_cq *chp) if (qhp == NULL) goto next_cqe; + if (flush_qhp != qhp) { + spin_lock(&qhp->lock); + + if (qhp->wq.flushed == 1) + goto next_cqe; + } + if (CQE_OPCODE(hw_cqe) == FW_RI_TERMINATE) goto next_cqe; @@ -390,6 +397,8 @@ void c4iw_flush_hw_cq(struct c4iw_cq *chp) next_cqe: t4_hwcq_consume(&chp->cq); ret = t4_next_hw_cqe(&chp->cq, &hw_cqe); + if (qhp && flush_qhp != qhp) + spin_unlock(&qhp->lock); } } diff --git a/drivers/infiniband/hw/cxgb4/device.c b/drivers/infiniband/hw/cxgb4/device.c index feeb8ee6f4a2..44161ca4d2a8 100644 --- a/drivers/infiniband/hw/cxgb4/device.c +++ b/drivers/infiniband/hw/cxgb4/device.c @@ -875,6 +875,11 @@ static int c4iw_rdev_open(struct c4iw_rdev *rdev) rdev->status_page->db_off = 0; + init_completion(&rdev->rqt_compl); + init_completion(&rdev->pbl_compl); + kref_init(&rdev->rqt_kref); + kref_init(&rdev->pbl_kref); + return 0; err_free_status_page_and_wr_log: if (c4iw_wr_log && rdev->wr_log) @@ -893,13 +898,15 @@ destroy_resource: static void c4iw_rdev_close(struct c4iw_rdev *rdev) { - destroy_workqueue(rdev->free_workq); kfree(rdev->wr_log); c4iw_release_dev_ucontext(rdev, &rdev->uctx); free_page((unsigned long)rdev->status_page); c4iw_pblpool_destroy(rdev); c4iw_rqtpool_destroy(rdev); + wait_for_completion(&rdev->pbl_compl); + wait_for_completion(&rdev->rqt_compl); c4iw_ocqp_pool_destroy(rdev); + destroy_workqueue(rdev->free_workq); c4iw_destroy_resource(&rdev->resource); } diff --git a/drivers/infiniband/hw/cxgb4/iw_cxgb4.h b/drivers/infiniband/hw/cxgb4/iw_cxgb4.h index cc929002c05e..831027717121 100644 --- a/drivers/infiniband/hw/cxgb4/iw_cxgb4.h +++ b/drivers/infiniband/hw/cxgb4/iw_cxgb4.h @@ -185,6 +185,10 @@ struct c4iw_rdev { struct wr_log_entry *wr_log; int wr_log_size; struct workqueue_struct *free_workq; + struct completion rqt_compl; + struct completion pbl_compl; + struct kref rqt_kref; + struct kref pbl_kref; }; static inline int c4iw_fatal_error(struct c4iw_rdev *rdev) @@ -1049,7 +1053,7 @@ u32 c4iw_pblpool_alloc(struct c4iw_rdev *rdev, int size); void c4iw_pblpool_free(struct c4iw_rdev *rdev, u32 addr, int size); u32 c4iw_ocqp_pool_alloc(struct c4iw_rdev *rdev, int size); void c4iw_ocqp_pool_free(struct c4iw_rdev *rdev, u32 addr, int size); -void c4iw_flush_hw_cq(struct c4iw_cq *chp); +void c4iw_flush_hw_cq(struct c4iw_cq *chp, struct c4iw_qp *flush_qhp); void c4iw_count_rcqes(struct t4_cq *cq, struct t4_wq *wq, int *count); int c4iw_ep_disconnect(struct c4iw_ep *ep, int abrupt, gfp_t gfp); int c4iw_flush_rq(struct t4_wq *wq, struct t4_cq *cq, int count); diff --git a/drivers/infiniband/hw/cxgb4/qp.c b/drivers/infiniband/hw/cxgb4/qp.c index de77b6027d69..ae167b686608 100644 --- a/drivers/infiniband/hw/cxgb4/qp.c +++ b/drivers/infiniband/hw/cxgb4/qp.c @@ -1343,12 +1343,12 @@ static void __flush_qp(struct c4iw_qp *qhp, struct c4iw_cq *rchp, qhp->wq.flushed = 1; t4_set_wq_in_error(&qhp->wq); - c4iw_flush_hw_cq(rchp); + c4iw_flush_hw_cq(rchp, qhp); c4iw_count_rcqes(&rchp->cq, &qhp->wq, &count); rq_flushed = c4iw_flush_rq(&qhp->wq, &rchp->cq, count); if (schp != rchp) - c4iw_flush_hw_cq(schp); + c4iw_flush_hw_cq(schp, qhp); sq_flushed = c4iw_flush_sq(qhp); spin_unlock(&qhp->lock); diff --git a/drivers/infiniband/hw/cxgb4/resource.c b/drivers/infiniband/hw/cxgb4/resource.c index 3cf25997ed2b..0ef25ae05e6f 100644 --- a/drivers/infiniband/hw/cxgb4/resource.c +++ b/drivers/infiniband/hw/cxgb4/resource.c @@ -260,12 +260,22 @@ u32 c4iw_pblpool_alloc(struct c4iw_rdev *rdev, int size) rdev->stats.pbl.cur += roundup(size, 1 << MIN_PBL_SHIFT); if (rdev->stats.pbl.cur > rdev->stats.pbl.max) rdev->stats.pbl.max = rdev->stats.pbl.cur; + kref_get(&rdev->pbl_kref); } else rdev->stats.pbl.fail++; mutex_unlock(&rdev->stats.lock); return (u32)addr; } +static void destroy_pblpool(struct kref *kref) +{ + struct c4iw_rdev *rdev; + + rdev = container_of(kref, struct c4iw_rdev, pbl_kref); + gen_pool_destroy(rdev->pbl_pool); + complete(&rdev->pbl_compl); +} + void c4iw_pblpool_free(struct c4iw_rdev *rdev, u32 addr, int size) { pr_debug("addr 0x%x size %d\n", addr, size); @@ -273,6 +283,7 @@ void c4iw_pblpool_free(struct c4iw_rdev *rdev, u32 addr, int size) rdev->stats.pbl.cur -= roundup(size, 1 << MIN_PBL_SHIFT); mutex_unlock(&rdev->stats.lock); gen_pool_free(rdev->pbl_pool, (unsigned long)addr, size); + kref_put(&rdev->pbl_kref, destroy_pblpool); } int c4iw_pblpool_create(struct c4iw_rdev *rdev) @@ -310,7 +321,7 @@ int c4iw_pblpool_create(struct c4iw_rdev *rdev) void c4iw_pblpool_destroy(struct c4iw_rdev *rdev) { - gen_pool_destroy(rdev->pbl_pool); + kref_put(&rdev->pbl_kref, destroy_pblpool); } /* @@ -331,12 +342,22 @@ u32 c4iw_rqtpool_alloc(struct c4iw_rdev *rdev, int size) rdev->stats.rqt.cur += roundup(size << 6, 1 << MIN_RQT_SHIFT); if (rdev->stats.rqt.cur > rdev->stats.rqt.max) rdev->stats.rqt.max = rdev->stats.rqt.cur; + kref_get(&rdev->rqt_kref); } else rdev->stats.rqt.fail++; mutex_unlock(&rdev->stats.lock); return (u32)addr; } +static void destroy_rqtpool(struct kref *kref) +{ + struct c4iw_rdev *rdev; + + rdev = container_of(kref, struct c4iw_rdev, rqt_kref); + gen_pool_destroy(rdev->rqt_pool); + complete(&rdev->rqt_compl); +} + void c4iw_rqtpool_free(struct c4iw_rdev *rdev, u32 addr, int size) { pr_debug("addr 0x%x size %d\n", addr, size << 6); @@ -344,6 +365,7 @@ void c4iw_rqtpool_free(struct c4iw_rdev *rdev, u32 addr, int size) rdev->stats.rqt.cur -= roundup(size << 6, 1 << MIN_RQT_SHIFT); mutex_unlock(&rdev->stats.lock); gen_pool_free(rdev->rqt_pool, (unsigned long)addr, size << 6); + kref_put(&rdev->rqt_kref, destroy_rqtpool); } int c4iw_rqtpool_create(struct c4iw_rdev *rdev) @@ -380,7 +402,7 @@ int c4iw_rqtpool_create(struct c4iw_rdev *rdev) void c4iw_rqtpool_destroy(struct c4iw_rdev *rdev) { - gen_pool_destroy(rdev->rqt_pool); + kref_put(&rdev->rqt_kref, destroy_rqtpool); } /* diff --git a/drivers/infiniband/hw/hfi1/affinity.c b/drivers/infiniband/hw/hfi1/affinity.c index a97055dd4fbd..b5fab55cc275 100644 --- a/drivers/infiniband/hw/hfi1/affinity.c +++ b/drivers/infiniband/hw/hfi1/affinity.c @@ -412,7 +412,6 @@ static void hfi1_cleanup_sdma_notifier(struct hfi1_msix_entry *msix) static int get_irq_affinity(struct hfi1_devdata *dd, struct hfi1_msix_entry *msix) { - int ret; cpumask_var_t diff; struct hfi1_affinity_node *entry; struct cpu_mask_set *set = NULL; @@ -424,10 +423,6 @@ static int get_irq_affinity(struct hfi1_devdata *dd, extra[0] = '\0'; cpumask_clear(&msix->mask); - ret = zalloc_cpumask_var(&diff, GFP_KERNEL); - if (!ret) - return -ENOMEM; - entry = node_affinity_lookup(dd->node); switch (msix->type) { @@ -458,6 +453,9 @@ static int get_irq_affinity(struct hfi1_devdata *dd, * finds its CPU here. */ if (cpu == -1 && set) { + if (!zalloc_cpumask_var(&diff, GFP_KERNEL)) + return -ENOMEM; + if (cpumask_equal(&set->mask, &set->used)) { /* * We've used up all the CPUs, bump up the generation @@ -469,6 +467,8 @@ static int get_irq_affinity(struct hfi1_devdata *dd, cpumask_andnot(diff, &set->mask, &set->used); cpu = cpumask_first(diff); cpumask_set_cpu(cpu, &set->used); + + free_cpumask_var(diff); } cpumask_set_cpu(cpu, &msix->mask); @@ -482,7 +482,6 @@ static int get_irq_affinity(struct hfi1_devdata *dd, hfi1_setup_sdma_notifier(msix); } - free_cpumask_var(diff); return 0; } diff --git a/drivers/infiniband/hw/hfi1/driver.c b/drivers/infiniband/hw/hfi1/driver.c index 46d1475b2154..bd837a048bf4 100644 --- a/drivers/infiniband/hw/hfi1/driver.c +++ b/drivers/infiniband/hw/hfi1/driver.c @@ -433,31 +433,43 @@ void hfi1_process_ecn_slowpath(struct rvt_qp *qp, struct hfi1_packet *pkt, bool do_cnp) { struct hfi1_ibport *ibp = to_iport(qp->ibqp.device, qp->port_num); + struct hfi1_pportdata *ppd = ppd_from_ibp(ibp); struct ib_other_headers *ohdr = pkt->ohdr; struct ib_grh *grh = pkt->grh; u32 rqpn = 0, bth1; - u16 pkey, rlid, dlid = ib_get_dlid(pkt->hdr); + u16 pkey; + u32 rlid, slid, dlid = 0; u8 hdr_type, sc, svc_type; bool is_mcast = false; + /* can be called from prescan */ if (pkt->etype == RHF_RCV_TYPE_BYPASS) { is_mcast = hfi1_is_16B_mcast(dlid); pkey = hfi1_16B_get_pkey(pkt->hdr); sc = hfi1_16B_get_sc(pkt->hdr); + dlid = hfi1_16B_get_dlid(pkt->hdr); + slid = hfi1_16B_get_slid(pkt->hdr); hdr_type = HFI1_PKT_TYPE_16B; } else { is_mcast = (dlid > be16_to_cpu(IB_MULTICAST_LID_BASE)) && (dlid != be16_to_cpu(IB_LID_PERMISSIVE)); pkey = ib_bth_get_pkey(ohdr); sc = hfi1_9B_get_sc5(pkt->hdr, pkt->rhf); + dlid = ib_get_dlid(pkt->hdr); + slid = ib_get_slid(pkt->hdr); hdr_type = HFI1_PKT_TYPE_9B; } switch (qp->ibqp.qp_type) { + case IB_QPT_UD: + dlid = ppd->lid; + rlid = slid; + rqpn = ib_get_sqpn(pkt->ohdr); + svc_type = IB_CC_SVCTYPE_UD; + break; case IB_QPT_SMI: case IB_QPT_GSI: - case IB_QPT_UD: - rlid = ib_get_slid(pkt->hdr); + rlid = slid; rqpn = ib_get_sqpn(pkt->ohdr); svc_type = IB_CC_SVCTYPE_UD; break; @@ -482,7 +494,6 @@ void hfi1_process_ecn_slowpath(struct rvt_qp *qp, struct hfi1_packet *pkt, dlid, rlid, sc, grh); if (!is_mcast && (bth1 & IB_BECN_SMASK)) { - struct hfi1_pportdata *ppd = ppd_from_ibp(ibp); u32 lqpn = bth1 & RVT_QPN_MASK; u8 sl = ibp->sc_to_sl[sc]; diff --git a/drivers/infiniband/hw/hfi1/hfi.h b/drivers/infiniband/hw/hfi1/hfi.h index 32c48265405e..cac2c62bc42d 100644 --- a/drivers/infiniband/hw/hfi1/hfi.h +++ b/drivers/infiniband/hw/hfi1/hfi.h @@ -1537,13 +1537,13 @@ void set_link_ipg(struct hfi1_pportdata *ppd); void process_becn(struct hfi1_pportdata *ppd, u8 sl, u32 rlid, u32 lqpn, u32 rqpn, u8 svc_type); void return_cnp(struct hfi1_ibport *ibp, struct rvt_qp *qp, u32 remote_qpn, - u32 pkey, u32 slid, u32 dlid, u8 sc5, + u16 pkey, u32 slid, u32 dlid, u8 sc5, const struct ib_grh *old_grh); void return_cnp_16B(struct hfi1_ibport *ibp, struct rvt_qp *qp, - u32 remote_qpn, u32 pkey, u32 slid, u32 dlid, + u32 remote_qpn, u16 pkey, u32 slid, u32 dlid, u8 sc5, const struct ib_grh *old_grh); typedef void (*hfi1_handle_cnp)(struct hfi1_ibport *ibp, struct rvt_qp *qp, - u32 remote_qpn, u32 pkey, u32 slid, u32 dlid, + u32 remote_qpn, u16 pkey, u32 slid, u32 dlid, u8 sc5, const struct ib_grh *old_grh); #define PKEY_CHECK_INVALID -1 @@ -2437,7 +2437,7 @@ static inline void hfi1_make_16b_hdr(struct hfi1_16b_header *hdr, ((slid >> OPA_16B_SLID_SHIFT) << OPA_16B_SLID_HIGH_SHIFT); lrh2 = (lrh2 & ~OPA_16B_DLID_MASK) | ((dlid >> OPA_16B_DLID_SHIFT) << OPA_16B_DLID_HIGH_SHIFT); - lrh2 = (lrh2 & ~OPA_16B_PKEY_MASK) | (pkey << OPA_16B_PKEY_SHIFT); + lrh2 = (lrh2 & ~OPA_16B_PKEY_MASK) | ((u32)pkey << OPA_16B_PKEY_SHIFT); lrh2 = (lrh2 & ~OPA_16B_L4_MASK) | l4; hdr->lrh[0] = lrh0; diff --git a/drivers/infiniband/hw/hfi1/init.c b/drivers/infiniband/hw/hfi1/init.c index 33eba2356742..6309edf811df 100644 --- a/drivers/infiniband/hw/hfi1/init.c +++ b/drivers/infiniband/hw/hfi1/init.c @@ -88,9 +88,9 @@ * pio buffers per ctxt, etc.) Zero means use one user context per CPU. */ int num_user_contexts = -1; -module_param_named(num_user_contexts, num_user_contexts, uint, S_IRUGO); +module_param_named(num_user_contexts, num_user_contexts, int, 0444); MODULE_PARM_DESC( - num_user_contexts, "Set max number of user contexts to use"); + num_user_contexts, "Set max number of user contexts to use (default: -1 will use the real (non-HT) CPU count)"); uint krcvqs[RXE_NUM_DATA_VL]; int krcvqsset; @@ -1209,30 +1209,49 @@ static void finalize_asic_data(struct hfi1_devdata *dd, kfree(ad); } -static void __hfi1_free_devdata(struct kobject *kobj) +/** + * hfi1_clean_devdata - cleans up per-unit data structure + * @dd: pointer to a valid devdata structure + * + * It cleans up all data structures set up by + * by hfi1_alloc_devdata(). + */ +static void hfi1_clean_devdata(struct hfi1_devdata *dd) { - struct hfi1_devdata *dd = - container_of(kobj, struct hfi1_devdata, kobj); struct hfi1_asic_data *ad; unsigned long flags; spin_lock_irqsave(&hfi1_devs_lock, flags); - idr_remove(&hfi1_unit_table, dd->unit); - list_del(&dd->list); + if (!list_empty(&dd->list)) { + idr_remove(&hfi1_unit_table, dd->unit); + list_del_init(&dd->list); + } ad = release_asic_data(dd); spin_unlock_irqrestore(&hfi1_devs_lock, flags); - if (ad) - finalize_asic_data(dd, ad); + + finalize_asic_data(dd, ad); free_platform_config(dd); rcu_barrier(); /* wait for rcu callbacks to complete */ free_percpu(dd->int_counter); free_percpu(dd->rcv_limit); free_percpu(dd->send_schedule); free_percpu(dd->tx_opstats); + dd->int_counter = NULL; + dd->rcv_limit = NULL; + dd->send_schedule = NULL; + dd->tx_opstats = NULL; sdma_clean(dd, dd->num_sdma); rvt_dealloc_device(&dd->verbs_dev.rdi); } +static void __hfi1_free_devdata(struct kobject *kobj) +{ + struct hfi1_devdata *dd = + container_of(kobj, struct hfi1_devdata, kobj); + + hfi1_clean_devdata(dd); +} + static struct kobj_type hfi1_devdata_type = { .release = __hfi1_free_devdata, }; @@ -1265,6 +1284,8 @@ struct hfi1_devdata *hfi1_alloc_devdata(struct pci_dev *pdev, size_t extra) return ERR_PTR(-ENOMEM); dd->num_pports = nports; dd->pport = (struct hfi1_pportdata *)(dd + 1); + dd->pcidev = pdev; + pci_set_drvdata(pdev, dd); INIT_LIST_HEAD(&dd->list); idr_preload(GFP_KERNEL); @@ -1331,9 +1352,7 @@ struct hfi1_devdata *hfi1_alloc_devdata(struct pci_dev *pdev, size_t extra) return dd; bail: - if (!list_empty(&dd->list)) - list_del_init(&dd->list); - rvt_dealloc_device(&dd->verbs_dev.rdi); + hfi1_clean_devdata(dd); return ERR_PTR(ret); } diff --git a/drivers/infiniband/hw/hfi1/pcie.c b/drivers/infiniband/hw/hfi1/pcie.c index 83d66e862207..c1c982908b4b 100644 --- a/drivers/infiniband/hw/hfi1/pcie.c +++ b/drivers/infiniband/hw/hfi1/pcie.c @@ -163,9 +163,6 @@ int hfi1_pcie_ddinit(struct hfi1_devdata *dd, struct pci_dev *pdev) resource_size_t addr; int ret = 0; - dd->pcidev = pdev; - pci_set_drvdata(pdev, dd); - addr = pci_resource_start(pdev, 0); len = pci_resource_len(pdev, 0); diff --git a/drivers/infiniband/hw/hfi1/platform.c b/drivers/infiniband/hw/hfi1/platform.c index d486355880cb..cbf7faa5038c 100644 --- a/drivers/infiniband/hw/hfi1/platform.c +++ b/drivers/infiniband/hw/hfi1/platform.c @@ -199,6 +199,7 @@ void free_platform_config(struct hfi1_devdata *dd) { /* Release memory allocated for eprom or fallback file read. */ kfree(dd->platform_config.data); + dd->platform_config.data = NULL; } void get_port_type(struct hfi1_pportdata *ppd) diff --git a/drivers/infiniband/hw/hfi1/qsfp.c b/drivers/infiniband/hw/hfi1/qsfp.c index 1869f639c3ae..b5966991d647 100644 --- a/drivers/infiniband/hw/hfi1/qsfp.c +++ b/drivers/infiniband/hw/hfi1/qsfp.c @@ -204,6 +204,8 @@ static void clean_i2c_bus(struct hfi1_i2c_bus *bus) void clean_up_i2c(struct hfi1_devdata *dd, struct hfi1_asic_data *ad) { + if (!ad) + return; clean_i2c_bus(ad->i2c_bus0); ad->i2c_bus0 = NULL; clean_i2c_bus(ad->i2c_bus1); diff --git a/drivers/infiniband/hw/hfi1/ruc.c b/drivers/infiniband/hw/hfi1/ruc.c index 3daa94bdae3a..c0071ca4147a 100644 --- a/drivers/infiniband/hw/hfi1/ruc.c +++ b/drivers/infiniband/hw/hfi1/ruc.c @@ -733,6 +733,20 @@ static inline void hfi1_make_ruc_bth(struct rvt_qp *qp, ohdr->bth[2] = cpu_to_be32(bth2); } +/** + * hfi1_make_ruc_header_16B - build a 16B header + * @qp: the queue pair + * @ohdr: a pointer to the destination header memory + * @bth0: bth0 passed in from the RC/UC builder + * @bth2: bth2 passed in from the RC/UC builder + * @middle: non zero implies indicates ahg "could" be used + * @ps: the current packet state + * + * This routine may disarm ahg under these situations: + * - packet needs a GRH + * - BECN needed + * - migration state not IB_MIG_MIGRATED + */ static inline void hfi1_make_ruc_header_16B(struct rvt_qp *qp, struct ib_other_headers *ohdr, u32 bth0, u32 bth2, int middle, @@ -777,6 +791,12 @@ static inline void hfi1_make_ruc_header_16B(struct rvt_qp *qp, else middle = 0; + if (qp->s_flags & RVT_S_ECN) { + qp->s_flags &= ~RVT_S_ECN; + /* we recently received a FECN, so return a BECN */ + becn = true; + middle = 0; + } if (middle) build_ahg(qp, bth2); else @@ -784,11 +804,6 @@ static inline void hfi1_make_ruc_header_16B(struct rvt_qp *qp, bth0 |= pkey; bth0 |= extra_bytes << 20; - if (qp->s_flags & RVT_S_ECN) { - qp->s_flags &= ~RVT_S_ECN; - /* we recently received a FECN, so return a BECN */ - becn = true; - } hfi1_make_ruc_bth(qp, ohdr, bth0, bth1, bth2); if (!ppd->lid) @@ -806,6 +821,20 @@ static inline void hfi1_make_ruc_header_16B(struct rvt_qp *qp, pkey, becn, 0, l4, priv->s_sc); } +/** + * hfi1_make_ruc_header_9B - build a 9B header + * @qp: the queue pair + * @ohdr: a pointer to the destination header memory + * @bth0: bth0 passed in from the RC/UC builder + * @bth2: bth2 passed in from the RC/UC builder + * @middle: non zero implies indicates ahg "could" be used + * @ps: the current packet state + * + * This routine may disarm ahg under these situations: + * - packet needs a GRH + * - BECN needed + * - migration state not IB_MIG_MIGRATED + */ static inline void hfi1_make_ruc_header_9B(struct rvt_qp *qp, struct ib_other_headers *ohdr, u32 bth0, u32 bth2, int middle, @@ -839,6 +868,12 @@ static inline void hfi1_make_ruc_header_9B(struct rvt_qp *qp, else middle = 0; + if (qp->s_flags & RVT_S_ECN) { + qp->s_flags &= ~RVT_S_ECN; + /* we recently received a FECN, so return a BECN */ + bth1 |= (IB_BECN_MASK << IB_BECN_SHIFT); + middle = 0; + } if (middle) build_ahg(qp, bth2); else @@ -846,11 +881,6 @@ static inline void hfi1_make_ruc_header_9B(struct rvt_qp *qp, bth0 |= pkey; bth0 |= extra_bytes << 20; - if (qp->s_flags & RVT_S_ECN) { - qp->s_flags &= ~RVT_S_ECN; - /* we recently received a FECN, so return a BECN */ - bth1 |= (IB_BECN_MASK << IB_BECN_SHIFT); - } hfi1_make_ruc_bth(qp, ohdr, bth0, bth1, bth2); hfi1_make_ib_hdr(&ps->s_txreq->phdr.hdr.ibh, lrh0, diff --git a/drivers/infiniband/hw/hfi1/ud.c b/drivers/infiniband/hw/hfi1/ud.c index bcf3b0bebac8..69c17a5ef038 100644 --- a/drivers/infiniband/hw/hfi1/ud.c +++ b/drivers/infiniband/hw/hfi1/ud.c @@ -628,7 +628,7 @@ int hfi1_lookup_pkey_idx(struct hfi1_ibport *ibp, u16 pkey) } void return_cnp_16B(struct hfi1_ibport *ibp, struct rvt_qp *qp, - u32 remote_qpn, u32 pkey, u32 slid, u32 dlid, + u32 remote_qpn, u16 pkey, u32 slid, u32 dlid, u8 sc5, const struct ib_grh *old_grh) { u64 pbc, pbc_flags = 0; @@ -687,7 +687,7 @@ void return_cnp_16B(struct hfi1_ibport *ibp, struct rvt_qp *qp, } void return_cnp(struct hfi1_ibport *ibp, struct rvt_qp *qp, u32 remote_qpn, - u32 pkey, u32 slid, u32 dlid, u8 sc5, + u16 pkey, u32 slid, u32 dlid, u8 sc5, const struct ib_grh *old_grh) { u64 pbc, pbc_flags = 0; diff --git a/drivers/infiniband/hw/hns/hns_roce_hem.c b/drivers/infiniband/hw/hns/hns_roce_hem.c index 0eeabfbee192..63b5b3edabcb 100644 --- a/drivers/infiniband/hw/hns/hns_roce_hem.c +++ b/drivers/infiniband/hw/hns/hns_roce_hem.c @@ -912,7 +912,7 @@ int hns_roce_init_hem_table(struct hns_roce_dev *hr_dev, obj_per_chunk = buf_chunk_size / obj_size; num_hem = (nobj + obj_per_chunk - 1) / obj_per_chunk; bt_chunk_num = bt_chunk_size / 8; - if (table->type >= HEM_TYPE_MTT) + if (type >= HEM_TYPE_MTT) num_bt_l0 = bt_chunk_num; table->hem = kcalloc(num_hem, sizeof(*table->hem), @@ -920,7 +920,7 @@ int hns_roce_init_hem_table(struct hns_roce_dev *hr_dev, if (!table->hem) goto err_kcalloc_hem_buf; - if (check_whether_bt_num_3(table->type, hop_num)) { + if (check_whether_bt_num_3(type, hop_num)) { unsigned long num_bt_l1; num_bt_l1 = (num_hem + bt_chunk_num - 1) / @@ -939,8 +939,8 @@ int hns_roce_init_hem_table(struct hns_roce_dev *hr_dev, goto err_kcalloc_l1_dma; } - if (check_whether_bt_num_2(table->type, hop_num) || - check_whether_bt_num_3(table->type, hop_num)) { + if (check_whether_bt_num_2(type, hop_num) || + check_whether_bt_num_3(type, hop_num)) { table->bt_l0 = kcalloc(num_bt_l0, sizeof(*table->bt_l0), GFP_KERNEL); if (!table->bt_l0) @@ -1039,14 +1039,14 @@ void hns_roce_cleanup_hem_table(struct hns_roce_dev *hr_dev, void hns_roce_cleanup_hem(struct hns_roce_dev *hr_dev) { hns_roce_cleanup_hem_table(hr_dev, &hr_dev->cq_table.table); - hns_roce_cleanup_hem_table(hr_dev, &hr_dev->qp_table.irrl_table); if (hr_dev->caps.trrl_entry_sz) hns_roce_cleanup_hem_table(hr_dev, &hr_dev->qp_table.trrl_table); + hns_roce_cleanup_hem_table(hr_dev, &hr_dev->qp_table.irrl_table); hns_roce_cleanup_hem_table(hr_dev, &hr_dev->qp_table.qp_table); hns_roce_cleanup_hem_table(hr_dev, &hr_dev->mr_table.mtpt_table); - hns_roce_cleanup_hem_table(hr_dev, &hr_dev->mr_table.mtt_table); if (hns_roce_check_whether_mhop(hr_dev, HEM_TYPE_CQE)) hns_roce_cleanup_hem_table(hr_dev, &hr_dev->mr_table.mtt_cqe_table); + hns_roce_cleanup_hem_table(hr_dev, &hr_dev->mr_table.mtt_table); } diff --git a/drivers/infiniband/hw/hns/hns_roce_hw_v2.c b/drivers/infiniband/hw/hns/hns_roce_hw_v2.c index 8b84ab7800d8..25916e8522ed 100644 --- a/drivers/infiniband/hw/hns/hns_roce_hw_v2.c +++ b/drivers/infiniband/hw/hns/hns_roce_hw_v2.c @@ -71,6 +71,11 @@ static int set_rwqe_data_seg(struct ib_qp *ibqp, struct ib_send_wr *wr, return -EINVAL; } + if (wr->opcode == IB_WR_RDMA_READ) { + dev_err(hr_dev->dev, "Not support inline data!\n"); + return -EINVAL; + } + for (i = 0; i < wr->num_sge; i++) { memcpy(wqe, ((void *)wr->sg_list[i].addr), wr->sg_list[i].length); @@ -148,7 +153,7 @@ static int hns_roce_v2_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, ibqp->qp_type != IB_QPT_GSI && ibqp->qp_type != IB_QPT_UD)) { dev_err(dev, "Not supported QP(0x%x)type!\n", ibqp->qp_type); - *bad_wr = NULL; + *bad_wr = wr; return -EOPNOTSUPP; } @@ -182,7 +187,8 @@ static int hns_roce_v2_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, qp->sq.wrid[(qp->sq.head + nreq) & (qp->sq.wqe_cnt - 1)] = wr->wr_id; - owner_bit = ~(qp->sq.head >> ilog2(qp->sq.wqe_cnt)) & 0x1; + owner_bit = + ~(((qp->sq.head + nreq) >> ilog2(qp->sq.wqe_cnt)) & 0x1); /* Corresponding to the QP type, wqe process separately */ if (ibqp->qp_type == IB_QPT_GSI) { @@ -456,6 +462,7 @@ static int hns_roce_v2_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, } else { dev_err(dev, "Illegal qp_type(0x%x)\n", ibqp->qp_type); spin_unlock_irqrestore(&qp->sq.lock, flags); + *bad_wr = wr; return -EOPNOTSUPP; } } @@ -2592,10 +2599,12 @@ static void modify_qp_init_to_init(struct ib_qp *ibqp, roce_set_field(qpc_mask->byte_4_sqpn_tst, V2_QPC_BYTE_4_SQPN_M, V2_QPC_BYTE_4_SQPN_S, 0); - roce_set_field(context->byte_56_dqpn_err, V2_QPC_BYTE_56_DQPN_M, - V2_QPC_BYTE_56_DQPN_S, hr_qp->qpn); - roce_set_field(qpc_mask->byte_56_dqpn_err, V2_QPC_BYTE_56_DQPN_M, - V2_QPC_BYTE_56_DQPN_S, 0); + if (attr_mask & IB_QP_DEST_QPN) { + roce_set_field(context->byte_56_dqpn_err, V2_QPC_BYTE_56_DQPN_M, + V2_QPC_BYTE_56_DQPN_S, hr_qp->qpn); + roce_set_field(qpc_mask->byte_56_dqpn_err, + V2_QPC_BYTE_56_DQPN_M, V2_QPC_BYTE_56_DQPN_S, 0); + } roce_set_field(context->byte_168_irrl_idx, V2_QPC_BYTE_168_SQ_SHIFT_BAK_M, V2_QPC_BYTE_168_SQ_SHIFT_BAK_S, @@ -2650,8 +2659,7 @@ static int modify_qp_init_to_rtr(struct ib_qp *ibqp, return -EINVAL; } - if ((attr_mask & IB_QP_ALT_PATH) || (attr_mask & IB_QP_ACCESS_FLAGS) || - (attr_mask & IB_QP_PKEY_INDEX) || (attr_mask & IB_QP_QKEY)) { + if (attr_mask & IB_QP_ALT_PATH) { dev_err(dev, "INIT2RTR attr_mask (0x%x) error\n", attr_mask); return -EINVAL; } @@ -2800,10 +2808,12 @@ static int modify_qp_init_to_rtr(struct ib_qp *ibqp, V2_QPC_BYTE_140_RR_MAX_S, 0); } - roce_set_field(context->byte_56_dqpn_err, V2_QPC_BYTE_56_DQPN_M, - V2_QPC_BYTE_56_DQPN_S, attr->dest_qp_num); - roce_set_field(qpc_mask->byte_56_dqpn_err, V2_QPC_BYTE_56_DQPN_M, - V2_QPC_BYTE_56_DQPN_S, 0); + if (attr_mask & IB_QP_DEST_QPN) { + roce_set_field(context->byte_56_dqpn_err, V2_QPC_BYTE_56_DQPN_M, + V2_QPC_BYTE_56_DQPN_S, attr->dest_qp_num); + roce_set_field(qpc_mask->byte_56_dqpn_err, + V2_QPC_BYTE_56_DQPN_M, V2_QPC_BYTE_56_DQPN_S, 0); + } /* Configure GID index */ port_num = rdma_ah_get_port_num(&attr->ah_attr); @@ -2845,7 +2855,7 @@ static int modify_qp_init_to_rtr(struct ib_qp *ibqp, if (ibqp->qp_type == IB_QPT_GSI || ibqp->qp_type == IB_QPT_UD) roce_set_field(context->byte_24_mtu_tc, V2_QPC_BYTE_24_MTU_M, V2_QPC_BYTE_24_MTU_S, IB_MTU_4096); - else + else if (attr_mask & IB_QP_PATH_MTU) roce_set_field(context->byte_24_mtu_tc, V2_QPC_BYTE_24_MTU_M, V2_QPC_BYTE_24_MTU_S, attr->path_mtu); @@ -2922,11 +2932,9 @@ static int modify_qp_rtr_to_rts(struct ib_qp *ibqp, return -EINVAL; } - /* If exist optional param, return error */ - if ((attr_mask & IB_QP_ALT_PATH) || (attr_mask & IB_QP_ACCESS_FLAGS) || - (attr_mask & IB_QP_QKEY) || (attr_mask & IB_QP_PATH_MIG_STATE) || - (attr_mask & IB_QP_CUR_STATE) || - (attr_mask & IB_QP_MIN_RNR_TIMER)) { + /* Not support alternate path and path migration */ + if ((attr_mask & IB_QP_ALT_PATH) || + (attr_mask & IB_QP_PATH_MIG_STATE)) { dev_err(dev, "RTR2RTS attr_mask (0x%x)error\n", attr_mask); return -EINVAL; } @@ -3161,7 +3169,8 @@ static int hns_roce_v2_modify_qp(struct ib_qp *ibqp, (cur_state == IB_QPS_RTR && new_state == IB_QPS_ERR) || (cur_state == IB_QPS_RTS && new_state == IB_QPS_ERR) || (cur_state == IB_QPS_SQD && new_state == IB_QPS_ERR) || - (cur_state == IB_QPS_SQE && new_state == IB_QPS_ERR)) { + (cur_state == IB_QPS_SQE && new_state == IB_QPS_ERR) || + (cur_state == IB_QPS_ERR && new_state == IB_QPS_ERR)) { /* Nothing */ ; } else { @@ -4478,7 +4487,7 @@ static int hns_roce_v2_create_eq(struct hns_roce_dev *hr_dev, ret = hns_roce_cmd_mbox(hr_dev, mailbox->dma, 0, eq->eqn, 0, eq_cmd, HNS_ROCE_CMD_TIMEOUT_MSECS); if (ret) { - dev_err(dev, "[mailbox cmd] creat eqc failed.\n"); + dev_err(dev, "[mailbox cmd] create eqc failed.\n"); goto err_cmd_mbox; } diff --git a/drivers/infiniband/hw/hns/hns_roce_qp.c b/drivers/infiniband/hw/hns/hns_roce_qp.c index e289a924e789..d4aad34c21e2 100644 --- a/drivers/infiniband/hw/hns/hns_roce_qp.c +++ b/drivers/infiniband/hw/hns/hns_roce_qp.c @@ -620,7 +620,7 @@ static int hns_roce_create_qp_common(struct hns_roce_dev *hr_dev, to_hr_ucontext(ib_pd->uobject->context), ucmd.db_addr, &hr_qp->rdb); if (ret) { - dev_err(dev, "rp record doorbell map failed!\n"); + dev_err(dev, "rq record doorbell map failed!\n"); goto err_mtt; } } diff --git a/drivers/infiniband/hw/mlx4/mr.c b/drivers/infiniband/hw/mlx4/mr.c index 17f4f151a97f..61d8b06375bb 100644 --- a/drivers/infiniband/hw/mlx4/mr.c +++ b/drivers/infiniband/hw/mlx4/mr.c @@ -346,7 +346,7 @@ int mlx4_ib_umem_calc_optimal_mtt_size(struct ib_umem *umem, u64 start_va, /* Add to the first block the misalignment that it suffers from. */ total_len += (first_block_start & ((1ULL << block_shift) - 1ULL)); last_block_end = current_block_start + current_block_len; - last_block_aligned_end = round_up(last_block_end, 1 << block_shift); + last_block_aligned_end = round_up(last_block_end, 1ULL << block_shift); total_len += (last_block_aligned_end - last_block_end); if (total_len & ((1ULL << block_shift) - 1ULL)) diff --git a/drivers/infiniband/hw/mlx4/qp.c b/drivers/infiniband/hw/mlx4/qp.c index 50af8915e7ec..199648adac74 100644 --- a/drivers/infiniband/hw/mlx4/qp.c +++ b/drivers/infiniband/hw/mlx4/qp.c @@ -673,7 +673,8 @@ static int set_qp_rss(struct mlx4_ib_dev *dev, struct mlx4_ib_rss *rss_ctx, MLX4_IB_RX_HASH_SRC_PORT_TCP | MLX4_IB_RX_HASH_DST_PORT_TCP | MLX4_IB_RX_HASH_SRC_PORT_UDP | - MLX4_IB_RX_HASH_DST_PORT_UDP)) { + MLX4_IB_RX_HASH_DST_PORT_UDP | + MLX4_IB_RX_HASH_INNER)) { pr_debug("RX Hash fields_mask has unsupported mask (0x%llx)\n", ucmd->rx_hash_fields_mask); return (-EOPNOTSUPP); diff --git a/drivers/infiniband/hw/mlx5/Kconfig b/drivers/infiniband/hw/mlx5/Kconfig index bce263b92821..fb4d77be019b 100644 --- a/drivers/infiniband/hw/mlx5/Kconfig +++ b/drivers/infiniband/hw/mlx5/Kconfig @@ -1,6 +1,7 @@ config MLX5_INFINIBAND tristate "Mellanox Connect-IB HCA support" depends on NETDEVICES && ETHERNET && PCI && MLX5_CORE + depends on INFINIBAND_USER_ACCESS || INFINIBAND_USER_ACCESS=n ---help--- This driver provides low-level InfiniBand support for Mellanox Connect-IB PCI Express host channel adapters (HCAs). diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c index daa919e5a442..b4d8ff8ab807 100644 --- a/drivers/infiniband/hw/mlx5/main.c +++ b/drivers/infiniband/hw/mlx5/main.c @@ -52,7 +52,6 @@ #include <linux/mlx5/port.h> #include <linux/mlx5/vport.h> #include <linux/mlx5/fs.h> -#include <linux/mlx5/fs_helpers.h> #include <linux/list.h> #include <rdma/ib_smi.h> #include <rdma/ib_umem.h> @@ -180,7 +179,7 @@ static int mlx5_netdev_event(struct notifier_block *this, if (rep_ndev == ndev) roce->netdev = (event == NETDEV_UNREGISTER) ? NULL : ndev; - } else if (ndev->dev.parent == &ibdev->mdev->pdev->dev) { + } else if (ndev->dev.parent == &mdev->pdev->dev) { roce->netdev = (event == NETDEV_UNREGISTER) ? NULL : ndev; } @@ -4757,7 +4756,7 @@ mlx5_ib_get_vector_affinity(struct ib_device *ibdev, int comp_vector) { struct mlx5_ib_dev *dev = to_mdev(ibdev); - return mlx5_get_vector_affinity(dev->mdev, comp_vector); + return mlx5_get_vector_affinity_hint(dev->mdev, comp_vector); } /* The mlx5_ib_multiport_mutex should be held when calling this function */ @@ -5427,9 +5426,7 @@ static void mlx5_ib_stage_cong_debugfs_cleanup(struct mlx5_ib_dev *dev) static int mlx5_ib_stage_uar_init(struct mlx5_ib_dev *dev) { dev->mdev->priv.uar = mlx5_get_uars_page(dev->mdev); - if (!dev->mdev->priv.uar) - return -ENOMEM; - return 0; + return PTR_ERR_OR_ZERO(dev->mdev->priv.uar); } static void mlx5_ib_stage_uar_cleanup(struct mlx5_ib_dev *dev) diff --git a/drivers/infiniband/hw/mlx5/mr.c b/drivers/infiniband/hw/mlx5/mr.c index 1520a2f20f98..90a9c461cedc 100644 --- a/drivers/infiniband/hw/mlx5/mr.c +++ b/drivers/infiniband/hw/mlx5/mr.c @@ -866,25 +866,28 @@ static int mr_umem_get(struct ib_pd *pd, u64 start, u64 length, int *order) { struct mlx5_ib_dev *dev = to_mdev(pd->device); + struct ib_umem *u; int err; - *umem = ib_umem_get(pd->uobject->context, start, length, - access_flags, 0); - err = PTR_ERR_OR_ZERO(*umem); + *umem = NULL; + + u = ib_umem_get(pd->uobject->context, start, length, access_flags, 0); + err = PTR_ERR_OR_ZERO(u); if (err) { - *umem = NULL; - mlx5_ib_err(dev, "umem get failed (%d)\n", err); + mlx5_ib_dbg(dev, "umem get failed (%d)\n", err); return err; } - mlx5_ib_cont_pages(*umem, start, MLX5_MKEY_PAGE_SHIFT_MASK, npages, + mlx5_ib_cont_pages(u, start, MLX5_MKEY_PAGE_SHIFT_MASK, npages, page_shift, ncont, order); if (!*npages) { mlx5_ib_warn(dev, "avoid zero region\n"); - ib_umem_release(*umem); + ib_umem_release(u); return -EINVAL; } + *umem = u; + mlx5_ib_dbg(dev, "npages %d, ncont %d, order %d, page_shift %d\n", *npages, *ncont, *order, *page_shift); @@ -1458,13 +1461,12 @@ int mlx5_ib_rereg_user_mr(struct ib_mr *ib_mr, int flags, u64 start, int access_flags = flags & IB_MR_REREG_ACCESS ? new_access_flags : mr->access_flags; - u64 addr = (flags & IB_MR_REREG_TRANS) ? virt_addr : mr->umem->address; - u64 len = (flags & IB_MR_REREG_TRANS) ? length : mr->umem->length; int page_shift = 0; int upd_flags = 0; int npages = 0; int ncont = 0; int order = 0; + u64 addr, len; int err; mlx5_ib_dbg(dev, "start 0x%llx, virt_addr 0x%llx, length 0x%llx, access_flags 0x%x\n", @@ -1472,6 +1474,17 @@ int mlx5_ib_rereg_user_mr(struct ib_mr *ib_mr, int flags, u64 start, atomic_sub(mr->npages, &dev->mdev->priv.reg_pages); + if (!mr->umem) + return -EINVAL; + + if (flags & IB_MR_REREG_TRANS) { + addr = virt_addr; + len = length; + } else { + addr = mr->umem->address; + len = mr->umem->length; + } + if (flags != IB_MR_REREG_PD) { /* * Replace umem. This needs to be done whether or not UMR is @@ -1479,6 +1492,7 @@ int mlx5_ib_rereg_user_mr(struct ib_mr *ib_mr, int flags, u64 start, */ flags |= IB_MR_REREG_TRANS; ib_umem_release(mr->umem); + mr->umem = NULL; err = mr_umem_get(pd, addr, len, access_flags, &mr->umem, &npages, &page_shift, &ncont, &order); if (err) diff --git a/drivers/infiniband/hw/mlx5/qp.c b/drivers/infiniband/hw/mlx5/qp.c index 7ed4b70f6447..87b7c1be2a11 100644 --- a/drivers/infiniband/hw/mlx5/qp.c +++ b/drivers/infiniband/hw/mlx5/qp.c @@ -259,7 +259,11 @@ static int set_rq_size(struct mlx5_ib_dev *dev, struct ib_qp_cap *cap, } else { if (ucmd) { qp->rq.wqe_cnt = ucmd->rq_wqe_count; + if (ucmd->rq_wqe_shift > BITS_PER_BYTE * sizeof(ucmd->rq_wqe_shift)) + return -EINVAL; qp->rq.wqe_shift = ucmd->rq_wqe_shift; + if ((1 << qp->rq.wqe_shift) / sizeof(struct mlx5_wqe_data_seg) < qp->wq_sig) + return -EINVAL; qp->rq.max_gs = (1 << qp->rq.wqe_shift) / sizeof(struct mlx5_wqe_data_seg) - qp->wq_sig; qp->rq.max_post = qp->rq.wqe_cnt; } else { @@ -2451,18 +2455,18 @@ enum { static int ib_rate_to_mlx5(struct mlx5_ib_dev *dev, u8 rate) { - if (rate == IB_RATE_PORT_CURRENT) { + if (rate == IB_RATE_PORT_CURRENT) return 0; - } else if (rate < IB_RATE_2_5_GBPS || rate > IB_RATE_300_GBPS) { + + if (rate < IB_RATE_2_5_GBPS || rate > IB_RATE_300_GBPS) return -EINVAL; - } else { - while (rate != IB_RATE_2_5_GBPS && - !(1 << (rate + MLX5_STAT_RATE_OFFSET) & - MLX5_CAP_GEN(dev->mdev, stat_rate_support))) - --rate; - } - return rate + MLX5_STAT_RATE_OFFSET; + while (rate != IB_RATE_PORT_CURRENT && + !(1 << (rate + MLX5_STAT_RATE_OFFSET) & + MLX5_CAP_GEN(dev->mdev, stat_rate_support))) + --rate; + + return rate ? rate + MLX5_STAT_RATE_OFFSET : rate; } static int modify_raw_packet_eth_prio(struct mlx5_core_dev *dev, diff --git a/drivers/infiniband/hw/nes/nes_nic.c b/drivers/infiniband/hw/nes/nes_nic.c index 0a75164cedea..007d5e8a0121 100644 --- a/drivers/infiniband/hw/nes/nes_nic.c +++ b/drivers/infiniband/hw/nes/nes_nic.c @@ -461,7 +461,7 @@ static bool nes_nic_send(struct sk_buff *skb, struct net_device *netdev) /** * nes_netdev_start_xmit */ -static int nes_netdev_start_xmit(struct sk_buff *skb, struct net_device *netdev) +static netdev_tx_t nes_netdev_start_xmit(struct sk_buff *skb, struct net_device *netdev) { struct nes_vnic *nesvnic = netdev_priv(netdev); struct nes_device *nesdev = nesvnic->nesdev; diff --git a/drivers/infiniband/sw/rxe/rxe_opcode.c b/drivers/infiniband/sw/rxe/rxe_opcode.c index 61927c165b59..4cf11063e0b5 100644 --- a/drivers/infiniband/sw/rxe/rxe_opcode.c +++ b/drivers/infiniband/sw/rxe/rxe_opcode.c @@ -390,7 +390,7 @@ struct rxe_opcode_info rxe_opcode[RXE_NUM_OPCODE] = { .name = "IB_OPCODE_RC_SEND_ONLY_INV", .mask = RXE_IETH_MASK | RXE_PAYLOAD_MASK | RXE_REQ_MASK | RXE_COMP_MASK | RXE_RWR_MASK | RXE_SEND_MASK - | RXE_END_MASK, + | RXE_END_MASK | RXE_START_MASK, .length = RXE_BTH_BYTES + RXE_IETH_BYTES, .offset = { [RXE_BTH] = 0, diff --git a/drivers/infiniband/sw/rxe/rxe_req.c b/drivers/infiniband/sw/rxe/rxe_req.c index 7bdaf71b8221..785199990457 100644 --- a/drivers/infiniband/sw/rxe/rxe_req.c +++ b/drivers/infiniband/sw/rxe/rxe_req.c @@ -728,7 +728,6 @@ next_wqe: rollback_state(wqe, qp, &rollback_wqe, rollback_psn); if (ret == -EAGAIN) { - kfree_skb(skb); rxe_run_task(&qp->req.task, 1); goto exit; } diff --git a/drivers/infiniband/sw/rxe/rxe_resp.c b/drivers/infiniband/sw/rxe/rxe_resp.c index a65c9969f7fc..955ff3b6da9c 100644 --- a/drivers/infiniband/sw/rxe/rxe_resp.c +++ b/drivers/infiniband/sw/rxe/rxe_resp.c @@ -742,7 +742,6 @@ static enum resp_states read_reply(struct rxe_qp *qp, err = rxe_xmit_packet(rxe, qp, &ack_pkt, skb); if (err) { pr_err("Failed sending RDMA reply.\n"); - kfree_skb(skb); return RESPST_ERR_RNR; } @@ -954,10 +953,8 @@ static int send_ack(struct rxe_qp *qp, struct rxe_pkt_info *pkt, } err = rxe_xmit_packet(rxe, qp, &ack_pkt, skb); - if (err) { + if (err) pr_err_ratelimited("Failed sending ack\n"); - kfree_skb(skb); - } err1: return err; @@ -1141,7 +1138,6 @@ static enum resp_states duplicate_request(struct rxe_qp *qp, if (rc) { pr_err("Failed resending result. This flow is not handled - skb ignored\n"); rxe_drop_ref(qp); - kfree_skb(skb_copy); rc = RESPST_CLEANUP; goto out; } diff --git a/drivers/infiniband/ulp/ipoib/ipoib_main.c b/drivers/infiniband/ulp/ipoib/ipoib_main.c index 161ba8c76285..cf291f90b58f 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_main.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_main.c @@ -1094,7 +1094,7 @@ drop_and_unlock: spin_unlock_irqrestore(&priv->lock, flags); } -static int ipoib_start_xmit(struct sk_buff *skb, struct net_device *dev) +static netdev_tx_t ipoib_start_xmit(struct sk_buff *skb, struct net_device *dev) { struct ipoib_dev_priv *priv = ipoib_priv(dev); struct rdma_netdev *rn = netdev_priv(dev); diff --git a/drivers/infiniband/ulp/srp/Kconfig b/drivers/infiniband/ulp/srp/Kconfig index c74ee9633041..99db8fe5173a 100644 --- a/drivers/infiniband/ulp/srp/Kconfig +++ b/drivers/infiniband/ulp/srp/Kconfig @@ -1,6 +1,6 @@ config INFINIBAND_SRP tristate "InfiniBand SCSI RDMA Protocol" - depends on SCSI + depends on SCSI && INFINIBAND_ADDR_TRANS select SCSI_SRP_ATTRS ---help--- Support for the SCSI RDMA Protocol over InfiniBand. This diff --git a/drivers/infiniband/ulp/srpt/Kconfig b/drivers/infiniband/ulp/srpt/Kconfig index 31ee83d528d9..fb8b7182f05e 100644 --- a/drivers/infiniband/ulp/srpt/Kconfig +++ b/drivers/infiniband/ulp/srpt/Kconfig @@ -1,6 +1,6 @@ config INFINIBAND_SRPT tristate "InfiniBand SCSI RDMA Protocol target support" - depends on INFINIBAND && TARGET_CORE + depends on INFINIBAND && INFINIBAND_ADDR_TRANS && TARGET_CORE ---help--- Support for the SCSI RDMA Protocol (SRP) Target driver. The diff --git a/drivers/input/input-leds.c b/drivers/input/input-leds.c index 766bf2660116..5f04b2d94635 100644 --- a/drivers/input/input-leds.c +++ b/drivers/input/input-leds.c @@ -88,6 +88,7 @@ static int input_leds_connect(struct input_handler *handler, const struct input_device_id *id) { struct input_leds *leds; + struct input_led *led; unsigned int num_leds; unsigned int led_code; int led_no; @@ -119,14 +120,13 @@ static int input_leds_connect(struct input_handler *handler, led_no = 0; for_each_set_bit(led_code, dev->ledbit, LED_CNT) { - struct input_led *led = &leds->leds[led_no]; + if (!input_led_info[led_code].name) + continue; + led = &leds->leds[led_no]; led->handle = &leds->handle; led->code = led_code; - if (!input_led_info[led_code].name) - continue; - led->cdev.name = kasprintf(GFP_KERNEL, "%s::%s", dev_name(&dev->dev), input_led_info[led_code].name); diff --git a/drivers/input/mouse/alps.c b/drivers/input/mouse/alps.c index 0a67f235ba88..38f9501acdf0 100644 --- a/drivers/input/mouse/alps.c +++ b/drivers/input/mouse/alps.c @@ -583,7 +583,7 @@ static void alps_process_trackstick_packet_v3(struct psmouse *psmouse) x = (s8)(((packet[0] & 0x20) << 2) | (packet[1] & 0x7f)); y = (s8)(((packet[0] & 0x10) << 3) | (packet[2] & 0x7f)); - z = packet[4] & 0x7c; + z = packet[4] & 0x7f; /* * The x and y values tend to be quite large, and when used diff --git a/drivers/input/rmi4/rmi_spi.c b/drivers/input/rmi4/rmi_spi.c index 76edbf2c1bce..082defc329a8 100644 --- a/drivers/input/rmi4/rmi_spi.c +++ b/drivers/input/rmi4/rmi_spi.c @@ -147,8 +147,11 @@ static int rmi_spi_xfer(struct rmi_spi_xport *rmi_spi, if (len > RMI_SPI_XFER_SIZE_LIMIT) return -EINVAL; - if (rmi_spi->xfer_buf_size < len) - rmi_spi_manage_pools(rmi_spi, len); + if (rmi_spi->xfer_buf_size < len) { + ret = rmi_spi_manage_pools(rmi_spi, len); + if (ret < 0) + return ret; + } if (addr == 0) /* diff --git a/drivers/input/touchscreen/Kconfig b/drivers/input/touchscreen/Kconfig index 4f15496fec8b..3e613afa10b4 100644 --- a/drivers/input/touchscreen/Kconfig +++ b/drivers/input/touchscreen/Kconfig @@ -362,7 +362,7 @@ config TOUCHSCREEN_HIDEEP If unsure, say N. - To compile this driver as a moudle, choose M here : the + To compile this driver as a module, choose M here : the module will be called hideep_ts. config TOUCHSCREEN_ILI210X diff --git a/drivers/input/touchscreen/atmel_mxt_ts.c b/drivers/input/touchscreen/atmel_mxt_ts.c index 5d9699fe1b55..09194721aed2 100644 --- a/drivers/input/touchscreen/atmel_mxt_ts.c +++ b/drivers/input/touchscreen/atmel_mxt_ts.c @@ -280,7 +280,8 @@ struct mxt_data { struct input_dev *input_dev; char phys[64]; /* device physical location */ struct mxt_object *object_table; - struct mxt_info info; + struct mxt_info *info; + void *raw_info_block; unsigned int irq; unsigned int max_x; unsigned int max_y; @@ -460,12 +461,13 @@ static int mxt_lookup_bootloader_address(struct mxt_data *data, bool retry) { u8 appmode = data->client->addr; u8 bootloader; + u8 family_id = data->info ? data->info->family_id : 0; switch (appmode) { case 0x4a: case 0x4b: /* Chips after 1664S use different scheme */ - if (retry || data->info.family_id >= 0xa2) { + if (retry || family_id >= 0xa2) { bootloader = appmode - 0x24; break; } @@ -692,7 +694,7 @@ mxt_get_object(struct mxt_data *data, u8 type) struct mxt_object *object; int i; - for (i = 0; i < data->info.object_num; i++) { + for (i = 0; i < data->info->object_num; i++) { object = data->object_table + i; if (object->type == type) return object; @@ -1462,12 +1464,12 @@ static int mxt_update_cfg(struct mxt_data *data, const struct firmware *cfg) data_pos += offset; } - if (cfg_info.family_id != data->info.family_id) { + if (cfg_info.family_id != data->info->family_id) { dev_err(dev, "Family ID mismatch!\n"); return -EINVAL; } - if (cfg_info.variant_id != data->info.variant_id) { + if (cfg_info.variant_id != data->info->variant_id) { dev_err(dev, "Variant ID mismatch!\n"); return -EINVAL; } @@ -1512,7 +1514,7 @@ static int mxt_update_cfg(struct mxt_data *data, const struct firmware *cfg) /* Malloc memory to store configuration */ cfg_start_ofs = MXT_OBJECT_START + - data->info.object_num * sizeof(struct mxt_object) + + data->info->object_num * sizeof(struct mxt_object) + MXT_INFO_CHECKSUM_SIZE; config_mem_size = data->mem_size - cfg_start_ofs; config_mem = kzalloc(config_mem_size, GFP_KERNEL); @@ -1563,20 +1565,6 @@ release_mem: return ret; } -static int mxt_get_info(struct mxt_data *data) -{ - struct i2c_client *client = data->client; - struct mxt_info *info = &data->info; - int error; - - /* Read 7-byte info block starting at address 0 */ - error = __mxt_read_reg(client, 0, sizeof(*info), info); - if (error) - return error; - - return 0; -} - static void mxt_free_input_device(struct mxt_data *data) { if (data->input_dev) { @@ -1591,9 +1579,10 @@ static void mxt_free_object_table(struct mxt_data *data) video_unregister_device(&data->dbg.vdev); v4l2_device_unregister(&data->dbg.v4l2); #endif - - kfree(data->object_table); data->object_table = NULL; + data->info = NULL; + kfree(data->raw_info_block); + data->raw_info_block = NULL; kfree(data->msg_buf); data->msg_buf = NULL; data->T5_address = 0; @@ -1609,34 +1598,18 @@ static void mxt_free_object_table(struct mxt_data *data) data->max_reportid = 0; } -static int mxt_get_object_table(struct mxt_data *data) +static int mxt_parse_object_table(struct mxt_data *data, + struct mxt_object *object_table) { struct i2c_client *client = data->client; - size_t table_size; - struct mxt_object *object_table; - int error; int i; u8 reportid; u16 end_address; - table_size = data->info.object_num * sizeof(struct mxt_object); - object_table = kzalloc(table_size, GFP_KERNEL); - if (!object_table) { - dev_err(&data->client->dev, "Failed to allocate memory\n"); - return -ENOMEM; - } - - error = __mxt_read_reg(client, MXT_OBJECT_START, table_size, - object_table); - if (error) { - kfree(object_table); - return error; - } - /* Valid Report IDs start counting from 1 */ reportid = 1; data->mem_size = 0; - for (i = 0; i < data->info.object_num; i++) { + for (i = 0; i < data->info->object_num; i++) { struct mxt_object *object = object_table + i; u8 min_id, max_id; @@ -1660,8 +1633,8 @@ static int mxt_get_object_table(struct mxt_data *data) switch (object->type) { case MXT_GEN_MESSAGE_T5: - if (data->info.family_id == 0x80 && - data->info.version < 0x20) { + if (data->info->family_id == 0x80 && + data->info->version < 0x20) { /* * On mXT224 firmware versions prior to V2.0 * read and discard unused CRC byte otherwise @@ -1716,24 +1689,102 @@ static int mxt_get_object_table(struct mxt_data *data) /* If T44 exists, T5 position has to be directly after */ if (data->T44_address && (data->T5_address != data->T44_address + 1)) { dev_err(&client->dev, "Invalid T44 position\n"); - error = -EINVAL; - goto free_object_table; + return -EINVAL; } data->msg_buf = kcalloc(data->max_reportid, data->T5_msg_size, GFP_KERNEL); - if (!data->msg_buf) { - dev_err(&client->dev, "Failed to allocate message buffer\n"); + if (!data->msg_buf) + return -ENOMEM; + + return 0; +} + +static int mxt_read_info_block(struct mxt_data *data) +{ + struct i2c_client *client = data->client; + int error; + size_t size; + void *id_buf, *buf; + uint8_t num_objects; + u32 calculated_crc; + u8 *crc_ptr; + + /* If info block already allocated, free it */ + if (data->raw_info_block) + mxt_free_object_table(data); + + /* Read 7-byte ID information block starting at address 0 */ + size = sizeof(struct mxt_info); + id_buf = kzalloc(size, GFP_KERNEL); + if (!id_buf) + return -ENOMEM; + + error = __mxt_read_reg(client, 0, size, id_buf); + if (error) + goto err_free_mem; + + /* Resize buffer to give space for rest of info block */ + num_objects = ((struct mxt_info *)id_buf)->object_num; + size += (num_objects * sizeof(struct mxt_object)) + + MXT_INFO_CHECKSUM_SIZE; + + buf = krealloc(id_buf, size, GFP_KERNEL); + if (!buf) { error = -ENOMEM; - goto free_object_table; + goto err_free_mem; + } + id_buf = buf; + + /* Read rest of info block */ + error = __mxt_read_reg(client, MXT_OBJECT_START, + size - MXT_OBJECT_START, + id_buf + MXT_OBJECT_START); + if (error) + goto err_free_mem; + + /* Extract & calculate checksum */ + crc_ptr = id_buf + size - MXT_INFO_CHECKSUM_SIZE; + data->info_crc = crc_ptr[0] | (crc_ptr[1] << 8) | (crc_ptr[2] << 16); + + calculated_crc = mxt_calculate_crc(id_buf, 0, + size - MXT_INFO_CHECKSUM_SIZE); + + /* + * CRC mismatch can be caused by data corruption due to I2C comms + * issue or else device is not using Object Based Protocol (eg i2c-hid) + */ + if ((data->info_crc == 0) || (data->info_crc != calculated_crc)) { + dev_err(&client->dev, + "Info Block CRC error calculated=0x%06X read=0x%06X\n", + calculated_crc, data->info_crc); + error = -EIO; + goto err_free_mem; + } + + data->raw_info_block = id_buf; + data->info = (struct mxt_info *)id_buf; + + dev_info(&client->dev, + "Family: %u Variant: %u Firmware V%u.%u.%02X Objects: %u\n", + data->info->family_id, data->info->variant_id, + data->info->version >> 4, data->info->version & 0xf, + data->info->build, data->info->object_num); + + /* Parse object table information */ + error = mxt_parse_object_table(data, id_buf + MXT_OBJECT_START); + if (error) { + dev_err(&client->dev, "Error %d parsing object table\n", error); + mxt_free_object_table(data); + goto err_free_mem; } - data->object_table = object_table; + data->object_table = (struct mxt_object *)(id_buf + MXT_OBJECT_START); return 0; -free_object_table: - mxt_free_object_table(data); +err_free_mem: + kfree(id_buf); return error; } @@ -2046,7 +2097,7 @@ static int mxt_initialize(struct mxt_data *data) int error; while (1) { - error = mxt_get_info(data); + error = mxt_read_info_block(data); if (!error) break; @@ -2077,16 +2128,9 @@ static int mxt_initialize(struct mxt_data *data) msleep(MXT_FW_RESET_TIME); } - /* Get object table information */ - error = mxt_get_object_table(data); - if (error) { - dev_err(&client->dev, "Error %d reading object table\n", error); - return error; - } - error = mxt_acquire_irq(data); if (error) - goto err_free_object_table; + return error; error = request_firmware_nowait(THIS_MODULE, true, MXT_CFG_NAME, &client->dev, GFP_KERNEL, data, @@ -2094,14 +2138,10 @@ static int mxt_initialize(struct mxt_data *data) if (error) { dev_err(&client->dev, "Failed to invoke firmware loader: %d\n", error); - goto err_free_object_table; + return error; } return 0; - -err_free_object_table: - mxt_free_object_table(data); - return error; } static int mxt_set_t7_power_cfg(struct mxt_data *data, u8 sleep) @@ -2162,7 +2202,7 @@ recheck: static u16 mxt_get_debug_value(struct mxt_data *data, unsigned int x, unsigned int y) { - struct mxt_info *info = &data->info; + struct mxt_info *info = data->info; struct mxt_dbg *dbg = &data->dbg; unsigned int ofs, page; unsigned int col = 0; @@ -2490,7 +2530,7 @@ static const struct video_device mxt_video_device = { static void mxt_debug_init(struct mxt_data *data) { - struct mxt_info *info = &data->info; + struct mxt_info *info = data->info; struct mxt_dbg *dbg = &data->dbg; struct mxt_object *object; int error; @@ -2576,7 +2616,6 @@ static int mxt_configure_objects(struct mxt_data *data, const struct firmware *cfg) { struct device *dev = &data->client->dev; - struct mxt_info *info = &data->info; int error; error = mxt_init_t7_power_cfg(data); @@ -2601,11 +2640,6 @@ static int mxt_configure_objects(struct mxt_data *data, mxt_debug_init(data); - dev_info(dev, - "Family: %u Variant: %u Firmware V%u.%u.%02X Objects: %u\n", - info->family_id, info->variant_id, info->version >> 4, - info->version & 0xf, info->build, info->object_num); - return 0; } @@ -2614,7 +2648,7 @@ static ssize_t mxt_fw_version_show(struct device *dev, struct device_attribute *attr, char *buf) { struct mxt_data *data = dev_get_drvdata(dev); - struct mxt_info *info = &data->info; + struct mxt_info *info = data->info; return scnprintf(buf, PAGE_SIZE, "%u.%u.%02X\n", info->version >> 4, info->version & 0xf, info->build); } @@ -2624,7 +2658,7 @@ static ssize_t mxt_hw_version_show(struct device *dev, struct device_attribute *attr, char *buf) { struct mxt_data *data = dev_get_drvdata(dev); - struct mxt_info *info = &data->info; + struct mxt_info *info = data->info; return scnprintf(buf, PAGE_SIZE, "%u.%u\n", info->family_id, info->variant_id); } @@ -2663,7 +2697,7 @@ static ssize_t mxt_object_show(struct device *dev, return -ENOMEM; error = 0; - for (i = 0; i < data->info.object_num; i++) { + for (i = 0; i < data->info->object_num; i++) { object = data->object_table + i; if (!mxt_object_readable(object->type)) @@ -3035,6 +3069,15 @@ static const struct dmi_system_id mxt_dmi_table[] = { .driver_data = samus_platform_data, }, { + /* Samsung Chromebook Pro */ + .ident = "Samsung Chromebook Pro", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Google"), + DMI_MATCH(DMI_PRODUCT_NAME, "Caroline"), + }, + .driver_data = samus_platform_data, + }, + { /* Other Google Chromebooks */ .ident = "Chromebook", .matches = { @@ -3254,6 +3297,11 @@ static SIMPLE_DEV_PM_OPS(mxt_pm_ops, mxt_suspend, mxt_resume); static const struct of_device_id mxt_of_match[] = { { .compatible = "atmel,maxtouch", }, + /* Compatibles listed below are deprecated */ + { .compatible = "atmel,qt602240_ts", }, + { .compatible = "atmel,atmel_mxt_ts", }, + { .compatible = "atmel,atmel_mxt_tp", }, + { .compatible = "atmel,mXT224", }, {}, }; MODULE_DEVICE_TABLE(of, mxt_of_match); diff --git a/drivers/iommu/amd_iommu.c b/drivers/iommu/amd_iommu.c index 2a99f0f14795..8fb8c737fffe 100644 --- a/drivers/iommu/amd_iommu.c +++ b/drivers/iommu/amd_iommu.c @@ -83,7 +83,6 @@ static DEFINE_SPINLOCK(amd_iommu_devtable_lock); static DEFINE_SPINLOCK(pd_bitmap_lock); -static DEFINE_SPINLOCK(iommu_table_lock); /* List of all available dev_data structures */ static LLIST_HEAD(dev_data_list); @@ -3562,6 +3561,7 @@ EXPORT_SYMBOL(amd_iommu_device_info); *****************************************************************************/ static struct irq_chip amd_ir_chip; +static DEFINE_SPINLOCK(iommu_table_lock); static void set_dte_irq_entry(u16 devid, struct irq_remap_table *table) { diff --git a/drivers/iommu/dma-iommu.c b/drivers/iommu/dma-iommu.c index f05f3cf90756..ddcbbdb5d658 100644 --- a/drivers/iommu/dma-iommu.c +++ b/drivers/iommu/dma-iommu.c @@ -167,40 +167,16 @@ EXPORT_SYMBOL(iommu_put_dma_cookie); * @list: Reserved region list from iommu_get_resv_regions() * * IOMMU drivers can use this to implement their .get_resv_regions callback - * for general non-IOMMU-specific reservations. Currently, this covers host - * bridge windows for PCI devices and GICv3 ITS region reservation on ACPI - * based ARM platforms that may require HW MSI reservation. + * for general non-IOMMU-specific reservations. Currently, this covers GICv3 + * ITS region reservation on ACPI based ARM platforms that may require HW MSI + * reservation. */ void iommu_dma_get_resv_regions(struct device *dev, struct list_head *list) { - struct pci_host_bridge *bridge; - struct resource_entry *window; - - if (!is_of_node(dev->iommu_fwspec->iommu_fwnode) && - iort_iommu_msi_get_resv_regions(dev, list) < 0) - return; - - if (!dev_is_pci(dev)) - return; - - bridge = pci_find_host_bridge(to_pci_dev(dev)->bus); - resource_list_for_each_entry(window, &bridge->windows) { - struct iommu_resv_region *region; - phys_addr_t start; - size_t length; - - if (resource_type(window->res) != IORESOURCE_MEM) - continue; - start = window->res->start - window->offset; - length = window->res->end - window->res->start + 1; - region = iommu_alloc_resv_region(start, length, 0, - IOMMU_RESV_RESERVED); - if (!region) - return; + if (!is_of_node(dev->iommu_fwspec->iommu_fwnode)) + iort_iommu_msi_get_resv_regions(dev, list); - list_add_tail(®ion->list, list); - } } EXPORT_SYMBOL(iommu_dma_get_resv_regions); @@ -229,6 +205,23 @@ static int cookie_init_hw_msi_region(struct iommu_dma_cookie *cookie, return 0; } +static void iova_reserve_pci_windows(struct pci_dev *dev, + struct iova_domain *iovad) +{ + struct pci_host_bridge *bridge = pci_find_host_bridge(dev->bus); + struct resource_entry *window; + unsigned long lo, hi; + + resource_list_for_each_entry(window, &bridge->windows) { + if (resource_type(window->res) != IORESOURCE_MEM) + continue; + + lo = iova_pfn(iovad, window->res->start - window->offset); + hi = iova_pfn(iovad, window->res->end - window->offset); + reserve_iova(iovad, lo, hi); + } +} + static int iova_reserve_iommu_regions(struct device *dev, struct iommu_domain *domain) { @@ -238,6 +231,9 @@ static int iova_reserve_iommu_regions(struct device *dev, LIST_HEAD(resv_regions); int ret = 0; + if (dev_is_pci(dev)) + iova_reserve_pci_windows(to_pci_dev(dev), iovad); + iommu_get_resv_regions(dev, &resv_regions); list_for_each_entry(region, &resv_regions, list) { unsigned long lo, hi; diff --git a/drivers/iommu/dmar.c b/drivers/iommu/dmar.c index accf58388bdb..460bed4fc5b1 100644 --- a/drivers/iommu/dmar.c +++ b/drivers/iommu/dmar.c @@ -1345,7 +1345,7 @@ void qi_flush_dev_iotlb(struct intel_iommu *iommu, u16 sid, u16 qdep, struct qi_desc desc; if (mask) { - BUG_ON(addr & ((1 << (VTD_PAGE_SHIFT + mask)) - 1)); + WARN_ON_ONCE(addr & ((1ULL << (VTD_PAGE_SHIFT + mask)) - 1)); addr |= (1ULL << (VTD_PAGE_SHIFT + mask - 1)) - 1; desc.high = QI_DEV_IOTLB_ADDR(addr) | QI_DEV_IOTLB_SIZE; } else diff --git a/drivers/iommu/intel_irq_remapping.c b/drivers/iommu/intel_irq_remapping.c index 66f69af2c219..3062a154a9fb 100644 --- a/drivers/iommu/intel_irq_remapping.c +++ b/drivers/iommu/intel_irq_remapping.c @@ -1136,7 +1136,7 @@ static void intel_ir_reconfigure_irte(struct irq_data *irqd, bool force) irte->dest_id = IRTE_DEST(cfg->dest_apicid); /* Update the hardware only if the interrupt is in remapped mode. */ - if (!force || ir_data->irq_2_iommu.mode == IRQ_REMAPPING) + if (force || ir_data->irq_2_iommu.mode == IRQ_REMAPPING) modify_irte(&ir_data->irq_2_iommu, irte); } diff --git a/drivers/iommu/rockchip-iommu.c b/drivers/iommu/rockchip-iommu.c index 5fc8656c60f9..0468acfa131f 100644 --- a/drivers/iommu/rockchip-iommu.c +++ b/drivers/iommu/rockchip-iommu.c @@ -1098,7 +1098,7 @@ static int rk_iommu_of_xlate(struct device *dev, data->iommu = platform_get_drvdata(iommu_dev); dev->archdata.iommu = data; - of_dev_put(iommu_dev); + platform_device_put(iommu_dev); return 0; } @@ -1175,8 +1175,15 @@ static int rk_iommu_probe(struct platform_device *pdev) for (i = 0; i < iommu->num_clocks; ++i) iommu->clocks[i].id = rk_iommu_clocks[i]; + /* + * iommu clocks should be present for all new devices and devicetrees + * but there are older devicetrees without clocks out in the wild. + * So clocks as optional for the time being. + */ err = devm_clk_bulk_get(iommu->dev, iommu->num_clocks, iommu->clocks); - if (err) + if (err == -ENOENT) + iommu->num_clocks = 0; + else if (err) return err; err = clk_bulk_prepare(iommu->num_clocks, iommu->clocks); diff --git a/drivers/irqchip/qcom-irq-combiner.c b/drivers/irqchip/qcom-irq-combiner.c index f31265937439..7f0c0be322e0 100644 --- a/drivers/irqchip/qcom-irq-combiner.c +++ b/drivers/irqchip/qcom-irq-combiner.c @@ -1,4 +1,4 @@ -/* Copyright (c) 2015-2016, The Linux Foundation. All rights reserved. +/* Copyright (c) 2015-2018, The Linux Foundation. All rights reserved. * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License version 2 and @@ -68,7 +68,7 @@ static void combiner_handle_irq(struct irq_desc *desc) bit = readl_relaxed(combiner->regs[reg].addr); status = bit & combiner->regs[reg].enabled; - if (!status) + if (bit && !status) pr_warn_ratelimited("Unexpected IRQ on CPU%d: (%08x %08lx %p)\n", smp_processor_id(), bit, combiner->regs[reg].enabled, diff --git a/drivers/md/bcache/alloc.c b/drivers/md/bcache/alloc.c index 004cc3cc6123..7fa2631b422c 100644 --- a/drivers/md/bcache/alloc.c +++ b/drivers/md/bcache/alloc.c @@ -290,7 +290,7 @@ do { \ if (kthread_should_stop() || \ test_bit(CACHE_SET_IO_DISABLE, &ca->set->flags)) { \ set_current_state(TASK_RUNNING); \ - return 0; \ + goto out; \ } \ \ schedule(); \ @@ -378,6 +378,9 @@ retry_invalidate: bch_prio_write(ca); } } +out: + wait_for_kthread_stop(); + return 0; } /* Allocation */ diff --git a/drivers/md/bcache/bcache.h b/drivers/md/bcache/bcache.h index d338b7086013..3a0cfb237af9 100644 --- a/drivers/md/bcache/bcache.h +++ b/drivers/md/bcache/bcache.h @@ -392,6 +392,8 @@ struct cached_dev { #define DEFAULT_CACHED_DEV_ERROR_LIMIT 64 atomic_t io_errors; unsigned error_limit; + + char backing_dev_name[BDEVNAME_SIZE]; }; enum alloc_reserve { @@ -464,6 +466,8 @@ struct cache { atomic_long_t meta_sectors_written; atomic_long_t btree_sectors_written; atomic_long_t sectors_written; + + char cache_dev_name[BDEVNAME_SIZE]; }; struct gc_stat { diff --git a/drivers/md/bcache/debug.c b/drivers/md/bcache/debug.c index 028f7b386e01..4e63c6f6c04d 100644 --- a/drivers/md/bcache/debug.c +++ b/drivers/md/bcache/debug.c @@ -106,7 +106,6 @@ void bch_btree_verify(struct btree *b) void bch_data_verify(struct cached_dev *dc, struct bio *bio) { - char name[BDEVNAME_SIZE]; struct bio *check; struct bio_vec bv, cbv; struct bvec_iter iter, citer = { 0 }; @@ -134,7 +133,7 @@ void bch_data_verify(struct cached_dev *dc, struct bio *bio) bv.bv_len), dc->disk.c, "verify failed at dev %s sector %llu", - bdevname(dc->bdev, name), + dc->backing_dev_name, (uint64_t) bio->bi_iter.bi_sector); kunmap_atomic(p1); diff --git a/drivers/md/bcache/io.c b/drivers/md/bcache/io.c index 7fac97ae036e..2ddf8515e6a5 100644 --- a/drivers/md/bcache/io.c +++ b/drivers/md/bcache/io.c @@ -52,7 +52,6 @@ void bch_submit_bbio(struct bio *bio, struct cache_set *c, /* IO errors */ void bch_count_backing_io_errors(struct cached_dev *dc, struct bio *bio) { - char buf[BDEVNAME_SIZE]; unsigned errors; WARN_ONCE(!dc, "NULL pointer of struct cached_dev"); @@ -60,7 +59,7 @@ void bch_count_backing_io_errors(struct cached_dev *dc, struct bio *bio) errors = atomic_add_return(1, &dc->io_errors); if (errors < dc->error_limit) pr_err("%s: IO error on backing device, unrecoverable", - bio_devname(bio, buf)); + dc->backing_dev_name); else bch_cached_dev_error(dc); } @@ -105,19 +104,18 @@ void bch_count_io_errors(struct cache *ca, } if (error) { - char buf[BDEVNAME_SIZE]; unsigned errors = atomic_add_return(1 << IO_ERROR_SHIFT, &ca->io_errors); errors >>= IO_ERROR_SHIFT; if (errors < ca->set->error_limit) pr_err("%s: IO error on %s%s", - bdevname(ca->bdev, buf), m, + ca->cache_dev_name, m, is_read ? ", recovering." : "."); else bch_cache_set_error(ca->set, "%s: too many IO errors %s", - bdevname(ca->bdev, buf), m); + ca->cache_dev_name, m); } } diff --git a/drivers/md/bcache/request.c b/drivers/md/bcache/request.c index a65e3365eeb9..8e3e8655ed63 100644 --- a/drivers/md/bcache/request.c +++ b/drivers/md/bcache/request.c @@ -649,11 +649,8 @@ static void backing_request_endio(struct bio *bio) */ if (unlikely(s->iop.writeback && bio->bi_opf & REQ_PREFLUSH)) { - char buf[BDEVNAME_SIZE]; - - bio_devname(bio, buf); pr_err("Can't flush %s: returned bi_status %i", - buf, bio->bi_status); + dc->backing_dev_name, bio->bi_status); } else { /* set to orig_bio->bi_status in bio_complete() */ s->iop.status = bio->bi_status; diff --git a/drivers/md/bcache/super.c b/drivers/md/bcache/super.c index d90d9e59ca00..3dea06b41d43 100644 --- a/drivers/md/bcache/super.c +++ b/drivers/md/bcache/super.c @@ -936,7 +936,6 @@ static void cancel_writeback_rate_update_dwork(struct cached_dev *dc) static void cached_dev_detach_finish(struct work_struct *w) { struct cached_dev *dc = container_of(w, struct cached_dev, detach); - char buf[BDEVNAME_SIZE]; struct closure cl; closure_init_stack(&cl); @@ -967,7 +966,7 @@ static void cached_dev_detach_finish(struct work_struct *w) mutex_unlock(&bch_register_lock); - pr_info("Caching disabled for %s", bdevname(dc->bdev, buf)); + pr_info("Caching disabled for %s", dc->backing_dev_name); /* Drop ref we took in cached_dev_detach() */ closure_put(&dc->disk.cl); @@ -999,29 +998,28 @@ int bch_cached_dev_attach(struct cached_dev *dc, struct cache_set *c, { uint32_t rtime = cpu_to_le32(get_seconds()); struct uuid_entry *u; - char buf[BDEVNAME_SIZE]; struct cached_dev *exist_dc, *t; - bdevname(dc->bdev, buf); - if ((set_uuid && memcmp(set_uuid, c->sb.set_uuid, 16)) || (!set_uuid && memcmp(dc->sb.set_uuid, c->sb.set_uuid, 16))) return -ENOENT; if (dc->disk.c) { - pr_err("Can't attach %s: already attached", buf); + pr_err("Can't attach %s: already attached", + dc->backing_dev_name); return -EINVAL; } if (test_bit(CACHE_SET_STOPPING, &c->flags)) { - pr_err("Can't attach %s: shutting down", buf); + pr_err("Can't attach %s: shutting down", + dc->backing_dev_name); return -EINVAL; } if (dc->sb.block_size < c->sb.block_size) { /* Will die */ pr_err("Couldn't attach %s: block size less than set's block size", - buf); + dc->backing_dev_name); return -EINVAL; } @@ -1029,7 +1027,7 @@ int bch_cached_dev_attach(struct cached_dev *dc, struct cache_set *c, list_for_each_entry_safe(exist_dc, t, &c->cached_devs, list) { if (!memcmp(dc->sb.uuid, exist_dc->sb.uuid, 16)) { pr_err("Tried to attach %s but duplicate UUID already attached", - buf); + dc->backing_dev_name); return -EINVAL; } @@ -1047,13 +1045,15 @@ int bch_cached_dev_attach(struct cached_dev *dc, struct cache_set *c, if (!u) { if (BDEV_STATE(&dc->sb) == BDEV_STATE_DIRTY) { - pr_err("Couldn't find uuid for %s in set", buf); + pr_err("Couldn't find uuid for %s in set", + dc->backing_dev_name); return -ENOENT; } u = uuid_find_empty(c); if (!u) { - pr_err("Not caching %s, no room for UUID", buf); + pr_err("Not caching %s, no room for UUID", + dc->backing_dev_name); return -EINVAL; } } @@ -1112,7 +1112,8 @@ int bch_cached_dev_attach(struct cached_dev *dc, struct cache_set *c, up_write(&dc->writeback_lock); pr_info("Caching %s as %s on set %pU", - bdevname(dc->bdev, buf), dc->disk.disk->disk_name, + dc->backing_dev_name, + dc->disk.disk->disk_name, dc->disk.c->sb.set_uuid); return 0; } @@ -1225,10 +1226,10 @@ static void register_bdev(struct cache_sb *sb, struct page *sb_page, struct block_device *bdev, struct cached_dev *dc) { - char name[BDEVNAME_SIZE]; const char *err = "cannot allocate memory"; struct cache_set *c; + bdevname(bdev, dc->backing_dev_name); memcpy(&dc->sb, sb, sizeof(struct cache_sb)); dc->bdev = bdev; dc->bdev->bd_holder = dc; @@ -1237,6 +1238,7 @@ static void register_bdev(struct cache_sb *sb, struct page *sb_page, bio_first_bvec_all(&dc->sb_bio)->bv_page = sb_page; get_page(sb_page); + if (cached_dev_init(dc, sb->block_size << 9)) goto err; @@ -1247,7 +1249,7 @@ static void register_bdev(struct cache_sb *sb, struct page *sb_page, if (bch_cache_accounting_add_kobjs(&dc->accounting, &dc->disk.kobj)) goto err; - pr_info("registered backing device %s", bdevname(bdev, name)); + pr_info("registered backing device %s", dc->backing_dev_name); list_add(&dc->list, &uncached_devices); list_for_each_entry(c, &bch_cache_sets, list) @@ -1259,7 +1261,7 @@ static void register_bdev(struct cache_sb *sb, struct page *sb_page, return; err: - pr_notice("error %s: %s", bdevname(bdev, name), err); + pr_notice("error %s: %s", dc->backing_dev_name, err); bcache_device_stop(&dc->disk); } @@ -1367,7 +1369,7 @@ int bch_flash_dev_create(struct cache_set *c, uint64_t size) bool bch_cached_dev_error(struct cached_dev *dc) { - char name[BDEVNAME_SIZE]; + struct cache_set *c; if (!dc || test_bit(BCACHE_DEV_CLOSING, &dc->disk.flags)) return false; @@ -1377,7 +1379,22 @@ bool bch_cached_dev_error(struct cached_dev *dc) smp_mb(); pr_err("stop %s: too many IO errors on backing device %s\n", - dc->disk.disk->disk_name, bdevname(dc->bdev, name)); + dc->disk.disk->disk_name, dc->backing_dev_name); + + /* + * If the cached device is still attached to a cache set, + * even dc->io_disable is true and no more I/O requests + * accepted, cache device internal I/O (writeback scan or + * garbage collection) may still prevent bcache device from + * being stopped. So here CACHE_SET_IO_DISABLE should be + * set to c->flags too, to make the internal I/O to cache + * device rejected and stopped immediately. + * If c is NULL, that means the bcache device is not attached + * to any cache set, then no CACHE_SET_IO_DISABLE bit to set. + */ + c = dc->disk.c; + if (c && test_and_set_bit(CACHE_SET_IO_DISABLE, &c->flags)) + pr_info("CACHE_SET_IO_DISABLE already set"); bcache_device_stop(&dc->disk); return true; @@ -1395,7 +1412,7 @@ bool bch_cache_set_error(struct cache_set *c, const char *fmt, ...) return false; if (test_and_set_bit(CACHE_SET_IO_DISABLE, &c->flags)) - pr_warn("CACHE_SET_IO_DISABLE already set"); + pr_info("CACHE_SET_IO_DISABLE already set"); /* XXX: we can be called from atomic context acquire_console_sem(); @@ -1539,6 +1556,20 @@ static void conditional_stop_bcache_device(struct cache_set *c, */ pr_warn("stop_when_cache_set_failed of %s is \"auto\" and cache is dirty, stop it to avoid potential data corruption.", d->disk->disk_name); + /* + * There might be a small time gap that cache set is + * released but bcache device is not. Inside this time + * gap, regular I/O requests will directly go into + * backing device as no cache set attached to. This + * behavior may also introduce potential inconsistence + * data in writeback mode while cache is dirty. + * Therefore before calling bcache_device_stop() due + * to a broken cache device, dc->io_disable should be + * explicitly set to true. + */ + dc->io_disable = true; + /* make others know io_disable is true earlier */ + smp_mb(); bcache_device_stop(d); } else { /* @@ -2003,12 +2034,10 @@ static int cache_alloc(struct cache *ca) static int register_cache(struct cache_sb *sb, struct page *sb_page, struct block_device *bdev, struct cache *ca) { - char name[BDEVNAME_SIZE]; const char *err = NULL; /* must be set for any error case */ int ret = 0; - bdevname(bdev, name); - + bdevname(bdev, ca->cache_dev_name); memcpy(&ca->sb, sb, sizeof(struct cache_sb)); ca->bdev = bdev; ca->bdev->bd_holder = ca; @@ -2045,14 +2074,14 @@ static int register_cache(struct cache_sb *sb, struct page *sb_page, goto out; } - pr_info("registered cache device %s", name); + pr_info("registered cache device %s", ca->cache_dev_name); out: kobject_put(&ca->kobj); err: if (err) - pr_notice("error %s: %s", name, err); + pr_notice("error %s: %s", ca->cache_dev_name, err); return ret; } diff --git a/drivers/md/bcache/writeback.c b/drivers/md/bcache/writeback.c index 4a9547cdcdc5..ad45ebe1a74b 100644 --- a/drivers/md/bcache/writeback.c +++ b/drivers/md/bcache/writeback.c @@ -244,8 +244,10 @@ static void dirty_endio(struct bio *bio) struct keybuf_key *w = bio->bi_private; struct dirty_io *io = w->private; - if (bio->bi_status) + if (bio->bi_status) { SET_KEY_DIRTY(&w->key, false); + bch_count_backing_io_errors(io->dc, bio); + } closure_put(&io->cl); } diff --git a/drivers/md/dm-bufio.c b/drivers/md/dm-bufio.c index 12aa9ca21d8c..dc385b70e4c3 100644 --- a/drivers/md/dm-bufio.c +++ b/drivers/md/dm-bufio.c @@ -1681,8 +1681,9 @@ struct dm_bufio_client *dm_bufio_client_create(struct block_device *bdev, unsign if (block_size <= KMALLOC_MAX_SIZE && (block_size < PAGE_SIZE || !is_power_of_2(block_size))) { - snprintf(slab_name, sizeof slab_name, "dm_bufio_cache-%u", c->block_size); - c->slab_cache = kmem_cache_create(slab_name, c->block_size, ARCH_KMALLOC_MINALIGN, + unsigned align = min(1U << __ffs(block_size), (unsigned)PAGE_SIZE); + snprintf(slab_name, sizeof slab_name, "dm_bufio_cache-%u", block_size); + c->slab_cache = kmem_cache_create(slab_name, block_size, align, SLAB_RECLAIM_ACCOUNT, NULL); if (!c->slab_cache) { r = -ENOMEM; diff --git a/drivers/md/dm-cache-background-tracker.c b/drivers/md/dm-cache-background-tracker.c index 1d0af0a21fc7..84814e819e4c 100644 --- a/drivers/md/dm-cache-background-tracker.c +++ b/drivers/md/dm-cache-background-tracker.c @@ -166,7 +166,7 @@ static bool max_work_reached(struct background_tracker *b) atomic_read(&b->pending_demotes) >= b->max_work; } -struct bt_work *alloc_work(struct background_tracker *b) +static struct bt_work *alloc_work(struct background_tracker *b) { if (max_work_reached(b)) return NULL; diff --git a/drivers/md/dm-integrity.c b/drivers/md/dm-integrity.c index 77d9fe58dae2..514fb4aec5d1 100644 --- a/drivers/md/dm-integrity.c +++ b/drivers/md/dm-integrity.c @@ -2440,7 +2440,7 @@ static void dm_integrity_free_journal_scatterlist(struct dm_integrity_c *ic, str unsigned i; for (i = 0; i < ic->journal_sections; i++) kvfree(sl[i]); - kfree(sl); + kvfree(sl); } static struct scatterlist **dm_integrity_alloc_journal_scatterlist(struct dm_integrity_c *ic, struct page_list *pl) diff --git a/drivers/md/dm-raid1.c b/drivers/md/dm-raid1.c index 580c49cc8079..5903e492bb34 100644 --- a/drivers/md/dm-raid1.c +++ b/drivers/md/dm-raid1.c @@ -23,6 +23,8 @@ #define MAX_RECOVERY 1 /* Maximum number of regions recovered in parallel. */ +#define MAX_NR_MIRRORS (DM_KCOPYD_MAX_REGIONS + 1) + #define DM_RAID1_HANDLE_ERRORS 0x01 #define DM_RAID1_KEEP_LOG 0x02 #define errors_handled(p) ((p)->features & DM_RAID1_HANDLE_ERRORS) @@ -255,7 +257,7 @@ static int mirror_flush(struct dm_target *ti) unsigned long error_bits; unsigned int i; - struct dm_io_region io[ms->nr_mirrors]; + struct dm_io_region io[MAX_NR_MIRRORS]; struct mirror *m; struct dm_io_request io_req = { .bi_op = REQ_OP_WRITE, @@ -651,7 +653,7 @@ static void write_callback(unsigned long error, void *context) static void do_write(struct mirror_set *ms, struct bio *bio) { unsigned int i; - struct dm_io_region io[ms->nr_mirrors], *dest = io; + struct dm_io_region io[MAX_NR_MIRRORS], *dest = io; struct mirror *m; struct dm_io_request io_req = { .bi_op = REQ_OP_WRITE, @@ -1083,7 +1085,7 @@ static int mirror_ctr(struct dm_target *ti, unsigned int argc, char **argv) argc -= args_used; if (!argc || sscanf(argv[0], "%u%c", &nr_mirrors, &dummy) != 1 || - nr_mirrors < 2 || nr_mirrors > DM_KCOPYD_MAX_REGIONS + 1) { + nr_mirrors < 2 || nr_mirrors > MAX_NR_MIRRORS) { ti->error = "Invalid number of mirrors"; dm_dirty_log_destroy(dl); return -EINVAL; @@ -1404,7 +1406,7 @@ static void mirror_status(struct dm_target *ti, status_type_t type, int num_feature_args = 0; struct mirror_set *ms = (struct mirror_set *) ti->private; struct dm_dirty_log *log = dm_rh_dirty_log(ms->rh); - char buffer[ms->nr_mirrors + 1]; + char buffer[MAX_NR_MIRRORS + 1]; switch (type) { case STATUSTYPE_INFO: diff --git a/drivers/md/dm.c b/drivers/md/dm.c index 4ea404dbcf0b..0a7b0107ca78 100644 --- a/drivers/md/dm.c +++ b/drivers/md/dm.c @@ -1020,7 +1020,8 @@ int dm_set_target_max_io_len(struct dm_target *ti, sector_t len) EXPORT_SYMBOL_GPL(dm_set_target_max_io_len); static struct dm_target *dm_dax_get_live_target(struct mapped_device *md, - sector_t sector, int *srcu_idx) + sector_t sector, int *srcu_idx) + __acquires(md->io_barrier) { struct dm_table *map; struct dm_target *ti; @@ -1037,7 +1038,7 @@ static struct dm_target *dm_dax_get_live_target(struct mapped_device *md, } static long dm_dax_direct_access(struct dax_device *dax_dev, pgoff_t pgoff, - long nr_pages, void **kaddr, pfn_t *pfn) + long nr_pages, void **kaddr, pfn_t *pfn) { struct mapped_device *md = dax_get_private(dax_dev); sector_t sector = pgoff * PAGE_SECTORS; @@ -1065,7 +1066,7 @@ static long dm_dax_direct_access(struct dax_device *dax_dev, pgoff_t pgoff, } static size_t dm_dax_copy_from_iter(struct dax_device *dax_dev, pgoff_t pgoff, - void *addr, size_t bytes, struct iov_iter *i) + void *addr, size_t bytes, struct iov_iter *i) { struct mapped_device *md = dax_get_private(dax_dev); sector_t sector = pgoff * PAGE_SECTORS; diff --git a/drivers/media/i2c/saa7115.c b/drivers/media/i2c/saa7115.c index e216cd768409..b07114b5efb2 100644 --- a/drivers/media/i2c/saa7115.c +++ b/drivers/media/i2c/saa7115.c @@ -20,7 +20,7 @@ // // VBI support (2004) and cleanups (2005) by Hans Verkuil <hverkuil@xs4all.nl> // -// Copyright (c) 2005-2006 Mauro Carvalho Chehab <mchehab@infradead.org> +// Copyright (c) 2005-2006 Mauro Carvalho Chehab <mchehab@kernel.org> // SAA7111, SAA7113 and SAA7118 support #include "saa711x_regs.h" diff --git a/drivers/media/i2c/saa711x_regs.h b/drivers/media/i2c/saa711x_regs.h index a50d480e101a..44fabe08234d 100644 --- a/drivers/media/i2c/saa711x_regs.h +++ b/drivers/media/i2c/saa711x_regs.h @@ -2,7 +2,7 @@ * SPDX-License-Identifier: GPL-2.0+ * saa711x - Philips SAA711x video decoder register specifications * - * Copyright (c) 2006 Mauro Carvalho Chehab <mchehab@infradead.org> + * Copyright (c) 2006 Mauro Carvalho Chehab <mchehab@kernel.org> */ #define R_00_CHIP_VERSION 0x00 diff --git a/drivers/media/i2c/tda7432.c b/drivers/media/i2c/tda7432.c index 1c5c61d829d6..9b4f21237810 100644 --- a/drivers/media/i2c/tda7432.c +++ b/drivers/media/i2c/tda7432.c @@ -8,7 +8,7 @@ * Muting and tone control by Jonathan Isom <jisom@ematic.com> * * Copyright (c) 2000 Eric Sandeen <eric_sandeen@bigfoot.com> - * Copyright (c) 2006 Mauro Carvalho Chehab <mchehab@infradead.org> + * Copyright (c) 2006 Mauro Carvalho Chehab <mchehab@kernel.org> * This code is placed under the terms of the GNU General Public License * Based on tda9855.c by Steve VanDeBogart (vandebo@uclink.berkeley.edu) * Which was based on tda8425.c by Greg Alexander (c) 1998 diff --git a/drivers/media/i2c/tvp5150.c b/drivers/media/i2c/tvp5150.c index 2476d812f669..1734ed4ede33 100644 --- a/drivers/media/i2c/tvp5150.c +++ b/drivers/media/i2c/tvp5150.c @@ -2,7 +2,7 @@ // // tvp5150 - Texas Instruments TVP5150A/AM1 and TVP5151 video decoder driver // -// Copyright (c) 2005,2006 Mauro Carvalho Chehab <mchehab@infradead.org> +// Copyright (c) 2005,2006 Mauro Carvalho Chehab <mchehab@kernel.org> #include <dt-bindings/media/tvp5150.h> #include <linux/i2c.h> diff --git a/drivers/media/i2c/tvp5150_reg.h b/drivers/media/i2c/tvp5150_reg.h index c43b7b844021..d3a764cae1a0 100644 --- a/drivers/media/i2c/tvp5150_reg.h +++ b/drivers/media/i2c/tvp5150_reg.h @@ -3,7 +3,7 @@ * * tvp5150 - Texas Instruments TVP5150A/AM1 video decoder registers * - * Copyright (c) 2005,2006 Mauro Carvalho Chehab <mchehab@infradead.org> + * Copyright (c) 2005,2006 Mauro Carvalho Chehab <mchehab@kernel.org> */ #define TVP5150_VD_IN_SRC_SEL_1 0x00 /* Video input source selection #1 */ diff --git a/drivers/media/i2c/tvp7002.c b/drivers/media/i2c/tvp7002.c index a26c1a3f7183..4599b7e28a8d 100644 --- a/drivers/media/i2c/tvp7002.c +++ b/drivers/media/i2c/tvp7002.c @@ -5,7 +5,7 @@ * Author: Santiago Nunez-Corrales <santiago.nunez@ridgerun.com> * * This code is partially based upon the TVP5150 driver - * written by Mauro Carvalho Chehab (mchehab@infradead.org), + * written by Mauro Carvalho Chehab <mchehab@kernel.org>, * the TVP514x driver written by Vaibhav Hiremath <hvaibhav@ti.com> * and the TVP7002 driver in the TI LSP 2.10.00.14. Revisions by * Muralidharan Karicheri and Snehaprabha Narnakaje (TI). diff --git a/drivers/media/i2c/tvp7002_reg.h b/drivers/media/i2c/tvp7002_reg.h index 3c8c8b0a6a4c..7f56ba689dfe 100644 --- a/drivers/media/i2c/tvp7002_reg.h +++ b/drivers/media/i2c/tvp7002_reg.h @@ -5,7 +5,7 @@ * Author: Santiago Nunez-Corrales <santiago.nunez@ridgerun.com> * * This code is partially based upon the TVP5150 driver - * written by Mauro Carvalho Chehab (mchehab@infradead.org), + * written by Mauro Carvalho Chehab <mchehab@kernel.org>, * the TVP514x driver written by Vaibhav Hiremath <hvaibhav@ti.com> * and the TVP7002 driver in the TI LSP 2.10.00.14 * diff --git a/drivers/media/media-devnode.c b/drivers/media/media-devnode.c index 67ac51eff15c..6b87a721dc49 100644 --- a/drivers/media/media-devnode.c +++ b/drivers/media/media-devnode.c @@ -4,7 +4,7 @@ * Copyright (C) 2010 Nokia Corporation * * Based on drivers/media/video/v4l2_dev.c code authored by - * Mauro Carvalho Chehab <mchehab@infradead.org> (version 2) + * Mauro Carvalho Chehab <mchehab@kernel.org> (version 2) * Alan Cox, <alan@lxorguk.ukuu.org.uk> (version 1) * * Contacts: Laurent Pinchart <laurent.pinchart@ideasonboard.com> diff --git a/drivers/media/pci/bt8xx/bttv-audio-hook.c b/drivers/media/pci/bt8xx/bttv-audio-hook.c index 9f1f9169fb5b..346fc7f58839 100644 --- a/drivers/media/pci/bt8xx/bttv-audio-hook.c +++ b/drivers/media/pci/bt8xx/bttv-audio-hook.c @@ -1,7 +1,7 @@ /* * Handlers for board audio hooks, splitted from bttv-cards * - * Copyright (c) 2006 Mauro Carvalho Chehab (mchehab@infradead.org) + * Copyright (c) 2006 Mauro Carvalho Chehab <mchehab@kernel.org> * This code is placed under the terms of the GNU General Public License */ diff --git a/drivers/media/pci/bt8xx/bttv-audio-hook.h b/drivers/media/pci/bt8xx/bttv-audio-hook.h index 159d07adeff8..be16a537a03a 100644 --- a/drivers/media/pci/bt8xx/bttv-audio-hook.h +++ b/drivers/media/pci/bt8xx/bttv-audio-hook.h @@ -1,7 +1,7 @@ /* * Handlers for board audio hooks, splitted from bttv-cards * - * Copyright (c) 2006 Mauro Carvalho Chehab (mchehab@infradead.org) + * Copyright (c) 2006 Mauro Carvalho Chehab <mchehab@kernel.org> * This code is placed under the terms of the GNU General Public License */ diff --git a/drivers/media/pci/bt8xx/bttv-cards.c b/drivers/media/pci/bt8xx/bttv-cards.c index 1902732f90e1..2616243b2c49 100644 --- a/drivers/media/pci/bt8xx/bttv-cards.c +++ b/drivers/media/pci/bt8xx/bttv-cards.c @@ -2447,7 +2447,7 @@ struct tvcard bttv_tvcards[] = { }, /* ---- card 0x88---------------------------------- */ [BTTV_BOARD_ACORP_Y878F] = { - /* Mauro Carvalho Chehab <mchehab@infradead.org> */ + /* Mauro Carvalho Chehab <mchehab@kernel.org> */ .name = "Acorp Y878F", .video_inputs = 3, /* .audio_inputs= 1, */ @@ -2688,7 +2688,7 @@ struct tvcard bttv_tvcards[] = { }, [BTTV_BOARD_ENLTV_FM_2] = { /* Encore TV Tuner Pro ENL TV-FM-2 - Mauro Carvalho Chehab <mchehab@infradead.org */ + Mauro Carvalho Chehab <mchehab@kernel.org> */ .name = "Encore ENL TV-FM-2", .video_inputs = 3, /* .audio_inputs= 1, */ diff --git a/drivers/media/pci/bt8xx/bttv-driver.c b/drivers/media/pci/bt8xx/bttv-driver.c index 707f57a9f940..de3f44b8dec6 100644 --- a/drivers/media/pci/bt8xx/bttv-driver.c +++ b/drivers/media/pci/bt8xx/bttv-driver.c @@ -13,7 +13,7 @@ (c) 2005-2006 Nickolay V. Shmyrev <nshmyrev@yandex.ru> Fixes to be fully V4L2 compliant by - (c) 2006 Mauro Carvalho Chehab <mchehab@infradead.org> + (c) 2006 Mauro Carvalho Chehab <mchehab@kernel.org> Cropping and overscan support Copyright (C) 2005, 2006 Michael H. Schimek <mschimek@gmx.at> diff --git a/drivers/media/pci/bt8xx/bttv-i2c.c b/drivers/media/pci/bt8xx/bttv-i2c.c index eccd1e3d717a..c76823eb399d 100644 --- a/drivers/media/pci/bt8xx/bttv-i2c.c +++ b/drivers/media/pci/bt8xx/bttv-i2c.c @@ -8,7 +8,7 @@ & Marcus Metzler (mocm@thp.uni-koeln.de) (c) 1999-2003 Gerd Knorr <kraxel@bytesex.org> - (c) 2005 Mauro Carvalho Chehab <mchehab@infradead.org> + (c) 2005 Mauro Carvalho Chehab <mchehab@kernel.org> - Multituner support and i2c address binding This program is free software; you can redistribute it and/or modify diff --git a/drivers/media/pci/cx23885/cx23885-input.c b/drivers/media/pci/cx23885/cx23885-input.c index be49589a61d2..395ff9bba759 100644 --- a/drivers/media/pci/cx23885/cx23885-input.c +++ b/drivers/media/pci/cx23885/cx23885-input.c @@ -13,7 +13,7 @@ * Copyright (C) 2008 <srinivasa.deevi at conexant dot com> * Copyright (C) 2005 Ludovico Cavedon <cavedon@sssup.it> * Markus Rechberger <mrechberger@gmail.com> - * Mauro Carvalho Chehab <mchehab@infradead.org> + * Mauro Carvalho Chehab <mchehab@kernel.org> * Sascha Sommer <saschasommer@freenet.de> * Copyright (C) 2004, 2005 Chris Pascoe * Copyright (C) 2003, 2004 Gerd Knorr diff --git a/drivers/media/pci/cx88/cx88-alsa.c b/drivers/media/pci/cx88/cx88-alsa.c index ab09bb55cf45..8a28fda703a2 100644 --- a/drivers/media/pci/cx88/cx88-alsa.c +++ b/drivers/media/pci/cx88/cx88-alsa.c @@ -4,7 +4,7 @@ * * (c) 2007 Trent Piepho <xyzzy@speakeasy.org> * (c) 2005,2006 Ricardo Cerqueira <v4l@cerqueira.org> - * (c) 2005 Mauro Carvalho Chehab <mchehab@infradead.org> + * (c) 2005 Mauro Carvalho Chehab <mchehab@kernel.org> * Based on a dummy cx88 module by Gerd Knorr <kraxel@bytesex.org> * Based on dummy.c by Jaroslav Kysela <perex@perex.cz> * @@ -103,7 +103,7 @@ MODULE_PARM_DESC(index, "Index value for cx88x capture interface(s)."); MODULE_DESCRIPTION("ALSA driver module for cx2388x based TV cards"); MODULE_AUTHOR("Ricardo Cerqueira"); -MODULE_AUTHOR("Mauro Carvalho Chehab <mchehab@infradead.org>"); +MODULE_AUTHOR("Mauro Carvalho Chehab <mchehab@kernel.org>"); MODULE_LICENSE("GPL"); MODULE_VERSION(CX88_VERSION); diff --git a/drivers/media/pci/cx88/cx88-blackbird.c b/drivers/media/pci/cx88/cx88-blackbird.c index 0e0952e60795..7a4876cf9f08 100644 --- a/drivers/media/pci/cx88/cx88-blackbird.c +++ b/drivers/media/pci/cx88/cx88-blackbird.c @@ -5,7 +5,7 @@ * (c) 2004 Jelle Foks <jelle@foks.us> * (c) 2004 Gerd Knorr <kraxel@bytesex.org> * - * (c) 2005-2006 Mauro Carvalho Chehab <mchehab@infradead.org> + * (c) 2005-2006 Mauro Carvalho Chehab <mchehab@kernel.org> * - video_ioctl2 conversion * * Includes parts from the ivtv driver <http://sourceforge.net/projects/ivtv/> diff --git a/drivers/media/pci/cx88/cx88-core.c b/drivers/media/pci/cx88/cx88-core.c index 8bfa5b7ed91b..60988e95b637 100644 --- a/drivers/media/pci/cx88/cx88-core.c +++ b/drivers/media/pci/cx88/cx88-core.c @@ -4,7 +4,7 @@ * * (c) 2003 Gerd Knorr <kraxel@bytesex.org> [SuSE Labs] * - * (c) 2005-2006 Mauro Carvalho Chehab <mchehab@infradead.org> + * (c) 2005-2006 Mauro Carvalho Chehab <mchehab@kernel.org> * - Multituner support * - video_ioctl2 conversion * - PAL/M fixes diff --git a/drivers/media/pci/cx88/cx88-i2c.c b/drivers/media/pci/cx88/cx88-i2c.c index f7692775fb5a..99f88a05a7c9 100644 --- a/drivers/media/pci/cx88/cx88-i2c.c +++ b/drivers/media/pci/cx88/cx88-i2c.c @@ -8,7 +8,7 @@ * (c) 2002 Yurij Sysoev <yurij@naturesoft.net> * (c) 1999-2003 Gerd Knorr <kraxel@bytesex.org> * - * (c) 2005 Mauro Carvalho Chehab <mchehab@infradead.org> + * (c) 2005 Mauro Carvalho Chehab <mchehab@kernel.org> * - Multituner support and i2c address binding * * This program is free software; you can redistribute it and/or modify diff --git a/drivers/media/pci/cx88/cx88-video.c b/drivers/media/pci/cx88/cx88-video.c index 9be682cdb644..7b113bad70d2 100644 --- a/drivers/media/pci/cx88/cx88-video.c +++ b/drivers/media/pci/cx88/cx88-video.c @@ -5,7 +5,7 @@ * * (c) 2003-04 Gerd Knorr <kraxel@bytesex.org> [SuSE Labs] * - * (c) 2005-2006 Mauro Carvalho Chehab <mchehab@infradead.org> + * (c) 2005-2006 Mauro Carvalho Chehab <mchehab@kernel.org> * - Multituner support * - video_ioctl2 conversion * - PAL/M fixes diff --git a/drivers/media/radio/radio-aimslab.c b/drivers/media/radio/radio-aimslab.c index 5ef635e72e10..4c52ac6d8bc5 100644 --- a/drivers/media/radio/radio-aimslab.c +++ b/drivers/media/radio/radio-aimslab.c @@ -4,7 +4,7 @@ * Copyright 1997 M. Kirkwood * * Converted to the radio-isa framework by Hans Verkuil <hans.verkuil@cisco.com> - * Converted to V4L2 API by Mauro Carvalho Chehab <mchehab@infradead.org> + * Converted to V4L2 API by Mauro Carvalho Chehab <mchehab@kernel.org> * Converted to new API by Alan Cox <alan@lxorguk.ukuu.org.uk> * Various bugfixes and enhancements by Russell Kroll <rkroll@exploits.org> * diff --git a/drivers/media/radio/radio-aztech.c b/drivers/media/radio/radio-aztech.c index 9e12c6027359..840b7d60462b 100644 --- a/drivers/media/radio/radio-aztech.c +++ b/drivers/media/radio/radio-aztech.c @@ -2,7 +2,7 @@ * radio-aztech.c - Aztech radio card driver * * Converted to the radio-isa framework by Hans Verkuil <hans.verkuil@xs4all.nl> - * Converted to V4L2 API by Mauro Carvalho Chehab <mchehab@infradead.org> + * Converted to V4L2 API by Mauro Carvalho Chehab <mchehab@kernel.org> * Adapted to support the Video for Linux API by * Russell Kroll <rkroll@exploits.org>. Based on original tuner code by: * diff --git a/drivers/media/radio/radio-gemtek.c b/drivers/media/radio/radio-gemtek.c index 3ff4c4e1435f..f051f8694ab9 100644 --- a/drivers/media/radio/radio-gemtek.c +++ b/drivers/media/radio/radio-gemtek.c @@ -15,7 +15,7 @@ * Various bugfixes and enhancements by Russell Kroll <rkroll@exploits.org> * * Converted to the radio-isa framework by Hans Verkuil <hans.verkuil@cisco.com> - * Converted to V4L2 API by Mauro Carvalho Chehab <mchehab@infradead.org> + * Converted to V4L2 API by Mauro Carvalho Chehab <mchehab@kernel.org> * * Note: this card seems to swap the left and right audio channels! * diff --git a/drivers/media/radio/radio-maxiradio.c b/drivers/media/radio/radio-maxiradio.c index 95f06f3b35dc..e4e758739246 100644 --- a/drivers/media/radio/radio-maxiradio.c +++ b/drivers/media/radio/radio-maxiradio.c @@ -27,7 +27,7 @@ * BUGS: * - card unmutes if you change frequency * - * (c) 2006, 2007 by Mauro Carvalho Chehab <mchehab@infradead.org>: + * (c) 2006, 2007 by Mauro Carvalho Chehab <mchehab@kernel.org>: * - Conversion to V4L2 API * - Uses video_ioctl2 for parsing and to add debug support */ diff --git a/drivers/media/radio/radio-rtrack2.c b/drivers/media/radio/radio-rtrack2.c index abeaedd8d437..5a1470eb753e 100644 --- a/drivers/media/radio/radio-rtrack2.c +++ b/drivers/media/radio/radio-rtrack2.c @@ -7,7 +7,7 @@ * Various bugfixes and enhancements by Russell Kroll <rkroll@exploits.org> * * Converted to the radio-isa framework by Hans Verkuil <hans.verkuil@cisco.com> - * Converted to V4L2 API by Mauro Carvalho Chehab <mchehab@infradead.org> + * Converted to V4L2 API by Mauro Carvalho Chehab <mchehab@kernel.org> * * Fully tested with actual hardware and the v4l2-compliance tool. */ diff --git a/drivers/media/radio/radio-sf16fmi.c b/drivers/media/radio/radio-sf16fmi.c index fc4e63d36e4c..4f9b97edd9eb 100644 --- a/drivers/media/radio/radio-sf16fmi.c +++ b/drivers/media/radio/radio-sf16fmi.c @@ -13,7 +13,7 @@ * No volume control - only mute/unmute - you have to use line volume * control on SB-part of SF16-FMI/SF16-FMP/SF16-FMD * - * Converted to V4L2 API by Mauro Carvalho Chehab <mchehab@infradead.org> + * Converted to V4L2 API by Mauro Carvalho Chehab <mchehab@kernel.org> */ #include <linux/kernel.h> /* __setup */ diff --git a/drivers/media/radio/radio-terratec.c b/drivers/media/radio/radio-terratec.c index 4f116ea294fb..1af8f29cc7d1 100644 --- a/drivers/media/radio/radio-terratec.c +++ b/drivers/media/radio/radio-terratec.c @@ -17,7 +17,7 @@ * Volume Control is done digitally * * Converted to the radio-isa framework by Hans Verkuil <hans.verkuil@cisco.com> - * Converted to V4L2 API by Mauro Carvalho Chehab <mchehab@infradead.org> + * Converted to V4L2 API by Mauro Carvalho Chehab <mchehab@kernel.org> */ #include <linux/module.h> /* Modules */ diff --git a/drivers/media/radio/radio-trust.c b/drivers/media/radio/radio-trust.c index 26a8c6002121..a4bad322ffff 100644 --- a/drivers/media/radio/radio-trust.c +++ b/drivers/media/radio/radio-trust.c @@ -12,7 +12,7 @@ * Scott McGrath (smcgrath@twilight.vtc.vsc.edu) * William McGrath (wmcgrath@twilight.vtc.vsc.edu) * - * Converted to V4L2 API by Mauro Carvalho Chehab <mchehab@infradead.org> + * Converted to V4L2 API by Mauro Carvalho Chehab <mchehab@kernel.org> */ #include <stdarg.h> diff --git a/drivers/media/radio/radio-typhoon.c b/drivers/media/radio/radio-typhoon.c index eb72a4d13758..d0d67ad85b8f 100644 --- a/drivers/media/radio/radio-typhoon.c +++ b/drivers/media/radio/radio-typhoon.c @@ -25,7 +25,7 @@ * The frequency change is necessary since the card never seems to be * completely silent. * - * Converted to V4L2 API by Mauro Carvalho Chehab <mchehab@infradead.org> + * Converted to V4L2 API by Mauro Carvalho Chehab <mchehab@kernel.org> */ #include <linux/module.h> /* Modules */ diff --git a/drivers/media/radio/radio-zoltrix.c b/drivers/media/radio/radio-zoltrix.c index 026e88eef29c..6007cd09b328 100644 --- a/drivers/media/radio/radio-zoltrix.c +++ b/drivers/media/radio/radio-zoltrix.c @@ -27,7 +27,7 @@ * 2002-07-15 - Fix Stereo typo * * 2006-07-24 - Converted to V4L2 API - * by Mauro Carvalho Chehab <mchehab@infradead.org> + * by Mauro Carvalho Chehab <mchehab@kernel.org> * * Converted to the radio-isa framework by Hans Verkuil <hans.verkuil@cisco.com> * diff --git a/drivers/media/rc/keymaps/rc-avermedia-m135a.c b/drivers/media/rc/keymaps/rc-avermedia-m135a.c index f6977df1a75b..d275d98d066a 100644 --- a/drivers/media/rc/keymaps/rc-avermedia-m135a.c +++ b/drivers/media/rc/keymaps/rc-avermedia-m135a.c @@ -12,7 +12,7 @@ * * On Avermedia M135A with IR model RM-JX, the same codes exist on both * Positivo (BR) and original IR, initial version and remote control codes - * added by Mauro Carvalho Chehab <mchehab@infradead.org> + * added by Mauro Carvalho Chehab <mchehab@kernel.org> * * Positivo also ships Avermedia M135A with model RM-K6, extra control * codes added by Herton Ronaldo Krzesinski <herton@mandriva.com.br> diff --git a/drivers/media/rc/keymaps/rc-encore-enltv-fm53.c b/drivers/media/rc/keymaps/rc-encore-enltv-fm53.c index e4e78c1f4123..057c13b765ef 100644 --- a/drivers/media/rc/keymaps/rc-encore-enltv-fm53.c +++ b/drivers/media/rc/keymaps/rc-encore-enltv-fm53.c @@ -9,7 +9,7 @@ #include <linux/module.h> /* Encore ENLTV-FM v5.3 - Mauro Carvalho Chehab <mchehab@infradead.org> + Mauro Carvalho Chehab <mchehab@kernel.org> */ static struct rc_map_table encore_enltv_fm53[] = { diff --git a/drivers/media/rc/keymaps/rc-encore-enltv2.c b/drivers/media/rc/keymaps/rc-encore-enltv2.c index c3d4437a6fda..cd0555924456 100644 --- a/drivers/media/rc/keymaps/rc-encore-enltv2.c +++ b/drivers/media/rc/keymaps/rc-encore-enltv2.c @@ -9,7 +9,7 @@ #include <linux/module.h> /* Encore ENLTV2-FM - silver plastic - "Wand Media" written at the botton - Mauro Carvalho Chehab <mchehab@infradead.org> */ + Mauro Carvalho Chehab <mchehab@kernel.org> */ static struct rc_map_table encore_enltv2[] = { { 0x4c, KEY_POWER2 }, diff --git a/drivers/media/rc/keymaps/rc-kaiomy.c b/drivers/media/rc/keymaps/rc-kaiomy.c index f0f88df18606..a00051339842 100644 --- a/drivers/media/rc/keymaps/rc-kaiomy.c +++ b/drivers/media/rc/keymaps/rc-kaiomy.c @@ -9,7 +9,7 @@ #include <linux/module.h> /* Kaiomy TVnPC U2 - Mauro Carvalho Chehab <mchehab@infradead.org> + Mauro Carvalho Chehab <mchehab@kernel.org> */ static struct rc_map_table kaiomy[] = { diff --git a/drivers/media/rc/keymaps/rc-kworld-plus-tv-analog.c b/drivers/media/rc/keymaps/rc-kworld-plus-tv-analog.c index 453e04377de7..db5edde3eeb1 100644 --- a/drivers/media/rc/keymaps/rc-kworld-plus-tv-analog.c +++ b/drivers/media/rc/keymaps/rc-kworld-plus-tv-analog.c @@ -9,7 +9,7 @@ #include <linux/module.h> /* Kworld Plus TV Analog Lite PCI IR - Mauro Carvalho Chehab <mchehab@infradead.org> + Mauro Carvalho Chehab <mchehab@kernel.org> */ static struct rc_map_table kworld_plus_tv_analog[] = { diff --git a/drivers/media/rc/keymaps/rc-pixelview-new.c b/drivers/media/rc/keymaps/rc-pixelview-new.c index 791130f108ff..e4e34f2ccf74 100644 --- a/drivers/media/rc/keymaps/rc-pixelview-new.c +++ b/drivers/media/rc/keymaps/rc-pixelview-new.c @@ -9,7 +9,7 @@ #include <linux/module.h> /* - Mauro Carvalho Chehab <mchehab@infradead.org> + Mauro Carvalho Chehab <mchehab@kernel.org> present on PV MPEG 8000GT */ diff --git a/drivers/media/tuners/tea5761.c b/drivers/media/tuners/tea5761.c index 88b3e80c38ad..d78a2bdb3e36 100644 --- a/drivers/media/tuners/tea5761.c +++ b/drivers/media/tuners/tea5761.c @@ -2,7 +2,7 @@ // For Philips TEA5761 FM Chip // I2C address is always 0x20 (0x10 at 7-bit mode). // -// Copyright (c) 2005-2007 Mauro Carvalho Chehab (mchehab@infradead.org) +// Copyright (c) 2005-2007 Mauro Carvalho Chehab <mchehab@kernel.org> #include <linux/i2c.h> #include <linux/slab.h> @@ -337,5 +337,5 @@ EXPORT_SYMBOL_GPL(tea5761_attach); EXPORT_SYMBOL_GPL(tea5761_autodetection); MODULE_DESCRIPTION("Philips TEA5761 FM tuner driver"); -MODULE_AUTHOR("Mauro Carvalho Chehab <mchehab@infradead.org>"); +MODULE_AUTHOR("Mauro Carvalho Chehab <mchehab@kernel.org>"); MODULE_LICENSE("GPL v2"); diff --git a/drivers/media/tuners/tea5767.c b/drivers/media/tuners/tea5767.c index 2b2c064d7dc3..016d0d5ec50b 100644 --- a/drivers/media/tuners/tea5767.c +++ b/drivers/media/tuners/tea5767.c @@ -2,7 +2,7 @@ // For Philips TEA5767 FM Chip used on some TV Cards like Prolink Pixelview // I2C address is always 0xC0. // -// Copyright (c) 2005 Mauro Carvalho Chehab (mchehab@infradead.org) +// Copyright (c) 2005 Mauro Carvalho Chehab <mchehab@kernel.org> // // tea5767 autodetection thanks to Torsten Seeboth and Atsushi Nakagawa // from their contributions on DScaler. @@ -469,5 +469,5 @@ EXPORT_SYMBOL_GPL(tea5767_attach); EXPORT_SYMBOL_GPL(tea5767_autodetection); MODULE_DESCRIPTION("Philips TEA5767 FM tuner driver"); -MODULE_AUTHOR("Mauro Carvalho Chehab <mchehab@infradead.org>"); +MODULE_AUTHOR("Mauro Carvalho Chehab <mchehab@kernel.org>"); MODULE_LICENSE("GPL v2"); diff --git a/drivers/media/tuners/tuner-xc2028-types.h b/drivers/media/tuners/tuner-xc2028-types.h index bb0437c36c03..50d017a4822a 100644 --- a/drivers/media/tuners/tuner-xc2028-types.h +++ b/drivers/media/tuners/tuner-xc2028-types.h @@ -5,7 +5,7 @@ * This file includes internal tipes to be used inside tuner-xc2028. * Shouldn't be included outside tuner-xc2028 * - * Copyright (c) 2007-2008 Mauro Carvalho Chehab (mchehab@infradead.org) + * Copyright (c) 2007-2008 Mauro Carvalho Chehab <mchehab@kernel.org> */ /* xc3028 firmware types */ diff --git a/drivers/media/tuners/tuner-xc2028.c b/drivers/media/tuners/tuner-xc2028.c index fca85e08ebd7..84744e138982 100644 --- a/drivers/media/tuners/tuner-xc2028.c +++ b/drivers/media/tuners/tuner-xc2028.c @@ -1,7 +1,7 @@ // SPDX-License-Identifier: GPL-2.0 // tuner-xc2028 // -// Copyright (c) 2007-2008 Mauro Carvalho Chehab (mchehab@infradead.org) +// Copyright (c) 2007-2008 Mauro Carvalho Chehab <mchehab@kernel.org> // // Copyright (c) 2007 Michel Ludwig (michel.ludwig@gmail.com) // - frontend interface @@ -1518,7 +1518,7 @@ EXPORT_SYMBOL(xc2028_attach); MODULE_DESCRIPTION("Xceive xc2028/xc3028 tuner driver"); MODULE_AUTHOR("Michel Ludwig <michel.ludwig@gmail.com>"); -MODULE_AUTHOR("Mauro Carvalho Chehab <mchehab@infradead.org>"); +MODULE_AUTHOR("Mauro Carvalho Chehab <mchehab@kernel.org>"); MODULE_LICENSE("GPL v2"); MODULE_FIRMWARE(XC2028_DEFAULT_FIRMWARE); MODULE_FIRMWARE(XC3028L_DEFAULT_FIRMWARE); diff --git a/drivers/media/tuners/tuner-xc2028.h b/drivers/media/tuners/tuner-xc2028.h index 03fd6d4233a4..7b58bc06e35c 100644 --- a/drivers/media/tuners/tuner-xc2028.h +++ b/drivers/media/tuners/tuner-xc2028.h @@ -2,7 +2,7 @@ * SPDX-License-Identifier: GPL-2.0 * tuner-xc2028 * - * Copyright (c) 2007-2008 Mauro Carvalho Chehab (mchehab@infradead.org) + * Copyright (c) 2007-2008 Mauro Carvalho Chehab <mchehab@kernel.org> */ #ifndef __TUNER_XC2028_H__ diff --git a/drivers/media/usb/em28xx/em28xx-camera.c b/drivers/media/usb/em28xx/em28xx-camera.c index 3c2694a16ed1..d1e66b503f4d 100644 --- a/drivers/media/usb/em28xx/em28xx-camera.c +++ b/drivers/media/usb/em28xx/em28xx-camera.c @@ -2,7 +2,7 @@ // // em28xx-camera.c - driver for Empia EM25xx/27xx/28xx USB video capture devices // -// Copyright (C) 2009 Mauro Carvalho Chehab <mchehab@infradead.org> +// Copyright (C) 2009 Mauro Carvalho Chehab <mchehab@kernel.org> // Copyright (C) 2013 Frank Schäfer <fschaefer.oss@googlemail.com> // // This program is free software; you can redistribute it and/or modify diff --git a/drivers/media/usb/em28xx/em28xx-cards.c b/drivers/media/usb/em28xx/em28xx-cards.c index 6e0e67d23876..7c3203d7044b 100644 --- a/drivers/media/usb/em28xx/em28xx-cards.c +++ b/drivers/media/usb/em28xx/em28xx-cards.c @@ -5,7 +5,7 @@ // // Copyright (C) 2005 Ludovico Cavedon <cavedon@sssup.it> // Markus Rechberger <mrechberger@gmail.com> -// Mauro Carvalho Chehab <mchehab@infradead.org> +// Mauro Carvalho Chehab <mchehab@kernel.org> // Sascha Sommer <saschasommer@freenet.de> // Copyright (C) 2012 Frank Schäfer <fschaefer.oss@googlemail.com> // diff --git a/drivers/media/usb/em28xx/em28xx-core.c b/drivers/media/usb/em28xx/em28xx-core.c index 36d341fb65dd..f28995383090 100644 --- a/drivers/media/usb/em28xx/em28xx-core.c +++ b/drivers/media/usb/em28xx/em28xx-core.c @@ -4,7 +4,7 @@ // // Copyright (C) 2005 Ludovico Cavedon <cavedon@sssup.it> // Markus Rechberger <mrechberger@gmail.com> -// Mauro Carvalho Chehab <mchehab@infradead.org> +// Mauro Carvalho Chehab <mchehab@kernel.org> // Sascha Sommer <saschasommer@freenet.de> // Copyright (C) 2012 Frank Schäfer <fschaefer.oss@googlemail.com> // @@ -32,7 +32,7 @@ #define DRIVER_AUTHOR "Ludovico Cavedon <cavedon@sssup.it>, " \ "Markus Rechberger <mrechberger@gmail.com>, " \ - "Mauro Carvalho Chehab <mchehab@infradead.org>, " \ + "Mauro Carvalho Chehab <mchehab@kernel.org>, " \ "Sascha Sommer <saschasommer@freenet.de>" MODULE_AUTHOR(DRIVER_AUTHOR); diff --git a/drivers/media/usb/em28xx/em28xx-dvb.c b/drivers/media/usb/em28xx/em28xx-dvb.c index a54cb8dc52c9..3f493e0b0716 100644 --- a/drivers/media/usb/em28xx/em28xx-dvb.c +++ b/drivers/media/usb/em28xx/em28xx-dvb.c @@ -2,7 +2,7 @@ // // DVB device driver for em28xx // -// (c) 2008-2011 Mauro Carvalho Chehab <mchehab@infradead.org> +// (c) 2008-2011 Mauro Carvalho Chehab <mchehab@kernel.org> // // (c) 2008 Devin Heitmueller <devin.heitmueller@gmail.com> // - Fixes for the driver to properly work with HVR-950 @@ -63,7 +63,7 @@ #include "tc90522.h" #include "qm1d1c0042.h" -MODULE_AUTHOR("Mauro Carvalho Chehab <mchehab@infradead.org>"); +MODULE_AUTHOR("Mauro Carvalho Chehab <mchehab@kernel.org>"); MODULE_LICENSE("GPL v2"); MODULE_DESCRIPTION(DRIVER_DESC " - digital TV interface"); MODULE_VERSION(EM28XX_VERSION); diff --git a/drivers/media/usb/em28xx/em28xx-i2c.c b/drivers/media/usb/em28xx/em28xx-i2c.c index 9151bccd859a..6458682bc6e2 100644 --- a/drivers/media/usb/em28xx/em28xx-i2c.c +++ b/drivers/media/usb/em28xx/em28xx-i2c.c @@ -4,7 +4,7 @@ // // Copyright (C) 2005 Ludovico Cavedon <cavedon@sssup.it> // Markus Rechberger <mrechberger@gmail.com> -// Mauro Carvalho Chehab <mchehab@infradead.org> +// Mauro Carvalho Chehab <mchehab@kernel.org> // Sascha Sommer <saschasommer@freenet.de> // Copyright (C) 2013 Frank Schäfer <fschaefer.oss@googlemail.com> // diff --git a/drivers/media/usb/em28xx/em28xx-input.c b/drivers/media/usb/em28xx/em28xx-input.c index 2dc1be00b8b8..f84a1208d5d3 100644 --- a/drivers/media/usb/em28xx/em28xx-input.c +++ b/drivers/media/usb/em28xx/em28xx-input.c @@ -4,7 +4,7 @@ // // Copyright (C) 2005 Ludovico Cavedon <cavedon@sssup.it> // Markus Rechberger <mrechberger@gmail.com> -// Mauro Carvalho Chehab <mchehab@infradead.org> +// Mauro Carvalho Chehab <mchehab@kernel.org> // Sascha Sommer <saschasommer@freenet.de> // // This program is free software; you can redistribute it and/or modify diff --git a/drivers/media/usb/em28xx/em28xx-video.c b/drivers/media/usb/em28xx/em28xx-video.c index d70ee13cc52e..68571bf36d28 100644 --- a/drivers/media/usb/em28xx/em28xx-video.c +++ b/drivers/media/usb/em28xx/em28xx-video.c @@ -5,7 +5,7 @@ // // Copyright (C) 2005 Ludovico Cavedon <cavedon@sssup.it> // Markus Rechberger <mrechberger@gmail.com> -// Mauro Carvalho Chehab <mchehab@infradead.org> +// Mauro Carvalho Chehab <mchehab@kernel.org> // Sascha Sommer <saschasommer@freenet.de> // Copyright (C) 2012 Frank Schäfer <fschaefer.oss@googlemail.com> // @@ -44,7 +44,7 @@ #define DRIVER_AUTHOR "Ludovico Cavedon <cavedon@sssup.it>, " \ "Markus Rechberger <mrechberger@gmail.com>, " \ - "Mauro Carvalho Chehab <mchehab@infradead.org>, " \ + "Mauro Carvalho Chehab <mchehab@kernel.org>, " \ "Sascha Sommer <saschasommer@freenet.de>" static unsigned int isoc_debug; diff --git a/drivers/media/usb/em28xx/em28xx.h b/drivers/media/usb/em28xx/em28xx.h index 63c7c6124707..b0378e77ddff 100644 --- a/drivers/media/usb/em28xx/em28xx.h +++ b/drivers/media/usb/em28xx/em28xx.h @@ -4,7 +4,7 @@ * * Copyright (C) 2005 Markus Rechberger <mrechberger@gmail.com> * Ludovico Cavedon <cavedon@sssup.it> - * Mauro Carvalho Chehab <mchehab@infradead.org> + * Mauro Carvalho Chehab <mchehab@kernel.org> * Copyright (C) 2012 Frank Schäfer <fschaefer.oss@googlemail.com> * * Based on the em2800 driver from Sascha Sommer <saschasommer@freenet.de> diff --git a/drivers/media/usb/gspca/zc3xx-reg.h b/drivers/media/usb/gspca/zc3xx-reg.h index a1bd94e8ce52..71fda38e85e0 100644 --- a/drivers/media/usb/gspca/zc3xx-reg.h +++ b/drivers/media/usb/gspca/zc3xx-reg.h @@ -1,7 +1,7 @@ /* * zc030x registers * - * Copyright (c) 2008 Mauro Carvalho Chehab <mchehab@infradead.org> + * Copyright (c) 2008 Mauro Carvalho Chehab <mchehab@kernel.org> * * The register aliases used here came from this driver: * http://zc0302.sourceforge.net/zc0302.php diff --git a/drivers/media/usb/tm6000/tm6000-cards.c b/drivers/media/usb/tm6000/tm6000-cards.c index 70939e96b856..23df50aa0a4a 100644 --- a/drivers/media/usb/tm6000/tm6000-cards.c +++ b/drivers/media/usb/tm6000/tm6000-cards.c @@ -1,7 +1,7 @@ // SPDX-License-Identifier: GPL-2.0 // tm6000-cards.c - driver for TM5600/TM6000/TM6010 USB video capture devices // -// Copyright (c) 2006-2007 Mauro Carvalho Chehab <mchehab@infradead.org> +// Copyright (c) 2006-2007 Mauro Carvalho Chehab <mchehab@kernel.org> #include <linux/init.h> #include <linux/module.h> diff --git a/drivers/media/usb/tm6000/tm6000-core.c b/drivers/media/usb/tm6000/tm6000-core.c index 23a1332d98e6..d3229aa45fcb 100644 --- a/drivers/media/usb/tm6000/tm6000-core.c +++ b/drivers/media/usb/tm6000/tm6000-core.c @@ -1,7 +1,7 @@ // SPDX-License-Identifier: GPL-2.0 // tm6000-core.c - driver for TM5600/TM6000/TM6010 USB video capture devices // -// Copyright (c) 2006-2007 Mauro Carvalho Chehab <mchehab@infradead.org> +// Copyright (c) 2006-2007 Mauro Carvalho Chehab <mchehab@kernel.org> // // Copyright (c) 2007 Michel Ludwig <michel.ludwig@gmail.com> // - DVB-T support diff --git a/drivers/media/usb/tm6000/tm6000-i2c.c b/drivers/media/usb/tm6000/tm6000-i2c.c index c9a62bbff27a..659b63febf85 100644 --- a/drivers/media/usb/tm6000/tm6000-i2c.c +++ b/drivers/media/usb/tm6000/tm6000-i2c.c @@ -1,7 +1,7 @@ // SPDX-License-Identifier: GPL-2.0 // tm6000-i2c.c - driver for TM5600/TM6000/TM6010 USB video capture devices // -// Copyright (c) 2006-2007 Mauro Carvalho Chehab <mchehab@infradead.org> +// Copyright (c) 2006-2007 Mauro Carvalho Chehab <mchehab@kernel.org> // // Copyright (c) 2007 Michel Ludwig <michel.ludwig@gmail.com> // - Fix SMBus Read Byte command diff --git a/drivers/media/usb/tm6000/tm6000-regs.h b/drivers/media/usb/tm6000/tm6000-regs.h index 21587fcf11e3..d10424673db9 100644 --- a/drivers/media/usb/tm6000/tm6000-regs.h +++ b/drivers/media/usb/tm6000/tm6000-regs.h @@ -2,7 +2,7 @@ * SPDX-License-Identifier: GPL-2.0 * tm6000-regs.h - driver for TM5600/TM6000/TM6010 USB video capture devices * - * Copyright (c) 2006-2007 Mauro Carvalho Chehab <mchehab@infradead.org> + * Copyright (c) 2006-2007 Mauro Carvalho Chehab <mchehab@kernel.org> */ /* diff --git a/drivers/media/usb/tm6000/tm6000-usb-isoc.h b/drivers/media/usb/tm6000/tm6000-usb-isoc.h index 5c615b0a7a46..b275dbce3a1b 100644 --- a/drivers/media/usb/tm6000/tm6000-usb-isoc.h +++ b/drivers/media/usb/tm6000/tm6000-usb-isoc.h @@ -2,7 +2,7 @@ * SPDX-License-Identifier: GPL-2.0 * tm6000-buf.c - driver for TM5600/TM6000/TM6010 USB video capture devices * - * Copyright (c) 2006-2007 Mauro Carvalho Chehab <mchehab@infradead.org> + * Copyright (c) 2006-2007 Mauro Carvalho Chehab <mchehab@kernel.org> */ #include <linux/videodev2.h> diff --git a/drivers/media/usb/tm6000/tm6000-video.c b/drivers/media/usb/tm6000/tm6000-video.c index b2399d4266da..aa85fe31c835 100644 --- a/drivers/media/usb/tm6000/tm6000-video.c +++ b/drivers/media/usb/tm6000/tm6000-video.c @@ -1,7 +1,7 @@ // SPDX-License-Identifier: GPL-2.0 // tm6000-video.c - driver for TM5600/TM6000/TM6010 USB video capture devices // -// Copyright (c) 2006-2007 Mauro Carvalho Chehab <mchehab@infradead.org> +// Copyright (c) 2006-2007 Mauro Carvalho Chehab <mchehab@kernel.org> // // Copyright (c) 2007 Michel Ludwig <michel.ludwig@gmail.com> // - Fixed module load/unload diff --git a/drivers/media/usb/tm6000/tm6000.h b/drivers/media/usb/tm6000/tm6000.h index e1e45770e28d..0864ed7314eb 100644 --- a/drivers/media/usb/tm6000/tm6000.h +++ b/drivers/media/usb/tm6000/tm6000.h @@ -2,7 +2,7 @@ * SPDX-License-Identifier: GPL-2.0 * tm6000.h - driver for TM5600/TM6000/TM6010 USB video capture devices * - * Copyright (c) 2006-2007 Mauro Carvalho Chehab <mchehab@infradead.org> + * Copyright (c) 2006-2007 Mauro Carvalho Chehab <mchehab@kernel.org> * * Copyright (c) 2007 Michel Ludwig <michel.ludwig@gmail.com> * - DVB-T support diff --git a/drivers/media/v4l2-core/v4l2-dev.c b/drivers/media/v4l2-core/v4l2-dev.c index 1d0b2208e8fb..c080dcc75393 100644 --- a/drivers/media/v4l2-core/v4l2-dev.c +++ b/drivers/media/v4l2-core/v4l2-dev.c @@ -10,7 +10,7 @@ * 2 of the License, or (at your option) any later version. * * Authors: Alan Cox, <alan@lxorguk.ukuu.org.uk> (version 1) - * Mauro Carvalho Chehab <mchehab@infradead.org> (version 2) + * Mauro Carvalho Chehab <mchehab@kernel.org> (version 2) * * Fixes: 20000516 Claudio Matsuoka <claudio@conectiva.com> * - Added procfs support @@ -1072,7 +1072,7 @@ static void __exit videodev_exit(void) subsys_initcall(videodev_init); module_exit(videodev_exit) -MODULE_AUTHOR("Alan Cox, Mauro Carvalho Chehab <mchehab@infradead.org>"); +MODULE_AUTHOR("Alan Cox, Mauro Carvalho Chehab <mchehab@kernel.org>"); MODULE_DESCRIPTION("Device registrar for Video4Linux drivers v2"); MODULE_LICENSE("GPL"); MODULE_ALIAS_CHARDEV_MAJOR(VIDEO_MAJOR); diff --git a/drivers/media/v4l2-core/v4l2-ioctl.c b/drivers/media/v4l2-core/v4l2-ioctl.c index f48c505550e0..de5d96dbe69e 100644 --- a/drivers/media/v4l2-core/v4l2-ioctl.c +++ b/drivers/media/v4l2-core/v4l2-ioctl.c @@ -9,7 +9,7 @@ * 2 of the License, or (at your option) any later version. * * Authors: Alan Cox, <alan@lxorguk.ukuu.org.uk> (version 1) - * Mauro Carvalho Chehab <mchehab@infradead.org> (version 2) + * Mauro Carvalho Chehab <mchehab@kernel.org> (version 2) */ #include <linux/mm.h> diff --git a/drivers/media/v4l2-core/videobuf-core.c b/drivers/media/v4l2-core/videobuf-core.c index 2b3981842b4b..7491b337002c 100644 --- a/drivers/media/v4l2-core/videobuf-core.c +++ b/drivers/media/v4l2-core/videobuf-core.c @@ -1,11 +1,11 @@ /* * generic helper functions for handling video4linux capture buffers * - * (c) 2007 Mauro Carvalho Chehab, <mchehab@infradead.org> + * (c) 2007 Mauro Carvalho Chehab, <mchehab@kernel.org> * * Highly based on video-buf written originally by: * (c) 2001,02 Gerd Knorr <kraxel@bytesex.org> - * (c) 2006 Mauro Carvalho Chehab, <mchehab@infradead.org> + * (c) 2006 Mauro Carvalho Chehab, <mchehab@kernel.org> * (c) 2006 Ted Walther and John Sokol * * This program is free software; you can redistribute it and/or modify @@ -38,7 +38,7 @@ static int debug; module_param(debug, int, 0644); MODULE_DESCRIPTION("helper module to manage video4linux buffers"); -MODULE_AUTHOR("Mauro Carvalho Chehab <mchehab@infradead.org>"); +MODULE_AUTHOR("Mauro Carvalho Chehab <mchehab@kernel.org>"); MODULE_LICENSE("GPL"); #define dprintk(level, fmt, arg...) \ diff --git a/drivers/media/v4l2-core/videobuf-dma-contig.c b/drivers/media/v4l2-core/videobuf-dma-contig.c index e02353e340dd..f46132504d88 100644 --- a/drivers/media/v4l2-core/videobuf-dma-contig.c +++ b/drivers/media/v4l2-core/videobuf-dma-contig.c @@ -7,7 +7,7 @@ * Copyright (c) 2008 Magnus Damm * * Based on videobuf-vmalloc.c, - * (c) 2007 Mauro Carvalho Chehab, <mchehab@infradead.org> + * (c) 2007 Mauro Carvalho Chehab, <mchehab@kernel.org> * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by diff --git a/drivers/media/v4l2-core/videobuf-dma-sg.c b/drivers/media/v4l2-core/videobuf-dma-sg.c index add2edb23eac..7770034aae28 100644 --- a/drivers/media/v4l2-core/videobuf-dma-sg.c +++ b/drivers/media/v4l2-core/videobuf-dma-sg.c @@ -6,11 +6,11 @@ * into PAGE_SIZE chunks). They also assume the driver does not need * to touch the video data. * - * (c) 2007 Mauro Carvalho Chehab, <mchehab@infradead.org> + * (c) 2007 Mauro Carvalho Chehab, <mchehab@kernel.org> * * Highly based on video-buf written originally by: * (c) 2001,02 Gerd Knorr <kraxel@bytesex.org> - * (c) 2006 Mauro Carvalho Chehab, <mchehab@infradead.org> + * (c) 2006 Mauro Carvalho Chehab, <mchehab@kernel.org> * (c) 2006 Ted Walther and John Sokol * * This program is free software; you can redistribute it and/or modify @@ -48,7 +48,7 @@ static int debug; module_param(debug, int, 0644); MODULE_DESCRIPTION("helper module to manage video4linux dma sg buffers"); -MODULE_AUTHOR("Mauro Carvalho Chehab <mchehab@infradead.org>"); +MODULE_AUTHOR("Mauro Carvalho Chehab <mchehab@kernel.org>"); MODULE_LICENSE("GPL"); #define dprintk(level, fmt, arg...) \ diff --git a/drivers/media/v4l2-core/videobuf-vmalloc.c b/drivers/media/v4l2-core/videobuf-vmalloc.c index 2ff7fcc77b11..45fe781aeeec 100644 --- a/drivers/media/v4l2-core/videobuf-vmalloc.c +++ b/drivers/media/v4l2-core/videobuf-vmalloc.c @@ -6,7 +6,7 @@ * into PAGE_SIZE chunks). They also assume the driver does not need * to touch the video data. * - * (c) 2007 Mauro Carvalho Chehab, <mchehab@infradead.org> + * (c) 2007 Mauro Carvalho Chehab, <mchehab@kernel.org> * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by @@ -41,7 +41,7 @@ static int debug; module_param(debug, int, 0644); MODULE_DESCRIPTION("helper module to manage video4linux vmalloc buffers"); -MODULE_AUTHOR("Mauro Carvalho Chehab <mchehab@infradead.org>"); +MODULE_AUTHOR("Mauro Carvalho Chehab <mchehab@kernel.org>"); MODULE_LICENSE("GPL"); #define dprintk(level, fmt, arg...) \ diff --git a/drivers/mtd/nand/onenand/omap2.c b/drivers/mtd/nand/onenand/omap2.c index 9c159f0dd9a6..321137158ff3 100644 --- a/drivers/mtd/nand/onenand/omap2.c +++ b/drivers/mtd/nand/onenand/omap2.c @@ -375,56 +375,42 @@ static int omap2_onenand_read_bufferram(struct mtd_info *mtd, int area, { struct omap2_onenand *c = container_of(mtd, struct omap2_onenand, mtd); struct onenand_chip *this = mtd->priv; - dma_addr_t dma_src, dma_dst; - int bram_offset; + struct device *dev = &c->pdev->dev; void *buf = (void *)buffer; + dma_addr_t dma_src, dma_dst; + int bram_offset, err; size_t xtra; - int ret; bram_offset = omap2_onenand_bufferram_offset(mtd, area) + area + offset; - if (bram_offset & 3 || (size_t)buf & 3 || count < 384) - goto out_copy; - - /* panic_write() may be in an interrupt context */ - if (in_interrupt() || oops_in_progress) + /* + * If the buffer address is not DMA-able, len is not long enough to make + * DMA transfers profitable or panic_write() may be in an interrupt + * context fallback to PIO mode. + */ + if (!virt_addr_valid(buf) || bram_offset & 3 || (size_t)buf & 3 || + count < 384 || in_interrupt() || oops_in_progress ) goto out_copy; - if (buf >= high_memory) { - struct page *p1; - - if (((size_t)buf & PAGE_MASK) != - ((size_t)(buf + count - 1) & PAGE_MASK)) - goto out_copy; - p1 = vmalloc_to_page(buf); - if (!p1) - goto out_copy; - buf = page_address(p1) + ((size_t)buf & ~PAGE_MASK); - } - xtra = count & 3; if (xtra) { count -= xtra; memcpy(buf + count, this->base + bram_offset + count, xtra); } + dma_dst = dma_map_single(dev, buf, count, DMA_FROM_DEVICE); dma_src = c->phys_base + bram_offset; - dma_dst = dma_map_single(&c->pdev->dev, buf, count, DMA_FROM_DEVICE); - if (dma_mapping_error(&c->pdev->dev, dma_dst)) { - dev_err(&c->pdev->dev, - "Couldn't DMA map a %d byte buffer\n", - count); - goto out_copy; - } - ret = omap2_onenand_dma_transfer(c, dma_src, dma_dst, count); - dma_unmap_single(&c->pdev->dev, dma_dst, count, DMA_FROM_DEVICE); - - if (ret) { - dev_err(&c->pdev->dev, "timeout waiting for DMA\n"); + if (dma_mapping_error(dev, dma_dst)) { + dev_err(dev, "Couldn't DMA map a %d byte buffer\n", count); goto out_copy; } - return 0; + err = omap2_onenand_dma_transfer(c, dma_src, dma_dst, count); + dma_unmap_single(dev, dma_dst, count, DMA_FROM_DEVICE); + if (!err) + return 0; + + dev_err(dev, "timeout waiting for DMA\n"); out_copy: memcpy(buf, this->base + bram_offset, count); @@ -437,49 +423,34 @@ static int omap2_onenand_write_bufferram(struct mtd_info *mtd, int area, { struct omap2_onenand *c = container_of(mtd, struct omap2_onenand, mtd); struct onenand_chip *this = mtd->priv; - dma_addr_t dma_src, dma_dst; - int bram_offset; + struct device *dev = &c->pdev->dev; void *buf = (void *)buffer; - int ret; + dma_addr_t dma_src, dma_dst; + int bram_offset, err; bram_offset = omap2_onenand_bufferram_offset(mtd, area) + area + offset; - if (bram_offset & 3 || (size_t)buf & 3 || count < 384) - goto out_copy; - - /* panic_write() may be in an interrupt context */ - if (in_interrupt() || oops_in_progress) + /* + * If the buffer address is not DMA-able, len is not long enough to make + * DMA transfers profitable or panic_write() may be in an interrupt + * context fallback to PIO mode. + */ + if (!virt_addr_valid(buf) || bram_offset & 3 || (size_t)buf & 3 || + count < 384 || in_interrupt() || oops_in_progress ) goto out_copy; - if (buf >= high_memory) { - struct page *p1; - - if (((size_t)buf & PAGE_MASK) != - ((size_t)(buf + count - 1) & PAGE_MASK)) - goto out_copy; - p1 = vmalloc_to_page(buf); - if (!p1) - goto out_copy; - buf = page_address(p1) + ((size_t)buf & ~PAGE_MASK); - } - - dma_src = dma_map_single(&c->pdev->dev, buf, count, DMA_TO_DEVICE); + dma_src = dma_map_single(dev, buf, count, DMA_TO_DEVICE); dma_dst = c->phys_base + bram_offset; - if (dma_mapping_error(&c->pdev->dev, dma_src)) { - dev_err(&c->pdev->dev, - "Couldn't DMA map a %d byte buffer\n", - count); - return -1; - } - - ret = omap2_onenand_dma_transfer(c, dma_src, dma_dst, count); - dma_unmap_single(&c->pdev->dev, dma_src, count, DMA_TO_DEVICE); - - if (ret) { - dev_err(&c->pdev->dev, "timeout waiting for DMA\n"); + if (dma_mapping_error(dev, dma_src)) { + dev_err(dev, "Couldn't DMA map a %d byte buffer\n", count); goto out_copy; } - return 0; + err = omap2_onenand_dma_transfer(c, dma_src, dma_dst, count); + dma_unmap_page(dev, dma_src, count, DMA_TO_DEVICE); + if (!err) + return 0; + + dev_err(dev, "timeout waiting for DMA\n"); out_copy: memcpy(this->base + bram_offset, buf, count); diff --git a/drivers/mtd/nand/raw/marvell_nand.c b/drivers/mtd/nand/raw/marvell_nand.c index 1d779a35ac8e..db5ec4e8bde9 100644 --- a/drivers/mtd/nand/raw/marvell_nand.c +++ b/drivers/mtd/nand/raw/marvell_nand.c @@ -1074,7 +1074,7 @@ static int marvell_nfc_hw_ecc_hmg_do_write_page(struct nand_chip *chip, return ret; ret = marvell_nfc_wait_op(chip, - chip->data_interface.timings.sdr.tPROG_max); + PSEC_TO_MSEC(chip->data_interface.timings.sdr.tPROG_max)); return ret; } @@ -1408,6 +1408,7 @@ marvell_nfc_hw_ecc_bch_write_chunk(struct nand_chip *chip, int chunk, struct marvell_nand_chip *marvell_nand = to_marvell_nand(chip); struct marvell_nfc *nfc = to_marvell_nfc(chip->controller); const struct marvell_hw_ecc_layout *lt = to_marvell_nand(chip)->layout; + u32 xtype; int ret; struct marvell_nfc_op nfc_op = { .ndcb[0] = NDCB0_CMD_TYPE(TYPE_WRITE) | NDCB0_LEN_OVRD, @@ -1423,7 +1424,12 @@ marvell_nfc_hw_ecc_bch_write_chunk(struct nand_chip *chip, int chunk, * last naked write. */ if (chunk == 0) { - nfc_op.ndcb[0] |= NDCB0_CMD_XTYPE(XTYPE_WRITE_DISPATCH) | + if (lt->nchunks == 1) + xtype = XTYPE_MONOLITHIC_RW; + else + xtype = XTYPE_WRITE_DISPATCH; + + nfc_op.ndcb[0] |= NDCB0_CMD_XTYPE(xtype) | NDCB0_ADDR_CYC(marvell_nand->addr_cyc) | NDCB0_CMD1(NAND_CMD_SEQIN); nfc_op.ndcb[1] |= NDCB1_ADDRS_PAGE(page); @@ -1494,7 +1500,7 @@ static int marvell_nfc_hw_ecc_bch_write_page(struct mtd_info *mtd, } ret = marvell_nfc_wait_op(chip, - chip->data_interface.timings.sdr.tPROG_max); + PSEC_TO_MSEC(chip->data_interface.timings.sdr.tPROG_max)); marvell_nfc_disable_hw_ecc(chip); diff --git a/drivers/mtd/nand/raw/nand_base.c b/drivers/mtd/nand/raw/nand_base.c index 72f3a89da513..f28c3a555861 100644 --- a/drivers/mtd/nand/raw/nand_base.c +++ b/drivers/mtd/nand/raw/nand_base.c @@ -706,12 +706,17 @@ static void nand_wait_status_ready(struct mtd_info *mtd, unsigned long timeo) */ int nand_soft_waitrdy(struct nand_chip *chip, unsigned long timeout_ms) { + const struct nand_sdr_timings *timings; u8 status = 0; int ret; if (!chip->exec_op) return -ENOTSUPP; + /* Wait tWB before polling the STATUS reg. */ + timings = nand_get_sdr_timings(&chip->data_interface); + ndelay(PSEC_TO_NSEC(timings->tWB_max)); + ret = nand_status_op(chip, NULL); if (ret) return ret; diff --git a/drivers/net/bonding/bond_alb.c b/drivers/net/bonding/bond_alb.c index 1ed9529e7bd1..5eb0df2e5464 100644 --- a/drivers/net/bonding/bond_alb.c +++ b/drivers/net/bonding/bond_alb.c @@ -450,7 +450,7 @@ static void rlb_update_client(struct rlb_client_info *client_info) { int i; - if (!client_info->slave) + if (!client_info->slave || !is_valid_ether_addr(client_info->mac_dst)) return; for (i = 0; i < RLB_ARP_BURST_SIZE; i++) { @@ -943,6 +943,10 @@ static void alb_send_lp_vid(struct slave *slave, u8 mac_addr[], skb->priority = TC_PRIO_CONTROL; skb->dev = slave->dev; + netdev_dbg(slave->bond->dev, + "Send learning packet: dev %s mac %pM vlan %d\n", + slave->dev->name, mac_addr, vid); + if (vid) __vlan_hwaccel_put_tag(skb, vlan_proto, vid); @@ -965,14 +969,13 @@ static int alb_upper_dev_walk(struct net_device *upper, void *_data) u8 *mac_addr = data->mac_addr; struct bond_vlan_tag *tags; - if (is_vlan_dev(upper) && vlan_get_encap_level(upper) == 0) { - if (strict_match && - ether_addr_equal_64bits(mac_addr, - upper->dev_addr)) { + if (is_vlan_dev(upper) && + bond->nest_level == vlan_get_encap_level(upper) - 1) { + if (upper->addr_assign_type == NET_ADDR_STOLEN) { alb_send_lp_vid(slave, mac_addr, vlan_dev_vlan_proto(upper), vlan_dev_vlan_id(upper)); - } else if (!strict_match) { + } else { alb_send_lp_vid(slave, upper->dev_addr, vlan_dev_vlan_proto(upper), vlan_dev_vlan_id(upper)); diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index 718e4914e3a0..1f1e97b26f95 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -1738,6 +1738,8 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev, if (bond_mode_uses_xmit_hash(bond)) bond_update_slave_arr(bond, NULL); + bond->nest_level = dev_get_nest_level(bond_dev); + netdev_info(bond_dev, "Enslaving %s as %s interface with %s link\n", slave_dev->name, bond_is_active_slave(new_slave) ? "an active" : "a backup", diff --git a/drivers/net/can/dev.c b/drivers/net/can/dev.c index b1779566c5bb..3c71f1cb205f 100644 --- a/drivers/net/can/dev.c +++ b/drivers/net/can/dev.c @@ -605,7 +605,7 @@ void can_bus_off(struct net_device *dev) { struct can_priv *priv = netdev_priv(dev); - netdev_dbg(dev, "bus-off\n"); + netdev_info(dev, "bus-off\n"); netif_carrier_off(dev); diff --git a/drivers/net/can/flexcan.c b/drivers/net/can/flexcan.c index 634c51e6b8ae..d53a45bf2a72 100644 --- a/drivers/net/can/flexcan.c +++ b/drivers/net/can/flexcan.c @@ -200,6 +200,7 @@ #define FLEXCAN_QUIRK_DISABLE_MECR BIT(4) /* Disable Memory error detection */ #define FLEXCAN_QUIRK_USE_OFF_TIMESTAMP BIT(5) /* Use timestamp based offloading */ #define FLEXCAN_QUIRK_BROKEN_PERR_STATE BIT(6) /* No interrupt for error passive */ +#define FLEXCAN_QUIRK_DEFAULT_BIG_ENDIAN BIT(7) /* default to BE register access */ /* Structure of the message buffer */ struct flexcan_mb { @@ -288,6 +289,12 @@ struct flexcan_priv { static const struct flexcan_devtype_data fsl_p1010_devtype_data = { .quirks = FLEXCAN_QUIRK_BROKEN_WERR_STATE | + FLEXCAN_QUIRK_BROKEN_PERR_STATE | + FLEXCAN_QUIRK_DEFAULT_BIG_ENDIAN, +}; + +static const struct flexcan_devtype_data fsl_imx25_devtype_data = { + .quirks = FLEXCAN_QUIRK_BROKEN_WERR_STATE | FLEXCAN_QUIRK_BROKEN_PERR_STATE, }; @@ -1251,9 +1258,9 @@ static void unregister_flexcandev(struct net_device *dev) static const struct of_device_id flexcan_of_match[] = { { .compatible = "fsl,imx6q-flexcan", .data = &fsl_imx6q_devtype_data, }, { .compatible = "fsl,imx28-flexcan", .data = &fsl_imx28_devtype_data, }, - { .compatible = "fsl,imx53-flexcan", .data = &fsl_p1010_devtype_data, }, - { .compatible = "fsl,imx35-flexcan", .data = &fsl_p1010_devtype_data, }, - { .compatible = "fsl,imx25-flexcan", .data = &fsl_p1010_devtype_data, }, + { .compatible = "fsl,imx53-flexcan", .data = &fsl_imx25_devtype_data, }, + { .compatible = "fsl,imx35-flexcan", .data = &fsl_imx25_devtype_data, }, + { .compatible = "fsl,imx25-flexcan", .data = &fsl_imx25_devtype_data, }, { .compatible = "fsl,p1010-flexcan", .data = &fsl_p1010_devtype_data, }, { .compatible = "fsl,vf610-flexcan", .data = &fsl_vf610_devtype_data, }, { .compatible = "fsl,ls1021ar2-flexcan", .data = &fsl_ls1021a_r2_devtype_data, }, @@ -1337,18 +1344,13 @@ static int flexcan_probe(struct platform_device *pdev) priv = netdev_priv(dev); - if (of_property_read_bool(pdev->dev.of_node, "big-endian")) { + if (of_property_read_bool(pdev->dev.of_node, "big-endian") || + devtype_data->quirks & FLEXCAN_QUIRK_DEFAULT_BIG_ENDIAN) { priv->read = flexcan_read_be; priv->write = flexcan_write_be; } else { - if (of_device_is_compatible(pdev->dev.of_node, - "fsl,p1010-flexcan")) { - priv->read = flexcan_read_be; - priv->write = flexcan_write_be; - } else { - priv->read = flexcan_read_le; - priv->write = flexcan_write_le; - } + priv->read = flexcan_read_le; + priv->write = flexcan_write_le; } priv->can.clock.freq = clock_freq; diff --git a/drivers/net/can/spi/hi311x.c b/drivers/net/can/spi/hi311x.c index 5590c559a8ca..53e320c92a8b 100644 --- a/drivers/net/can/spi/hi311x.c +++ b/drivers/net/can/spi/hi311x.c @@ -91,6 +91,7 @@ #define HI3110_STAT_BUSOFF BIT(2) #define HI3110_STAT_ERRP BIT(3) #define HI3110_STAT_ERRW BIT(4) +#define HI3110_STAT_TXMTY BIT(7) #define HI3110_BTR0_SJW_SHIFT 6 #define HI3110_BTR0_BRP_SHIFT 0 @@ -427,8 +428,10 @@ static int hi3110_get_berr_counter(const struct net_device *net, struct hi3110_priv *priv = netdev_priv(net); struct spi_device *spi = priv->spi; + mutex_lock(&priv->hi3110_lock); bec->txerr = hi3110_read(spi, HI3110_READ_TEC); bec->rxerr = hi3110_read(spi, HI3110_READ_REC); + mutex_unlock(&priv->hi3110_lock); return 0; } @@ -735,10 +738,7 @@ static irqreturn_t hi3110_can_ist(int irq, void *dev_id) } } - if (intf == 0) - break; - - if (intf & HI3110_INT_TXCPLT) { + if (priv->tx_len && statf & HI3110_STAT_TXMTY) { net->stats.tx_packets++; net->stats.tx_bytes += priv->tx_len - 1; can_led_event(net, CAN_LED_EVENT_TX); @@ -748,6 +748,9 @@ static irqreturn_t hi3110_can_ist(int irq, void *dev_id) } netif_wake_queue(net); } + + if (intf == 0) + break; } mutex_unlock(&priv->hi3110_lock); return IRQ_HANDLED; diff --git a/drivers/net/can/usb/kvaser_usb.c b/drivers/net/can/usb/kvaser_usb.c index 63587b8e6825..daed57d3d209 100644 --- a/drivers/net/can/usb/kvaser_usb.c +++ b/drivers/net/can/usb/kvaser_usb.c @@ -1179,7 +1179,7 @@ static void kvaser_usb_rx_can_msg(const struct kvaser_usb *dev, skb = alloc_can_skb(priv->netdev, &cf); if (!skb) { - stats->tx_dropped++; + stats->rx_dropped++; return; } diff --git a/drivers/net/dsa/mv88e6xxx/chip.c b/drivers/net/dsa/mv88e6xxx/chip.c index 3d2091099f7f..5b4374f21d76 100644 --- a/drivers/net/dsa/mv88e6xxx/chip.c +++ b/drivers/net/dsa/mv88e6xxx/chip.c @@ -3370,6 +3370,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .num_internal_phys = 5, .max_vid = 4095, .port_base_addr = 0x10, + .phy_base_addr = 0x0, .global1_addr = 0x1b, .global2_addr = 0x1c, .age_time_coeff = 15000, @@ -3391,6 +3392,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .num_internal_phys = 0, .max_vid = 4095, .port_base_addr = 0x10, + .phy_base_addr = 0x0, .global1_addr = 0x1b, .global2_addr = 0x1c, .age_time_coeff = 15000, @@ -3410,6 +3412,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .num_internal_phys = 8, .max_vid = 4095, .port_base_addr = 0x10, + .phy_base_addr = 0x0, .global1_addr = 0x1b, .global2_addr = 0x1c, .age_time_coeff = 15000, @@ -3431,6 +3434,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .num_internal_phys = 5, .max_vid = 4095, .port_base_addr = 0x10, + .phy_base_addr = 0x0, .global1_addr = 0x1b, .global2_addr = 0x1c, .age_time_coeff = 15000, @@ -3452,6 +3456,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .num_internal_phys = 0, .max_vid = 4095, .port_base_addr = 0x10, + .phy_base_addr = 0x0, .global1_addr = 0x1b, .global2_addr = 0x1c, .age_time_coeff = 15000, @@ -3472,6 +3477,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .num_gpio = 11, .max_vid = 4095, .port_base_addr = 0x10, + .phy_base_addr = 0x10, .global1_addr = 0x1b, .global2_addr = 0x1c, .age_time_coeff = 3750, @@ -3493,6 +3499,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .num_internal_phys = 5, .max_vid = 4095, .port_base_addr = 0x10, + .phy_base_addr = 0x0, .global1_addr = 0x1b, .global2_addr = 0x1c, .age_time_coeff = 15000, @@ -3514,6 +3521,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .num_internal_phys = 0, .max_vid = 4095, .port_base_addr = 0x10, + .phy_base_addr = 0x0, .global1_addr = 0x1b, .global2_addr = 0x1c, .age_time_coeff = 15000, @@ -3535,6 +3543,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .num_internal_phys = 5, .max_vid = 4095, .port_base_addr = 0x10, + .phy_base_addr = 0x0, .global1_addr = 0x1b, .global2_addr = 0x1c, .age_time_coeff = 15000, @@ -3557,6 +3566,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .num_gpio = 15, .max_vid = 4095, .port_base_addr = 0x10, + .phy_base_addr = 0x0, .global1_addr = 0x1b, .global2_addr = 0x1c, .age_time_coeff = 15000, @@ -3578,6 +3588,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .num_internal_phys = 5, .max_vid = 4095, .port_base_addr = 0x10, + .phy_base_addr = 0x0, .global1_addr = 0x1b, .global2_addr = 0x1c, .age_time_coeff = 15000, @@ -3600,6 +3611,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .num_gpio = 15, .max_vid = 4095, .port_base_addr = 0x10, + .phy_base_addr = 0x0, .global1_addr = 0x1b, .global2_addr = 0x1c, .age_time_coeff = 15000, @@ -3621,6 +3633,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .num_internal_phys = 0, .max_vid = 4095, .port_base_addr = 0x10, + .phy_base_addr = 0x0, .global1_addr = 0x1b, .global2_addr = 0x1c, .age_time_coeff = 15000, @@ -3641,6 +3654,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .num_gpio = 16, .max_vid = 8191, .port_base_addr = 0x0, + .phy_base_addr = 0x0, .global1_addr = 0x1b, .global2_addr = 0x1c, .tag_protocol = DSA_TAG_PROTO_DSA, @@ -3663,6 +3677,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .num_gpio = 16, .max_vid = 8191, .port_base_addr = 0x0, + .phy_base_addr = 0x0, .global1_addr = 0x1b, .global2_addr = 0x1c, .age_time_coeff = 3750, @@ -3684,6 +3699,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .num_internal_phys = 11, .max_vid = 8191, .port_base_addr = 0x0, + .phy_base_addr = 0x0, .global1_addr = 0x1b, .global2_addr = 0x1c, .age_time_coeff = 3750, @@ -3707,6 +3723,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .num_gpio = 15, .max_vid = 4095, .port_base_addr = 0x10, + .phy_base_addr = 0x0, .global1_addr = 0x1b, .global2_addr = 0x1c, .age_time_coeff = 15000, @@ -3730,6 +3747,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .num_gpio = 16, .max_vid = 8191, .port_base_addr = 0x0, + .phy_base_addr = 0x0, .global1_addr = 0x1b, .global2_addr = 0x1c, .age_time_coeff = 3750, @@ -3753,6 +3771,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .num_gpio = 15, .max_vid = 4095, .port_base_addr = 0x10, + .phy_base_addr = 0x0, .global1_addr = 0x1b, .global2_addr = 0x1c, .age_time_coeff = 15000, @@ -3776,6 +3795,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .num_gpio = 15, .max_vid = 4095, .port_base_addr = 0x10, + .phy_base_addr = 0x0, .global1_addr = 0x1b, .global2_addr = 0x1c, .age_time_coeff = 15000, @@ -3798,6 +3818,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .num_gpio = 11, .max_vid = 4095, .port_base_addr = 0x10, + .phy_base_addr = 0x10, .global1_addr = 0x1b, .global2_addr = 0x1c, .age_time_coeff = 3750, @@ -3820,6 +3841,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .num_internal_phys = 5, .max_vid = 4095, .port_base_addr = 0x10, + .phy_base_addr = 0x0, .global1_addr = 0x1b, .global2_addr = 0x1c, .age_time_coeff = 15000, @@ -3841,6 +3863,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .num_internal_phys = 5, .max_vid = 4095, .port_base_addr = 0x10, + .phy_base_addr = 0x0, .global1_addr = 0x1b, .global2_addr = 0x1c, .age_time_coeff = 15000, @@ -3863,6 +3886,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .num_gpio = 15, .max_vid = 4095, .port_base_addr = 0x10, + .phy_base_addr = 0x0, .global1_addr = 0x1b, .global2_addr = 0x1c, .age_time_coeff = 15000, @@ -3885,6 +3909,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .num_gpio = 16, .max_vid = 8191, .port_base_addr = 0x0, + .phy_base_addr = 0x0, .global1_addr = 0x1b, .global2_addr = 0x1c, .age_time_coeff = 3750, @@ -3907,6 +3932,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .num_gpio = 16, .max_vid = 8191, .port_base_addr = 0x0, + .phy_base_addr = 0x0, .global1_addr = 0x1b, .global2_addr = 0x1c, .age_time_coeff = 3750, diff --git a/drivers/net/dsa/mv88e6xxx/chip.h b/drivers/net/dsa/mv88e6xxx/chip.h index 80490f66bc06..12b7f4649b25 100644 --- a/drivers/net/dsa/mv88e6xxx/chip.h +++ b/drivers/net/dsa/mv88e6xxx/chip.h @@ -114,6 +114,7 @@ struct mv88e6xxx_info { unsigned int num_gpio; unsigned int max_vid; unsigned int port_base_addr; + unsigned int phy_base_addr; unsigned int global1_addr; unsigned int global2_addr; unsigned int age_time_coeff; diff --git a/drivers/net/dsa/mv88e6xxx/global2.c b/drivers/net/dsa/mv88e6xxx/global2.c index 0ce627fded48..8d22d66d84b7 100644 --- a/drivers/net/dsa/mv88e6xxx/global2.c +++ b/drivers/net/dsa/mv88e6xxx/global2.c @@ -1118,7 +1118,7 @@ int mv88e6xxx_g2_irq_mdio_setup(struct mv88e6xxx_chip *chip, err = irq; goto out; } - bus->irq[chip->info->port_base_addr + phy] = irq; + bus->irq[chip->info->phy_base_addr + phy] = irq; } return 0; out: diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_nic.c b/drivers/net/ethernet/aquantia/atlantic/aq_nic.c index 32f6d2e24d66..1a1a6380c128 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_nic.c +++ b/drivers/net/ethernet/aquantia/atlantic/aq_nic.c @@ -95,6 +95,7 @@ void aq_nic_cfg_start(struct aq_nic_s *self) /*rss rings */ cfg->vecs = min(cfg->aq_hw_caps->vecs, AQ_CFG_VECS_DEF); cfg->vecs = min(cfg->vecs, num_online_cpus()); + cfg->vecs = min(cfg->vecs, self->irqvecs); /* cfg->vecs should be power of 2 for RSS */ if (cfg->vecs >= 8U) cfg->vecs = 8U; @@ -246,6 +247,8 @@ void aq_nic_ndev_init(struct aq_nic_s *self) self->ndev->hw_features |= aq_hw_caps->hw_features; self->ndev->features = aq_hw_caps->hw_features; + self->ndev->vlan_features |= NETIF_F_HW_CSUM | NETIF_F_RXCSUM | + NETIF_F_RXHASH | NETIF_F_SG | NETIF_F_LRO; self->ndev->priv_flags = aq_hw_caps->hw_priv_flags; self->ndev->priv_flags |= IFF_LIVE_ADDR_CHANGE; diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_nic.h b/drivers/net/ethernet/aquantia/atlantic/aq_nic.h index 219b550d1665..faa533a0ec47 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_nic.h +++ b/drivers/net/ethernet/aquantia/atlantic/aq_nic.h @@ -80,6 +80,7 @@ struct aq_nic_s { struct pci_dev *pdev; unsigned int msix_entry_mask; + u32 irqvecs; }; static inline struct device *aq_nic_get_dev(struct aq_nic_s *self) diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_pci_func.c b/drivers/net/ethernet/aquantia/atlantic/aq_pci_func.c index ecc6306f940f..a50e08bb4748 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_pci_func.c +++ b/drivers/net/ethernet/aquantia/atlantic/aq_pci_func.c @@ -267,16 +267,16 @@ static int aq_pci_probe(struct pci_dev *pdev, numvecs = min(numvecs, num_online_cpus()); /*enable interrupts */ #if !AQ_CFG_FORCE_LEGACY_INT - err = pci_alloc_irq_vectors(self->pdev, numvecs, numvecs, - PCI_IRQ_MSIX); - - if (err < 0) { - err = pci_alloc_irq_vectors(self->pdev, 1, 1, - PCI_IRQ_MSI | PCI_IRQ_LEGACY); - if (err < 0) - goto err_hwinit; + numvecs = pci_alloc_irq_vectors(self->pdev, 1, numvecs, + PCI_IRQ_MSIX | PCI_IRQ_MSI | + PCI_IRQ_LEGACY); + + if (numvecs < 0) { + err = numvecs; + goto err_hwinit; } #endif + self->irqvecs = numvecs; /* net device init */ aq_nic_cfg_start(self); @@ -298,9 +298,9 @@ err_free_aq_hw: kfree(self->aq_hw); err_ioremap: free_netdev(ndev); -err_pci_func: - pci_release_regions(pdev); err_ndev: + pci_release_regions(pdev); +err_pci_func: pci_disable_device(pdev); return err; } diff --git a/drivers/net/ethernet/broadcom/bcmsysport.c b/drivers/net/ethernet/broadcom/bcmsysport.c index f9a3c1a76d5d..f33b25fbca63 100644 --- a/drivers/net/ethernet/broadcom/bcmsysport.c +++ b/drivers/net/ethernet/broadcom/bcmsysport.c @@ -2144,14 +2144,21 @@ static const struct net_device_ops bcm_sysport_netdev_ops = { .ndo_select_queue = bcm_sysport_select_queue, }; -static int bcm_sysport_map_queues(struct net_device *dev, +static int bcm_sysport_map_queues(struct notifier_block *nb, struct dsa_notifier_register_info *info) { - struct bcm_sysport_priv *priv = netdev_priv(dev); struct bcm_sysport_tx_ring *ring; + struct bcm_sysport_priv *priv; struct net_device *slave_dev; unsigned int num_tx_queues; unsigned int q, start, port; + struct net_device *dev; + + priv = container_of(nb, struct bcm_sysport_priv, dsa_notifier); + if (priv->netdev != info->master) + return 0; + + dev = info->master; /* We can't be setting up queue inspection for non directly attached * switches @@ -2174,11 +2181,12 @@ static int bcm_sysport_map_queues(struct net_device *dev, if (priv->is_lite) netif_set_real_num_tx_queues(slave_dev, slave_dev->num_tx_queues / 2); + num_tx_queues = slave_dev->real_num_tx_queues; if (priv->per_port_num_tx_queues && priv->per_port_num_tx_queues != num_tx_queues) - netdev_warn(slave_dev, "asymetric number of per-port queues\n"); + netdev_warn(slave_dev, "asymmetric number of per-port queues\n"); priv->per_port_num_tx_queues = num_tx_queues; @@ -2201,7 +2209,7 @@ static int bcm_sysport_map_queues(struct net_device *dev, return 0; } -static int bcm_sysport_dsa_notifier(struct notifier_block *unused, +static int bcm_sysport_dsa_notifier(struct notifier_block *nb, unsigned long event, void *ptr) { struct dsa_notifier_register_info *info; @@ -2211,7 +2219,7 @@ static int bcm_sysport_dsa_notifier(struct notifier_block *unused, info = ptr; - return notifier_from_errno(bcm_sysport_map_queues(info->master, info)); + return notifier_from_errno(bcm_sysport_map_queues(nb, info)); } #define REV_FMT "v%2x.%02x" diff --git a/drivers/net/ethernet/broadcom/tg3.c b/drivers/net/ethernet/broadcom/tg3.c index 08bbb639be1a..9f59b1270a7c 100644 --- a/drivers/net/ethernet/broadcom/tg3.c +++ b/drivers/net/ethernet/broadcom/tg3.c @@ -8733,14 +8733,15 @@ static void tg3_free_consistent(struct tg3 *tp) tg3_mem_rx_release(tp); tg3_mem_tx_release(tp); - /* Protect tg3_get_stats64() from reading freed tp->hw_stats. */ - tg3_full_lock(tp, 0); + /* tp->hw_stats can be referenced safely: + * 1. under rtnl_lock + * 2. or under tp->lock if TG3_FLAG_INIT_COMPLETE is set. + */ if (tp->hw_stats) { dma_free_coherent(&tp->pdev->dev, sizeof(struct tg3_hw_stats), tp->hw_stats, tp->stats_mapping); tp->hw_stats = NULL; } - tg3_full_unlock(tp); } /* @@ -14178,7 +14179,7 @@ static void tg3_get_stats64(struct net_device *dev, struct tg3 *tp = netdev_priv(dev); spin_lock_bh(&tp->lock); - if (!tp->hw_stats) { + if (!tp->hw_stats || !tg3_flag(tp, INIT_COMPLETE)) { *stats = tp->net_stats_prev; spin_unlock_bh(&tp->lock); return; diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c index 24d2865b8806..005283c7cdfe 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c @@ -3433,8 +3433,8 @@ static int adap_config_hma(struct adapter *adapter) sgl = adapter->hma.sgt->sgl; node = dev_to_node(adapter->pdev_dev); for_each_sg(sgl, iter, sgt->orig_nents, i) { - newpage = alloc_pages_node(node, __GFP_NOWARN | GFP_KERNEL, - page_order); + newpage = alloc_pages_node(node, __GFP_NOWARN | GFP_KERNEL | + __GFP_ZERO, page_order); if (!newpage) { dev_err(adapter->pdev_dev, "Not enough memory for HMA page allocation\n"); @@ -5474,6 +5474,7 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent) } spin_lock_init(&adapter->mbox_lock); INIT_LIST_HEAD(&adapter->mlist.list); + adapter->mbox_log->size = T4_OS_LOG_MBOX_CMDS; pci_set_drvdata(pdev, adapter); if (func != ent->driver_data) { @@ -5508,8 +5509,6 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent) goto out_free_adapter; } - adapter->mbox_log->size = T4_OS_LOG_MBOX_CMDS; - /* PCI device has been enabled */ adapter->flags |= DEV_ENABLED; memset(adapter->chan_map, 0xff, sizeof(adapter->chan_map)); diff --git a/drivers/net/ethernet/freescale/ucc_geth_ethtool.c b/drivers/net/ethernet/freescale/ucc_geth_ethtool.c index 4df282ed22c7..0beee2cc2ddd 100644 --- a/drivers/net/ethernet/freescale/ucc_geth_ethtool.c +++ b/drivers/net/ethernet/freescale/ucc_geth_ethtool.c @@ -61,7 +61,7 @@ static const char hw_stat_gstrings[][ETH_GSTRING_LEN] = { static const char tx_fw_stat_gstrings[][ETH_GSTRING_LEN] = { "tx-single-collision", "tx-multiple-collision", - "tx-late-collsion", + "tx-late-collision", "tx-aborted-frames", "tx-lost-frames", "tx-carrier-sense-errors", diff --git a/drivers/net/ethernet/intel/ice/ice_controlq.c b/drivers/net/ethernet/intel/ice/ice_controlq.c index 5909a4407e38..7c511f144ed6 100644 --- a/drivers/net/ethernet/intel/ice/ice_controlq.c +++ b/drivers/net/ethernet/intel/ice/ice_controlq.c @@ -1014,10 +1014,10 @@ ice_clean_rq_elem(struct ice_hw *hw, struct ice_ctl_q_info *cq, desc = ICE_CTL_Q_DESC(cq->rq, ntc); desc_idx = ntc; + cq->rq_last_status = (enum ice_aq_err)le16_to_cpu(desc->retval); flags = le16_to_cpu(desc->flags); if (flags & ICE_AQ_FLAG_ERR) { ret_code = ICE_ERR_AQ_ERROR; - cq->rq_last_status = (enum ice_aq_err)le16_to_cpu(desc->retval); ice_debug(hw, ICE_DBG_AQ_MSG, "Control Receive Queue Event received with error 0x%x\n", cq->rq_last_status); diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_ipsec.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_ipsec.c index 68af127987bc..cead23e3db0c 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_ipsec.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_ipsec.c @@ -943,8 +943,8 @@ err2: kfree(ipsec->ip_tbl); kfree(ipsec->rx_tbl); kfree(ipsec->tx_tbl); + kfree(ipsec); err1: - kfree(adapter->ipsec); netdev_err(adapter->netdev, "Unable to allocate memory for SA tables"); } diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c index 3123267dfba9..9592f3e3e42e 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c @@ -3427,6 +3427,9 @@ static s32 ixgbe_reset_hw_X550em(struct ixgbe_hw *hw) hw->phy.sfp_setup_needed = false; } + if (status == IXGBE_ERR_SFP_NOT_SUPPORTED) + return status; + /* Reset PHY */ if (!hw->phy.reset_disable && hw->phy.ops.reset) hw->phy.ops.reset(hw); diff --git a/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c b/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c index e3d04f226d57..850f8af95e49 100644 --- a/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c +++ b/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c @@ -4137,7 +4137,7 @@ out_drop: return NETDEV_TX_OK; } -static int ixgbevf_xmit_frame(struct sk_buff *skb, struct net_device *netdev) +static netdev_tx_t ixgbevf_xmit_frame(struct sk_buff *skb, struct net_device *netdev) { struct ixgbevf_adapter *adapter = netdev_priv(netdev); struct ixgbevf_ring *tx_ring; diff --git a/drivers/net/ethernet/marvell/mvpp2.c b/drivers/net/ethernet/marvell/mvpp2.c index 4202f9b5b966..6f410235987c 100644 --- a/drivers/net/ethernet/marvell/mvpp2.c +++ b/drivers/net/ethernet/marvell/mvpp2.c @@ -942,6 +942,7 @@ struct mvpp2 { struct clk *pp_clk; struct clk *gop_clk; struct clk *mg_clk; + struct clk *mg_core_clk; struct clk *axi_clk; /* List of pointers to port structures */ @@ -8768,18 +8769,27 @@ static int mvpp2_probe(struct platform_device *pdev) err = clk_prepare_enable(priv->mg_clk); if (err < 0) goto err_gop_clk; + + priv->mg_core_clk = devm_clk_get(&pdev->dev, "mg_core_clk"); + if (IS_ERR(priv->mg_core_clk)) { + priv->mg_core_clk = NULL; + } else { + err = clk_prepare_enable(priv->mg_core_clk); + if (err < 0) + goto err_mg_clk; + } } priv->axi_clk = devm_clk_get(&pdev->dev, "axi_clk"); if (IS_ERR(priv->axi_clk)) { err = PTR_ERR(priv->axi_clk); if (err == -EPROBE_DEFER) - goto err_gop_clk; + goto err_mg_core_clk; priv->axi_clk = NULL; } else { err = clk_prepare_enable(priv->axi_clk); if (err < 0) - goto err_gop_clk; + goto err_mg_core_clk; } /* Get system's tclk rate */ @@ -8793,7 +8803,7 @@ static int mvpp2_probe(struct platform_device *pdev) if (priv->hw_version == MVPP22) { err = dma_set_mask(&pdev->dev, MVPP2_DESC_DMA_MASK); if (err) - goto err_mg_clk; + goto err_axi_clk; /* Sadly, the BM pools all share the same register to * store the high 32 bits of their address. So they * must all have the same high 32 bits, which forces @@ -8801,14 +8811,14 @@ static int mvpp2_probe(struct platform_device *pdev) */ err = dma_set_coherent_mask(&pdev->dev, DMA_BIT_MASK(32)); if (err) - goto err_mg_clk; + goto err_axi_clk; } /* Initialize network controller */ err = mvpp2_init(pdev, priv); if (err < 0) { dev_err(&pdev->dev, "failed to initialize controller\n"); - goto err_mg_clk; + goto err_axi_clk; } /* Initialize ports */ @@ -8821,7 +8831,7 @@ static int mvpp2_probe(struct platform_device *pdev) if (priv->port_count == 0) { dev_err(&pdev->dev, "no ports enabled\n"); err = -ENODEV; - goto err_mg_clk; + goto err_axi_clk; } /* Statistics must be gathered regularly because some of them (like @@ -8849,8 +8859,13 @@ err_port_probe: mvpp2_port_remove(priv->port_list[i]); i++; } -err_mg_clk: +err_axi_clk: clk_disable_unprepare(priv->axi_clk); + +err_mg_core_clk: + if (priv->hw_version == MVPP22) + clk_disable_unprepare(priv->mg_core_clk); +err_mg_clk: if (priv->hw_version == MVPP22) clk_disable_unprepare(priv->mg_clk); err_gop_clk: @@ -8897,6 +8912,7 @@ static int mvpp2_remove(struct platform_device *pdev) return 0; clk_disable_unprepare(priv->axi_clk); + clk_disable_unprepare(priv->mg_core_clk); clk_disable_unprepare(priv->mg_clk); clk_disable_unprepare(priv->pp_clk); clk_disable_unprepare(priv->gop_clk); diff --git a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c index a30a2e95d13f..f11b45001cad 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c @@ -1027,6 +1027,22 @@ static int mlx4_en_set_coalesce(struct net_device *dev, if (!coal->tx_max_coalesced_frames_irq) return -EINVAL; + if (coal->tx_coalesce_usecs > MLX4_EN_MAX_COAL_TIME || + coal->rx_coalesce_usecs > MLX4_EN_MAX_COAL_TIME || + coal->rx_coalesce_usecs_low > MLX4_EN_MAX_COAL_TIME || + coal->rx_coalesce_usecs_high > MLX4_EN_MAX_COAL_TIME) { + netdev_info(dev, "%s: maximum coalesce time supported is %d usecs\n", + __func__, MLX4_EN_MAX_COAL_TIME); + return -ERANGE; + } + + if (coal->tx_max_coalesced_frames > MLX4_EN_MAX_COAL_PKTS || + coal->rx_max_coalesced_frames > MLX4_EN_MAX_COAL_PKTS) { + netdev_info(dev, "%s: maximum coalesced frames supported is %d\n", + __func__, MLX4_EN_MAX_COAL_PKTS); + return -ERANGE; + } + priv->rx_frames = (coal->rx_max_coalesced_frames == MLX4_EN_AUTO_CONF) ? MLX4_EN_RX_COAL_TARGET : diff --git a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c index e0adac4a9a19..9670b33fc9b1 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c @@ -3324,12 +3324,11 @@ int mlx4_en_init_netdev(struct mlx4_en_dev *mdev, int port, MAX_TX_RINGS, GFP_KERNEL); if (!priv->tx_ring[t]) { err = -ENOMEM; - goto err_free_tx; + goto out; } priv->tx_cq[t] = kzalloc(sizeof(struct mlx4_en_cq *) * MAX_TX_RINGS, GFP_KERNEL); if (!priv->tx_cq[t]) { - kfree(priv->tx_ring[t]); err = -ENOMEM; goto out; } @@ -3582,11 +3581,6 @@ int mlx4_en_init_netdev(struct mlx4_en_dev *mdev, int port, return 0; -err_free_tx: - while (t--) { - kfree(priv->tx_ring[t]); - kfree(priv->tx_cq[t]); - } out: mlx4_en_destroy_netdev(dev); return err; diff --git a/drivers/net/ethernet/mellanox/mlx4/main.c b/drivers/net/ethernet/mellanox/mlx4/main.c index bfef69235d71..211578ffc70d 100644 --- a/drivers/net/ethernet/mellanox/mlx4/main.c +++ b/drivers/net/ethernet/mellanox/mlx4/main.c @@ -1317,7 +1317,7 @@ static int mlx4_mf_unbond(struct mlx4_dev *dev) ret = mlx4_unbond_fs_rules(dev); if (ret) - mlx4_warn(dev, "multifunction unbond for flow rules failedi (%d)\n", ret); + mlx4_warn(dev, "multifunction unbond for flow rules failed (%d)\n", ret); ret1 = mlx4_unbond_mac_table(dev); if (ret1) { mlx4_warn(dev, "multifunction unbond for MAC table failed (%d)\n", ret1); diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h index f7c81133594f..ace6545f82e6 100644 --- a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h +++ b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h @@ -132,6 +132,9 @@ #define MLX4_EN_TX_COAL_PKTS 16 #define MLX4_EN_TX_COAL_TIME 0x10 +#define MLX4_EN_MAX_COAL_PKTS U16_MAX +#define MLX4_EN_MAX_COAL_TIME U16_MAX + #define MLX4_EN_RX_RATE_LOW 400000 #define MLX4_EN_RX_COAL_TIME_LOW 0 #define MLX4_EN_RX_RATE_HIGH 450000 @@ -552,8 +555,8 @@ struct mlx4_en_priv { u16 rx_usecs_low; u32 pkt_rate_high; u16 rx_usecs_high; - u16 sample_interval; - u16 adaptive_rx_coal; + u32 sample_interval; + u32 adaptive_rx_coal; u32 msg_enable; u32 loopback_ok; u32 validate_loopback; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c b/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c index 3d46ef48d5b8..c641d5656b2d 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c @@ -1007,12 +1007,14 @@ static void mlx5e_trust_update_sq_inline_mode(struct mlx5e_priv *priv) mutex_lock(&priv->state_lock); - if (!test_bit(MLX5E_STATE_OPENED, &priv->state)) - goto out; - new_channels.params = priv->channels.params; mlx5e_trust_update_tx_min_inline_mode(priv, &new_channels.params); + if (!test_bit(MLX5E_STATE_OPENED, &priv->state)) { + priv->channels.params = new_channels.params; + goto out; + } + /* Skip if tx_min_inline is the same */ if (new_channels.params.tx_min_inline_mode == priv->channels.params.tx_min_inline_mode) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c index d8f68e4d1018..876c3e4c6193 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c @@ -877,13 +877,14 @@ static const struct net_device_ops mlx5e_netdev_ops_rep = { }; static void mlx5e_build_rep_params(struct mlx5_core_dev *mdev, - struct mlx5e_params *params) + struct mlx5e_params *params, u16 mtu) { u8 cq_period_mode = MLX5_CAP_GEN(mdev, cq_period_start_from_cqe) ? MLX5_CQ_PERIOD_MODE_START_FROM_CQE : MLX5_CQ_PERIOD_MODE_START_FROM_EQE; params->hard_mtu = MLX5E_ETH_HARD_MTU; + params->sw_mtu = mtu; params->log_sq_size = MLX5E_REP_PARAMS_LOG_SQ_SIZE; params->rq_wq_type = MLX5_WQ_TYPE_LINKED_LIST; params->log_rq_mtu_frames = MLX5E_REP_PARAMS_LOG_RQ_SIZE; @@ -931,7 +932,7 @@ static void mlx5e_init_rep(struct mlx5_core_dev *mdev, priv->channels.params.num_channels = profile->max_nch(mdev); - mlx5e_build_rep_params(mdev, &priv->channels.params); + mlx5e_build_rep_params(mdev, &priv->channels.params, netdev->mtu); mlx5e_build_rep_netdev(netdev); mlx5e_timestamp_init(priv); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_selftest.c b/drivers/net/ethernet/mellanox/mlx5/core/en_selftest.c index 707976482c09..027f54ac1ca2 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_selftest.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_selftest.c @@ -290,7 +290,7 @@ static int mlx5e_test_loopback(struct mlx5e_priv *priv) if (!test_bit(MLX5E_STATE_OPENED, &priv->state)) { netdev_err(priv->netdev, - "\tCan't perform loobpack test while device is down\n"); + "\tCan't perform loopback test while device is down\n"); return -ENODEV; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c index 4197001f9801..b94276db3ce9 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c @@ -1261,6 +1261,10 @@ static int __parse_cls_flower(struct mlx5e_priv *priv, f->mask); addr_type = key->addr_type; + /* the HW doesn't support frag first/later */ + if (mask->flags & FLOW_DIS_FIRST_FRAG) + return -EOPNOTSUPP; + if (mask->flags & FLOW_DIS_IS_FRAGMENT) { MLX5_SET(fte_match_set_lyr_2_4, headers_c, frag, 1); MLX5_SET(fte_match_set_lyr_2_4, headers_v, frag, @@ -1864,7 +1868,8 @@ static bool modify_header_match_supported(struct mlx5_flow_spec *spec, } ip_proto = MLX5_GET(fte_match_set_lyr_2_4, headers_v, ip_protocol); - if (modify_ip_header && ip_proto != IPPROTO_TCP && ip_proto != IPPROTO_UDP) { + if (modify_ip_header && ip_proto != IPPROTO_TCP && + ip_proto != IPPROTO_UDP && ip_proto != IPPROTO_ICMP) { pr_info("can't offload re-write of ip proto %d\n", ip_proto); return false; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c index 20297108528a..5532aa3675c7 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c @@ -255,7 +255,7 @@ mlx5e_txwqe_build_dsegs(struct mlx5e_txqsq *sq, struct sk_buff *skb, dma_addr = dma_map_single(sq->pdev, skb_data, headlen, DMA_TO_DEVICE); if (unlikely(dma_mapping_error(sq->pdev, dma_addr))) - return -ENOMEM; + goto dma_unmap_wqe_err; dseg->addr = cpu_to_be64(dma_addr); dseg->lkey = sq->mkey_be; @@ -273,7 +273,7 @@ mlx5e_txwqe_build_dsegs(struct mlx5e_txqsq *sq, struct sk_buff *skb, dma_addr = skb_frag_dma_map(sq->pdev, frag, 0, fsz, DMA_TO_DEVICE); if (unlikely(dma_mapping_error(sq->pdev, dma_addr))) - return -ENOMEM; + goto dma_unmap_wqe_err; dseg->addr = cpu_to_be64(dma_addr); dseg->lkey = sq->mkey_be; @@ -285,6 +285,10 @@ mlx5e_txwqe_build_dsegs(struct mlx5e_txqsq *sq, struct sk_buff *skb, } return num_dma; + +dma_unmap_wqe_err: + mlx5e_dma_unmap_wqe_err(sq, num_dma); + return -ENOMEM; } static inline void @@ -380,17 +384,15 @@ static netdev_tx_t mlx5e_sq_xmit(struct mlx5e_txqsq *sq, struct sk_buff *skb, num_dma = mlx5e_txwqe_build_dsegs(sq, skb, skb_data, headlen, (struct mlx5_wqe_data_seg *)cseg + ds_cnt); if (unlikely(num_dma < 0)) - goto dma_unmap_wqe_err; + goto err_drop; mlx5e_txwqe_complete(sq, skb, opcode, ds_cnt + num_dma, num_bytes, num_dma, wi, cseg); return NETDEV_TX_OK; -dma_unmap_wqe_err: +err_drop: sq->stats.dropped++; - mlx5e_dma_unmap_wqe_err(sq, wi->num_dma); - dev_kfree_skb_any(skb); return NETDEV_TX_OK; @@ -645,17 +647,15 @@ netdev_tx_t mlx5i_sq_xmit(struct mlx5e_txqsq *sq, struct sk_buff *skb, num_dma = mlx5e_txwqe_build_dsegs(sq, skb, skb_data, headlen, (struct mlx5_wqe_data_seg *)cseg + ds_cnt); if (unlikely(num_dma < 0)) - goto dma_unmap_wqe_err; + goto err_drop; mlx5e_txwqe_complete(sq, skb, opcode, ds_cnt + num_dma, num_bytes, num_dma, wi, cseg); return NETDEV_TX_OK; -dma_unmap_wqe_err: +err_drop: sq->stats.dropped++; - mlx5e_dma_unmap_wqe_err(sq, wi->num_dma); - dev_kfree_skb_any(skb); return NETDEV_TX_OK; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eq.c b/drivers/net/ethernet/mellanox/mlx5/core/eq.c index c1c94974e16b..1814f803bd2c 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eq.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eq.c @@ -34,6 +34,9 @@ #include <linux/module.h> #include <linux/mlx5/driver.h> #include <linux/mlx5/cmd.h> +#ifdef CONFIG_RFS_ACCEL +#include <linux/cpu_rmap.h> +#endif #include "mlx5_core.h" #include "fpga/core.h" #include "eswitch.h" @@ -923,3 +926,28 @@ int mlx5_core_eq_query(struct mlx5_core_dev *dev, struct mlx5_eq *eq, MLX5_SET(query_eq_in, in, eq_number, eq->eqn); return mlx5_cmd_exec(dev, in, sizeof(in), out, outlen); } + +/* This function should only be called after mlx5_cmd_force_teardown_hca */ +void mlx5_core_eq_free_irqs(struct mlx5_core_dev *dev) +{ + struct mlx5_eq_table *table = &dev->priv.eq_table; + struct mlx5_eq *eq; + +#ifdef CONFIG_RFS_ACCEL + if (dev->rmap) { + free_irq_cpu_rmap(dev->rmap); + dev->rmap = NULL; + } +#endif + list_for_each_entry(eq, &table->comp_eqs_list, list) + free_irq(eq->irqn, eq); + + free_irq(table->pages_eq.irqn, &table->pages_eq); + free_irq(table->async_eq.irqn, &table->async_eq); + free_irq(table->cmd_eq.irqn, &table->cmd_eq); +#ifdef CONFIG_INFINIBAND_ON_DEMAND_PAGING + if (MLX5_CAP_GEN(dev, pg)) + free_irq(table->pfault_eq.irqn, &table->pfault_eq); +#endif + pci_free_irq_vectors(dev->pdev); +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c index 332bc56306bf..1352d13eedb3 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c @@ -2175,26 +2175,35 @@ int mlx5_eswitch_get_vport_stats(struct mlx5_eswitch *esw, memset(vf_stats, 0, sizeof(*vf_stats)); vf_stats->rx_packets = MLX5_GET_CTR(out, received_eth_unicast.packets) + + MLX5_GET_CTR(out, received_ib_unicast.packets) + MLX5_GET_CTR(out, received_eth_multicast.packets) + + MLX5_GET_CTR(out, received_ib_multicast.packets) + MLX5_GET_CTR(out, received_eth_broadcast.packets); vf_stats->rx_bytes = MLX5_GET_CTR(out, received_eth_unicast.octets) + + MLX5_GET_CTR(out, received_ib_unicast.octets) + MLX5_GET_CTR(out, received_eth_multicast.octets) + + MLX5_GET_CTR(out, received_ib_multicast.octets) + MLX5_GET_CTR(out, received_eth_broadcast.octets); vf_stats->tx_packets = MLX5_GET_CTR(out, transmitted_eth_unicast.packets) + + MLX5_GET_CTR(out, transmitted_ib_unicast.packets) + MLX5_GET_CTR(out, transmitted_eth_multicast.packets) + + MLX5_GET_CTR(out, transmitted_ib_multicast.packets) + MLX5_GET_CTR(out, transmitted_eth_broadcast.packets); vf_stats->tx_bytes = MLX5_GET_CTR(out, transmitted_eth_unicast.octets) + + MLX5_GET_CTR(out, transmitted_ib_unicast.octets) + MLX5_GET_CTR(out, transmitted_eth_multicast.octets) + + MLX5_GET_CTR(out, transmitted_ib_multicast.octets) + MLX5_GET_CTR(out, transmitted_eth_broadcast.octets); vf_stats->multicast = - MLX5_GET_CTR(out, received_eth_multicast.packets); + MLX5_GET_CTR(out, received_eth_multicast.packets) + + MLX5_GET_CTR(out, received_ib_multicast.packets); vf_stats->broadcast = MLX5_GET_CTR(out, received_eth_broadcast.packets); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c index de51e7c39bc8..c39c1692e674 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c @@ -187,6 +187,7 @@ static void del_sw_ns(struct fs_node *node); static void del_sw_hw_rule(struct fs_node *node); static bool mlx5_flow_dests_cmp(struct mlx5_flow_destination *d1, struct mlx5_flow_destination *d2); +static void cleanup_root_ns(struct mlx5_flow_root_namespace *root_ns); static struct mlx5_flow_rule * find_flow_rule(struct fs_fte *fte, struct mlx5_flow_destination *dest); @@ -481,7 +482,8 @@ static void del_sw_hw_rule(struct fs_node *node) if (rule->dest_attr.type == MLX5_FLOW_DESTINATION_TYPE_COUNTER && --fte->dests_size) { - modify_mask = BIT(MLX5_SET_FTE_MODIFY_ENABLE_MASK_ACTION); + modify_mask = BIT(MLX5_SET_FTE_MODIFY_ENABLE_MASK_ACTION) | + BIT(MLX5_SET_FTE_MODIFY_ENABLE_MASK_FLOW_COUNTERS); fte->action.action &= ~MLX5_FLOW_CONTEXT_ACTION_COUNT; update_fte = true; goto out; @@ -2351,23 +2353,27 @@ static int create_anchor_flow_table(struct mlx5_flow_steering *steering) static int init_root_ns(struct mlx5_flow_steering *steering) { + int err; + steering->root_ns = create_root_ns(steering, FS_FT_NIC_RX); if (!steering->root_ns) - goto cleanup; + return -ENOMEM; - if (init_root_tree(steering, &root_fs, &steering->root_ns->ns.node)) - goto cleanup; + err = init_root_tree(steering, &root_fs, &steering->root_ns->ns.node); + if (err) + goto out_err; set_prio_attrs(steering->root_ns); - - if (create_anchor_flow_table(steering)) - goto cleanup; + err = create_anchor_flow_table(steering); + if (err) + goto out_err; return 0; -cleanup: - mlx5_cleanup_fs(steering->dev); - return -ENOMEM; +out_err: + cleanup_root_ns(steering->root_ns); + steering->root_ns = NULL; + return err; } static void clean_tree(struct fs_node *node) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c index 63a8ea31601c..e2c465b0b3f8 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c @@ -1587,6 +1587,14 @@ static int mlx5_try_fast_unload(struct mlx5_core_dev *dev) mlx5_enter_error_state(dev, true); + /* Some platforms requiring freeing the IRQ's in the shutdown + * flow. If they aren't freed they can't be allocated after + * kexec. There is no need to cleanup the mlx5_core software + * contexts. + */ + mlx5_irq_clear_affinity_hints(dev); + mlx5_core_eq_free_irqs(dev); + return 0; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h index 7d001fe6e631..023882d9a22e 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h @@ -128,6 +128,8 @@ int mlx5_core_eq_query(struct mlx5_core_dev *dev, struct mlx5_eq *eq, u32 *out, int outlen); int mlx5_start_eqs(struct mlx5_core_dev *dev); void mlx5_stop_eqs(struct mlx5_core_dev *dev); +/* This function should only be called after mlx5_cmd_force_teardown_hca */ +void mlx5_core_eq_free_irqs(struct mlx5_core_dev *dev); struct mlx5_eq *mlx5_eqn2eq(struct mlx5_core_dev *dev, int eqn); u32 mlx5_eq_poll_irq_disabled(struct mlx5_eq *eq); void mlx5_cq_tasklet_cb(unsigned long data); diff --git a/drivers/net/ethernet/mellanox/mlxsw/core.c b/drivers/net/ethernet/mellanox/mlxsw/core.c index 93ea56620a24..e13ac3b8dff7 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/core.c +++ b/drivers/net/ethernet/mellanox/mlxsw/core.c @@ -1100,11 +1100,11 @@ err_emad_init: err_alloc_lag_mapping: mlxsw_ports_fini(mlxsw_core); err_ports_init: - mlxsw_bus->fini(bus_priv); -err_bus_init: if (!reload) devlink_resources_unregister(devlink, NULL); err_register_resources: + mlxsw_bus->fini(bus_priv); +err_bus_init: if (!reload) devlink_free(devlink); err_devlink_alloc: diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c index c11c9a635866..4ed01182a82c 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c @@ -1718,13 +1718,11 @@ __mlxsw_sp_port_mdb_del(struct mlxsw_sp_port *mlxsw_sp_port, struct net_device *dev = mlxsw_sp_port->dev; int err; - if (bridge_port->bridge_device->multicast_enabled) { - if (bridge_port->bridge_device->multicast_enabled) { - err = mlxsw_sp_port_smid_set(mlxsw_sp_port, mid->mid, - false); - if (err) - netdev_err(dev, "Unable to remove port from SMID\n"); - } + if (bridge_port->bridge_device->multicast_enabled && + !bridge_port->mrouter) { + err = mlxsw_sp_port_smid_set(mlxsw_sp_port, mid->mid, false); + if (err) + netdev_err(dev, "Unable to remove port from SMID\n"); } err = mlxsw_sp_port_remove_from_mid(mlxsw_sp_port, mid); diff --git a/drivers/net/ethernet/netronome/nfp/flower/action.c b/drivers/net/ethernet/netronome/nfp/flower/action.c index b3567a596fc1..80df9a5d4217 100644 --- a/drivers/net/ethernet/netronome/nfp/flower/action.c +++ b/drivers/net/ethernet/netronome/nfp/flower/action.c @@ -183,17 +183,21 @@ static int nfp_fl_set_ipv4_udp_tun(struct nfp_fl_set_ipv4_udp_tun *set_tun, const struct tc_action *action, struct nfp_fl_pre_tunnel *pre_tun, - enum nfp_flower_tun_type tun_type) + enum nfp_flower_tun_type tun_type, + struct net_device *netdev) { size_t act_size = sizeof(struct nfp_fl_set_ipv4_udp_tun); struct ip_tunnel_info *ip_tun = tcf_tunnel_info(action); u32 tmp_set_ip_tun_type_index = 0; /* Currently support one pre-tunnel so index is always 0. */ int pretun_idx = 0; + struct net *net; if (ip_tun->options_len) return -EOPNOTSUPP; + net = dev_net(netdev); + set_tun->head.jump_id = NFP_FL_ACTION_OPCODE_SET_IPV4_TUNNEL; set_tun->head.len_lw = act_size >> NFP_FL_LW_SIZ; @@ -204,6 +208,7 @@ nfp_fl_set_ipv4_udp_tun(struct nfp_fl_set_ipv4_udp_tun *set_tun, set_tun->tun_type_index = cpu_to_be32(tmp_set_ip_tun_type_index); set_tun->tun_id = ip_tun->key.tun_id; + set_tun->ttl = net->ipv4.sysctl_ip_default_ttl; /* Complete pre_tunnel action. */ pre_tun->ipv4_dst = ip_tun->key.u.ipv4.dst; @@ -511,7 +516,8 @@ nfp_flower_loop_action(const struct tc_action *a, *a_len += sizeof(struct nfp_fl_pre_tunnel); set_tun = (void *)&nfp_fl->action_data[*a_len]; - err = nfp_fl_set_ipv4_udp_tun(set_tun, a, pre_tun, *tun_type); + err = nfp_fl_set_ipv4_udp_tun(set_tun, a, pre_tun, *tun_type, + netdev); if (err) return err; *a_len += sizeof(struct nfp_fl_set_ipv4_udp_tun); diff --git a/drivers/net/ethernet/netronome/nfp/flower/cmsg.h b/drivers/net/ethernet/netronome/nfp/flower/cmsg.h index b6c0fd053a50..bee4367a2c38 100644 --- a/drivers/net/ethernet/netronome/nfp/flower/cmsg.h +++ b/drivers/net/ethernet/netronome/nfp/flower/cmsg.h @@ -190,7 +190,10 @@ struct nfp_fl_set_ipv4_udp_tun { __be16 reserved; __be64 tun_id __packed; __be32 tun_type_index; - __be32 extra[3]; + __be16 reserved2; + u8 ttl; + u8 reserved3; + __be32 extra[2]; }; /* Metadata with L2 (1W/4B) diff --git a/drivers/net/ethernet/netronome/nfp/flower/main.c b/drivers/net/ethernet/netronome/nfp/flower/main.c index ad02592a82b7..84e3b9f5abb1 100644 --- a/drivers/net/ethernet/netronome/nfp/flower/main.c +++ b/drivers/net/ethernet/netronome/nfp/flower/main.c @@ -52,8 +52,6 @@ #define NFP_FLOWER_ALLOWED_VER 0x0001000000010000UL -#define NFP_FLOWER_FRAME_HEADROOM 158 - static const char *nfp_flower_extra_cap(struct nfp_app *app, struct nfp_net *nn) { return "FLOWER"; @@ -360,7 +358,7 @@ nfp_flower_spawn_phy_reprs(struct nfp_app *app, struct nfp_flower_priv *priv) } SET_NETDEV_DEV(repr, &priv->nn->pdev->dev); - nfp_net_get_mac_addr(app->pf, port); + nfp_net_get_mac_addr(app->pf, repr, port); cmsg_port_id = nfp_flower_cmsg_phys_port(phys_port); err = nfp_repr_init(app, repr, @@ -559,22 +557,6 @@ static void nfp_flower_clean(struct nfp_app *app) app->priv = NULL; } -static int -nfp_flower_check_mtu(struct nfp_app *app, struct net_device *netdev, - int new_mtu) -{ - /* The flower fw reserves NFP_FLOWER_FRAME_HEADROOM bytes of the - * supported max MTU to allow for appending tunnel headers. To prevent - * unexpected behaviour this needs to be accounted for. - */ - if (new_mtu > netdev->max_mtu - NFP_FLOWER_FRAME_HEADROOM) { - nfp_err(app->cpp, "New MTU (%d) is not valid\n", new_mtu); - return -EINVAL; - } - - return 0; -} - static bool nfp_flower_check_ack(struct nfp_flower_priv *app_priv) { bool ret; @@ -656,7 +638,6 @@ const struct nfp_app_type app_flower = { .init = nfp_flower_init, .clean = nfp_flower_clean, - .check_mtu = nfp_flower_check_mtu, .repr_change_mtu = nfp_flower_repr_change_mtu, .vnic_alloc = nfp_flower_vnic_alloc, diff --git a/drivers/net/ethernet/netronome/nfp/nfp_app_nic.c b/drivers/net/ethernet/netronome/nfp/nfp_app_nic.c index 2a2f2fbc8850..b9618c37403f 100644 --- a/drivers/net/ethernet/netronome/nfp/nfp_app_nic.c +++ b/drivers/net/ethernet/netronome/nfp/nfp_app_nic.c @@ -69,7 +69,7 @@ int nfp_app_nic_vnic_alloc(struct nfp_app *app, struct nfp_net *nn, if (err) return err < 0 ? err : 0; - nfp_net_get_mac_addr(app->pf, nn->port); + nfp_net_get_mac_addr(app->pf, nn->dp.netdev, nn->port); return 0; } diff --git a/drivers/net/ethernet/netronome/nfp/nfp_main.h b/drivers/net/ethernet/netronome/nfp/nfp_main.h index add46e28212b..42211083b51f 100644 --- a/drivers/net/ethernet/netronome/nfp/nfp_main.h +++ b/drivers/net/ethernet/netronome/nfp/nfp_main.h @@ -171,7 +171,9 @@ void nfp_net_pci_remove(struct nfp_pf *pf); int nfp_hwmon_register(struct nfp_pf *pf); void nfp_hwmon_unregister(struct nfp_pf *pf); -void nfp_net_get_mac_addr(struct nfp_pf *pf, struct nfp_port *port); +void +nfp_net_get_mac_addr(struct nfp_pf *pf, struct net_device *netdev, + struct nfp_port *port); bool nfp_ctrl_tx(struct nfp_net *nn, struct sk_buff *skb); diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_main.c b/drivers/net/ethernet/netronome/nfp/nfp_net_main.c index 15fa47f622aa..45cd2092e498 100644 --- a/drivers/net/ethernet/netronome/nfp/nfp_net_main.c +++ b/drivers/net/ethernet/netronome/nfp/nfp_net_main.c @@ -67,23 +67,26 @@ /** * nfp_net_get_mac_addr() - Get the MAC address. * @pf: NFP PF handle + * @netdev: net_device to set MAC address on * @port: NFP port structure * * First try to get the MAC address from NSP ETH table. If that * fails generate a random address. */ -void nfp_net_get_mac_addr(struct nfp_pf *pf, struct nfp_port *port) +void +nfp_net_get_mac_addr(struct nfp_pf *pf, struct net_device *netdev, + struct nfp_port *port) { struct nfp_eth_table_port *eth_port; eth_port = __nfp_port_get_eth_port(port); if (!eth_port) { - eth_hw_addr_random(port->netdev); + eth_hw_addr_random(netdev); return; } - ether_addr_copy(port->netdev->dev_addr, eth_port->mac_addr); - ether_addr_copy(port->netdev->perm_addr, eth_port->mac_addr); + ether_addr_copy(netdev->dev_addr, eth_port->mac_addr); + ether_addr_copy(netdev->perm_addr, eth_port->mac_addr); } static struct nfp_eth_table_port * @@ -511,16 +514,18 @@ static int nfp_net_pci_map_mem(struct nfp_pf *pf) return PTR_ERR(mem); } - min_size = NFP_MAC_STATS_SIZE * (pf->eth_tbl->max_index + 1); - pf->mac_stats_mem = nfp_rtsym_map(pf->rtbl, "_mac_stats", - "net.macstats", min_size, - &pf->mac_stats_bar); - if (IS_ERR(pf->mac_stats_mem)) { - if (PTR_ERR(pf->mac_stats_mem) != -ENOENT) { - err = PTR_ERR(pf->mac_stats_mem); - goto err_unmap_ctrl; + if (pf->eth_tbl) { + min_size = NFP_MAC_STATS_SIZE * (pf->eth_tbl->max_index + 1); + pf->mac_stats_mem = nfp_rtsym_map(pf->rtbl, "_mac_stats", + "net.macstats", min_size, + &pf->mac_stats_bar); + if (IS_ERR(pf->mac_stats_mem)) { + if (PTR_ERR(pf->mac_stats_mem) != -ENOENT) { + err = PTR_ERR(pf->mac_stats_mem); + goto err_unmap_ctrl; + } + pf->mac_stats_mem = NULL; } - pf->mac_stats_mem = NULL; } pf->vf_cfg_mem = nfp_net_pf_map_rtsym(pf, "net.vfcfg", diff --git a/drivers/net/ethernet/ni/nixge.c b/drivers/net/ethernet/ni/nixge.c index 27364b7572fc..b092894dd128 100644 --- a/drivers/net/ethernet/ni/nixge.c +++ b/drivers/net/ethernet/ni/nixge.c @@ -1170,7 +1170,7 @@ static void *nixge_get_nvmem_address(struct device *dev) cell = nvmem_cell_get(dev, "address"); if (IS_ERR(cell)) - return cell; + return NULL; mac = nvmem_cell_read(cell, &cell_size); nvmem_cell_put(cell); @@ -1183,7 +1183,7 @@ static int nixge_probe(struct platform_device *pdev) struct nixge_priv *priv; struct net_device *ndev; struct resource *dmares; - const char *mac_addr; + const u8 *mac_addr; int err; ndev = alloc_etherdev(sizeof(*priv)); @@ -1202,10 +1202,12 @@ static int nixge_probe(struct platform_device *pdev) ndev->max_mtu = NIXGE_JUMBO_MTU; mac_addr = nixge_get_nvmem_address(&pdev->dev); - if (mac_addr && is_valid_ether_addr(mac_addr)) + if (mac_addr && is_valid_ether_addr(mac_addr)) { ether_addr_copy(ndev->dev_addr, mac_addr); - else + kfree(mac_addr); + } else { eth_hw_addr_random(ndev); + } priv = netdev_priv(ndev); priv->ndev = ndev; diff --git a/drivers/net/ethernet/qlogic/qed/qed_l2.c b/drivers/net/ethernet/qlogic/qed/qed_l2.c index e874504e8b28..8667799d0069 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_l2.c +++ b/drivers/net/ethernet/qlogic/qed/qed_l2.c @@ -115,8 +115,7 @@ int qed_l2_alloc(struct qed_hwfn *p_hwfn) void qed_l2_setup(struct qed_hwfn *p_hwfn) { - if (p_hwfn->hw_info.personality != QED_PCI_ETH && - p_hwfn->hw_info.personality != QED_PCI_ETH_ROCE) + if (!QED_IS_L2_PERSONALITY(p_hwfn)) return; mutex_init(&p_hwfn->p_l2_info->lock); @@ -126,8 +125,7 @@ void qed_l2_free(struct qed_hwfn *p_hwfn) { u32 i; - if (p_hwfn->hw_info.personality != QED_PCI_ETH && - p_hwfn->hw_info.personality != QED_PCI_ETH_ROCE) + if (!QED_IS_L2_PERSONALITY(p_hwfn)) return; if (!p_hwfn->p_l2_info) diff --git a/drivers/net/ethernet/qlogic/qed/qed_ll2.c b/drivers/net/ethernet/qlogic/qed/qed_ll2.c index 74fc626b1ec1..38502815d681 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_ll2.c +++ b/drivers/net/ethernet/qlogic/qed/qed_ll2.c @@ -2370,7 +2370,7 @@ static int qed_ll2_start_xmit(struct qed_dev *cdev, struct sk_buff *skb) u8 flags = 0; if (unlikely(skb->ip_summed != CHECKSUM_NONE)) { - DP_INFO(cdev, "Cannot transmit a checksumed packet\n"); + DP_INFO(cdev, "Cannot transmit a checksummed packet\n"); return -EINVAL; } diff --git a/drivers/net/ethernet/qlogic/qed/qed_main.c b/drivers/net/ethernet/qlogic/qed/qed_main.c index 9854aa9139af..7870ae2a6f7e 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_main.c +++ b/drivers/net/ethernet/qlogic/qed/qed_main.c @@ -680,7 +680,7 @@ static int qed_nic_stop(struct qed_dev *cdev) tasklet_disable(p_hwfn->sp_dpc); p_hwfn->b_sp_dpc_enabled = false; DP_VERBOSE(cdev, NETIF_MSG_IFDOWN, - "Disabled sp taskelt [hwfn %d] at %p\n", + "Disabled sp tasklet [hwfn %d] at %p\n", i, p_hwfn->sp_dpc); } } diff --git a/drivers/net/ethernet/qlogic/qed/qed_roce.c b/drivers/net/ethernet/qlogic/qed/qed_roce.c index fb7c2d1562ae..6acfd43c1a4f 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_roce.c +++ b/drivers/net/ethernet/qlogic/qed/qed_roce.c @@ -848,7 +848,7 @@ int qed_roce_query_qp(struct qed_hwfn *p_hwfn, if (!(qp->resp_offloaded)) { DP_NOTICE(p_hwfn, - "The responder's qp should be offloded before requester's\n"); + "The responder's qp should be offloaded before requester's\n"); return -EINVAL; } diff --git a/drivers/net/ethernet/qlogic/qede/qede_rdma.c b/drivers/net/ethernet/qlogic/qede/qede_rdma.c index 50b142fad6b8..1900bf7e67d1 100644 --- a/drivers/net/ethernet/qlogic/qede/qede_rdma.c +++ b/drivers/net/ethernet/qlogic/qede/qede_rdma.c @@ -238,7 +238,7 @@ qede_rdma_get_free_event_node(struct qede_dev *edev) } if (!found) { - event_node = kzalloc(sizeof(*event_node), GFP_KERNEL); + event_node = kzalloc(sizeof(*event_node), GFP_ATOMIC); if (!event_node) { DP_NOTICE(edev, "qedr: Could not allocate memory for rdma work\n"); diff --git a/drivers/net/ethernet/realtek/8139too.c b/drivers/net/ethernet/realtek/8139too.c index d24b47b8e0b2..d118da5a10a2 100644 --- a/drivers/net/ethernet/realtek/8139too.c +++ b/drivers/net/ethernet/realtek/8139too.c @@ -2224,7 +2224,7 @@ static void rtl8139_poll_controller(struct net_device *dev) struct rtl8139_private *tp = netdev_priv(dev); const int irq = tp->pci_dev->irq; - disable_irq(irq); + disable_irq_nosync(irq); rtl8139_interrupt(irq, dev); enable_irq(irq); } diff --git a/drivers/net/ethernet/realtek/r8169.c b/drivers/net/ethernet/realtek/r8169.c index 604ae78381ae..c7aac1fc99e8 100644 --- a/drivers/net/ethernet/realtek/r8169.c +++ b/drivers/net/ethernet/realtek/r8169.c @@ -4981,6 +4981,9 @@ static void rtl_pll_power_down(struct rtl8169_private *tp) static void rtl_pll_power_up(struct rtl8169_private *tp) { rtl_generic_op(tp, tp->pll_power_ops.up); + + /* give MAC/PHY some time to resume */ + msleep(20); } static void rtl_init_pll_power_ops(struct rtl8169_private *tp) diff --git a/drivers/net/ethernet/sfc/ef10.c b/drivers/net/ethernet/sfc/ef10.c index 63036d9bf3e6..d90a7b1f4088 100644 --- a/drivers/net/ethernet/sfc/ef10.c +++ b/drivers/net/ethernet/sfc/ef10.c @@ -4784,8 +4784,9 @@ expire: * will set rule->filter_id to EFX_ARFS_FILTER_ID_PENDING, meaning that * the rule is not removed by efx_rps_hash_del() below. */ - ret = efx_ef10_filter_remove_internal(efx, 1U << spec->priority, - filter_idx, true) == 0; + if (ret) + ret = efx_ef10_filter_remove_internal(efx, 1U << spec->priority, + filter_idx, true) == 0; /* While we can't safely dereference rule (we dropped the lock), we can * still test it for NULL. */ diff --git a/drivers/net/ethernet/sfc/rx.c b/drivers/net/ethernet/sfc/rx.c index 64a94f242027..d2e254f2f72b 100644 --- a/drivers/net/ethernet/sfc/rx.c +++ b/drivers/net/ethernet/sfc/rx.c @@ -839,6 +839,8 @@ static void efx_filter_rfs_work(struct work_struct *data) int rc; rc = efx->type->filter_insert(efx, &req->spec, true); + if (rc >= 0) + rc %= efx->type->max_rx_ip_filters; if (efx->rps_hash_table) { spin_lock_bh(&efx->rps_hash_lock); rule = efx_rps_hash_find(efx, &req->spec); diff --git a/drivers/net/ethernet/sun/niu.c b/drivers/net/ethernet/sun/niu.c index f081de4f38d7..88c12474a0c3 100644 --- a/drivers/net/ethernet/sun/niu.c +++ b/drivers/net/ethernet/sun/niu.c @@ -3443,7 +3443,7 @@ static int niu_process_rx_pkt(struct napi_struct *napi, struct niu *np, len = (val & RCR_ENTRY_L2_LEN) >> RCR_ENTRY_L2_LEN_SHIFT; - len -= ETH_FCS_LEN; + append_size = len + ETH_HLEN + ETH_FCS_LEN; addr = (val & RCR_ENTRY_PKT_BUF_ADDR) << RCR_ENTRY_PKT_BUF_ADDR_SHIFT; @@ -3453,7 +3453,6 @@ static int niu_process_rx_pkt(struct napi_struct *napi, struct niu *np, RCR_ENTRY_PKTBUFSZ_SHIFT]; off = addr & ~PAGE_MASK; - append_size = rcr_size; if (num_rcr == 1) { int ptype; @@ -3466,7 +3465,7 @@ static int niu_process_rx_pkt(struct napi_struct *napi, struct niu *np, else skb_checksum_none_assert(skb); } else if (!(val & RCR_ENTRY_MULTI)) - append_size = len - skb->len; + append_size = append_size - skb->len; niu_rx_skb_append(skb, page, off, append_size, rcr_size); if ((page->index + rp->rbr_block_size) - rcr_size == addr) { diff --git a/drivers/net/ethernet/ti/cpsw.c b/drivers/net/ethernet/ti/cpsw.c index 74f828412055..28d893b93d30 100644 --- a/drivers/net/ethernet/ti/cpsw.c +++ b/drivers/net/ethernet/ti/cpsw.c @@ -1340,6 +1340,8 @@ static inline void cpsw_add_dual_emac_def_ale_entries( cpsw_ale_add_ucast(cpsw->ale, priv->mac_addr, HOST_PORT_NUM, ALE_VLAN | ALE_SECURE, slave->port_vlan); + cpsw_ale_control_set(cpsw->ale, slave_port, + ALE_PORT_DROP_UNKNOWN_VLAN, 1); } static void soft_reset_slave(struct cpsw_slave *slave) diff --git a/drivers/net/hyperv/netvsc_drv.c b/drivers/net/hyperv/netvsc_drv.c index ecc84954c511..da07ccdf84bf 100644 --- a/drivers/net/hyperv/netvsc_drv.c +++ b/drivers/net/hyperv/netvsc_drv.c @@ -1840,7 +1840,8 @@ static int netvsc_vf_join(struct net_device *vf_netdev, goto rx_handler_failed; } - ret = netdev_upper_dev_link(vf_netdev, ndev, NULL); + ret = netdev_master_upper_dev_link(vf_netdev, ndev, + NULL, NULL, NULL); if (ret != 0) { netdev_err(vf_netdev, "can not set master device %s (err = %d)\n", diff --git a/drivers/net/hyperv/rndis_filter.c b/drivers/net/hyperv/rndis_filter.c index 6b127be781d9..e7ca5b5f39ed 100644 --- a/drivers/net/hyperv/rndis_filter.c +++ b/drivers/net/hyperv/rndis_filter.c @@ -1288,7 +1288,7 @@ struct netvsc_device *rndis_filter_device_add(struct hv_device *dev, rndis_device->link_state ? "down" : "up"); if (net_device->nvsp_version < NVSP_PROTOCOL_VERSION_5) - return net_device; + goto out; rndis_filter_query_link_speed(rndis_device, net_device); diff --git a/drivers/net/ieee802154/atusb.c b/drivers/net/ieee802154/atusb.c index 9fb9b565a002..4f684cbcdc57 100644 --- a/drivers/net/ieee802154/atusb.c +++ b/drivers/net/ieee802154/atusb.c @@ -1045,7 +1045,7 @@ static int atusb_probe(struct usb_interface *interface, atusb->tx_dr.bRequest = ATUSB_TX; atusb->tx_dr.wValue = cpu_to_le16(0); - atusb->tx_urb = usb_alloc_urb(0, GFP_ATOMIC); + atusb->tx_urb = usb_alloc_urb(0, GFP_KERNEL); if (!atusb->tx_urb) goto fail; diff --git a/drivers/net/ieee802154/mcr20a.c b/drivers/net/ieee802154/mcr20a.c index 55a22c761808..de0d7f28a181 100644 --- a/drivers/net/ieee802154/mcr20a.c +++ b/drivers/net/ieee802154/mcr20a.c @@ -1267,7 +1267,7 @@ mcr20a_probe(struct spi_device *spi) ret = mcr20a_get_platform_data(spi, pdata); if (ret < 0) { dev_crit(&spi->dev, "mcr20a_get_platform_data failed.\n"); - return ret; + goto free_pdata; } /* init reset gpio */ @@ -1275,7 +1275,7 @@ mcr20a_probe(struct spi_device *spi) ret = devm_gpio_request_one(&spi->dev, pdata->rst_gpio, GPIOF_OUT_INIT_HIGH, "reset"); if (ret) - return ret; + goto free_pdata; } /* reset mcr20a */ @@ -1291,7 +1291,8 @@ mcr20a_probe(struct spi_device *spi) hw = ieee802154_alloc_hw(sizeof(*lp), &mcr20a_hw_ops); if (!hw) { dev_crit(&spi->dev, "ieee802154_alloc_hw failed\n"); - return -ENOMEM; + ret = -ENOMEM; + goto free_pdata; } /* init mcr20a local data */ @@ -1308,8 +1309,10 @@ mcr20a_probe(struct spi_device *spi) /* init buf */ lp->buf = devm_kzalloc(&spi->dev, SPI_COMMAND_BUFFER, GFP_KERNEL); - if (!lp->buf) - return -ENOMEM; + if (!lp->buf) { + ret = -ENOMEM; + goto free_dev; + } mcr20a_setup_tx_spi_messages(lp); mcr20a_setup_rx_spi_messages(lp); @@ -1366,6 +1369,8 @@ mcr20a_probe(struct spi_device *spi) free_dev: ieee802154_free_hw(lp->hw); +free_pdata: + kfree(pdata); return ret; } diff --git a/drivers/net/phy/broadcom.c b/drivers/net/phy/broadcom.c index 3bb6b66dc7bf..f9c25912eb98 100644 --- a/drivers/net/phy/broadcom.c +++ b/drivers/net/phy/broadcom.c @@ -720,6 +720,15 @@ static struct phy_driver broadcom_drivers[] = { .get_strings = bcm_phy_get_strings, .get_stats = bcm53xx_phy_get_stats, .probe = bcm53xx_phy_probe, +}, { + .phy_id = PHY_ID_BCM89610, + .phy_id_mask = 0xfffffff0, + .name = "Broadcom BCM89610", + .features = PHY_GBIT_FEATURES, + .flags = PHY_HAS_INTERRUPT, + .config_init = bcm54xx_config_init, + .ack_interrupt = bcm_phy_ack_intr, + .config_intr = bcm_phy_config_intr, } }; module_phy_driver(broadcom_drivers); @@ -741,6 +750,7 @@ static struct mdio_device_id __maybe_unused broadcom_tbl[] = { { PHY_ID_BCMAC131, 0xfffffff0 }, { PHY_ID_BCM5241, 0xfffffff0 }, { PHY_ID_BCM5395, 0xfffffff0 }, + { PHY_ID_BCM89610, 0xfffffff0 }, { } }; diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c index ac23322a32e1..9e4ba8e80a18 100644 --- a/drivers/net/phy/phy_device.c +++ b/drivers/net/phy/phy_device.c @@ -535,8 +535,17 @@ static int get_phy_id(struct mii_bus *bus, int addr, u32 *phy_id, /* Grab the bits from PHYIR1, and put them in the upper half */ phy_reg = mdiobus_read(bus, addr, MII_PHYSID1); - if (phy_reg < 0) + if (phy_reg < 0) { + /* if there is no device, return without an error so scanning + * the bus works properly + */ + if (phy_reg == -EIO || phy_reg == -ENODEV) { + *phy_id = 0xffffffff; + return 0; + } + return -EIO; + } *phy_id = (phy_reg & 0xffff) << 16; diff --git a/drivers/net/phy/sfp-bus.c b/drivers/net/phy/sfp-bus.c index 0381da78d228..fd6c23f69c2f 100644 --- a/drivers/net/phy/sfp-bus.c +++ b/drivers/net/phy/sfp-bus.c @@ -125,7 +125,7 @@ void sfp_parse_support(struct sfp_bus *bus, const struct sfp_eeprom_id *id, if (id->base.br_nominal) { if (id->base.br_nominal != 255) { br_nom = id->base.br_nominal * 100; - br_min = br_nom + id->base.br_nominal * id->ext.br_min; + br_min = br_nom - id->base.br_nominal * id->ext.br_min; br_max = br_nom + id->base.br_nominal * id->ext.br_max; } else if (id->ext.br_max) { br_nom = 250 * id->ext.br_max; diff --git a/drivers/net/usb/qmi_wwan.c b/drivers/net/usb/qmi_wwan.c index c853e7410f5a..42565dd33aa6 100644 --- a/drivers/net/usb/qmi_wwan.c +++ b/drivers/net/usb/qmi_wwan.c @@ -1098,6 +1098,7 @@ static const struct usb_device_id products[] = { {QMI_FIXED_INTF(0x05c6, 0x9080, 8)}, {QMI_FIXED_INTF(0x05c6, 0x9083, 3)}, {QMI_FIXED_INTF(0x05c6, 0x9084, 4)}, + {QMI_FIXED_INTF(0x05c6, 0x90b2, 3)}, /* ublox R410M */ {QMI_FIXED_INTF(0x05c6, 0x920d, 0)}, {QMI_FIXED_INTF(0x05c6, 0x920d, 5)}, {QMI_QUIRK_SET_DTR(0x05c6, 0x9625, 4)}, /* YUGA CLM920-NC5 */ @@ -1343,6 +1344,18 @@ static int qmi_wwan_probe(struct usb_interface *intf, id->driver_info = (unsigned long)&qmi_wwan_info; } + /* There are devices where the same interface number can be + * configured as different functions. We should only bind to + * vendor specific functions when matching on interface number + */ + if (id->match_flags & USB_DEVICE_ID_MATCH_INT_NUMBER && + desc->bInterfaceClass != USB_CLASS_VENDOR_SPEC) { + dev_dbg(&intf->dev, + "Rejecting interface number match for class %02x\n", + desc->bInterfaceClass); + return -ENODEV; + } + /* Quectel EC20 quirk where we've QMI on interface 4 instead of 0 */ if (quectel_ec20_detected(intf) && desc->bInterfaceNumber == 0) { dev_dbg(&intf->dev, "Quectel EC20 quirk, skipping interface 0\n"); diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/firmware.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/firmware.c index 9277f4c2bfeb..94e177d7c9b5 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/firmware.c +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/firmware.c @@ -459,7 +459,7 @@ static void brcmf_fw_free_request(struct brcmf_fw_request *req) kfree(req); } -static void brcmf_fw_request_nvram_done(const struct firmware *fw, void *ctx) +static int brcmf_fw_request_nvram_done(const struct firmware *fw, void *ctx) { struct brcmf_fw *fwctx = ctx; struct brcmf_fw_item *cur; @@ -498,13 +498,10 @@ static void brcmf_fw_request_nvram_done(const struct firmware *fw, void *ctx) brcmf_dbg(TRACE, "nvram %p len %d\n", nvram, nvram_length); cur->nv_data.data = nvram; cur->nv_data.len = nvram_length; - return; + return 0; fail: - brcmf_dbg(TRACE, "failed: dev=%s\n", dev_name(fwctx->dev)); - fwctx->done(fwctx->dev, -ENOENT, NULL); - brcmf_fw_free_request(fwctx->req); - kfree(fwctx); + return -ENOENT; } static int brcmf_fw_request_next_item(struct brcmf_fw *fwctx, bool async) @@ -553,20 +550,27 @@ static void brcmf_fw_request_done(const struct firmware *fw, void *ctx) brcmf_dbg(TRACE, "enter: firmware %s %sfound\n", cur->path, fw ? "" : "not "); - if (fw) { - if (cur->type == BRCMF_FW_TYPE_BINARY) - cur->binary = fw; - else if (cur->type == BRCMF_FW_TYPE_NVRAM) - brcmf_fw_request_nvram_done(fw, fwctx); - else - release_firmware(fw); - } else if (cur->type == BRCMF_FW_TYPE_NVRAM) { - brcmf_fw_request_nvram_done(NULL, fwctx); - } else if (!(cur->flags & BRCMF_FW_REQF_OPTIONAL)) { + if (!fw) ret = -ENOENT; + + switch (cur->type) { + case BRCMF_FW_TYPE_NVRAM: + ret = brcmf_fw_request_nvram_done(fw, fwctx); + break; + case BRCMF_FW_TYPE_BINARY: + cur->binary = fw; + break; + default: + /* something fishy here so bail out early */ + brcmf_err("unknown fw type: %d\n", cur->type); + release_firmware(fw); + ret = -EINVAL; goto fail; } + if (ret < 0 && !(cur->flags & BRCMF_FW_REQF_OPTIONAL)) + goto fail; + do { if (++fwctx->curpos == fwctx->req->n_items) { ret = 0; diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/scan.h b/drivers/net/wireless/intel/iwlwifi/fw/api/scan.h index 7af3a0f51b77..a17c4a79b8d4 100644 --- a/drivers/net/wireless/intel/iwlwifi/fw/api/scan.h +++ b/drivers/net/wireless/intel/iwlwifi/fw/api/scan.h @@ -8,6 +8,7 @@ * Copyright(c) 2012 - 2014 Intel Corporation. All rights reserved. * Copyright(c) 2013 - 2015 Intel Mobile Communications GmbH * Copyright(c) 2016 - 2017 Intel Deutschland GmbH + * Copyright(c) 2018 Intel Corporation * * This program is free software; you can redistribute it and/or modify * it under the terms of version 2 of the GNU General Public License as @@ -30,7 +31,7 @@ * Copyright(c) 2012 - 2014 Intel Corporation. All rights reserved. * Copyright(c) 2013 - 2015 Intel Mobile Communications GmbH * Copyright(c) 2016 - 2017 Intel Deutschland GmbH - * Copyright(c) 2018 Intel Corporation + * Copyright(c) 2018 Intel Corporation * All rights reserved. * * Redistribution and use in source and binary forms, with or without @@ -749,13 +750,9 @@ struct iwl_scan_req_umac { } __packed; #define IWL_SCAN_REQ_UMAC_SIZE_V8 sizeof(struct iwl_scan_req_umac) -#define IWL_SCAN_REQ_UMAC_SIZE_V7 (sizeof(struct iwl_scan_req_umac) - \ - 4 * sizeof(u8)) -#define IWL_SCAN_REQ_UMAC_SIZE_V6 (sizeof(struct iwl_scan_req_umac) - \ - 2 * sizeof(u8) - sizeof(__le16)) -#define IWL_SCAN_REQ_UMAC_SIZE_V1 (sizeof(struct iwl_scan_req_umac) - \ - 2 * sizeof(__le32) - 2 * sizeof(u8) - \ - sizeof(__le16)) +#define IWL_SCAN_REQ_UMAC_SIZE_V7 48 +#define IWL_SCAN_REQ_UMAC_SIZE_V6 44 +#define IWL_SCAN_REQ_UMAC_SIZE_V1 36 /** * struct iwl_umac_scan_abort diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c index 8928613e033e..ca0174680af9 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c +++ b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c @@ -76,6 +76,7 @@ #include "iwl-io.h" #include "iwl-csr.h" #include "fw/acpi.h" +#include "fw/api/nvm-reg.h" /* NVM offsets (in words) definitions */ enum nvm_offsets { @@ -146,8 +147,8 @@ static const u8 iwl_ext_nvm_channels[] = { 149, 153, 157, 161, 165, 169, 173, 177, 181 }; -#define IWL_NUM_CHANNELS ARRAY_SIZE(iwl_nvm_channels) -#define IWL_NUM_CHANNELS_EXT ARRAY_SIZE(iwl_ext_nvm_channels) +#define IWL_NVM_NUM_CHANNELS ARRAY_SIZE(iwl_nvm_channels) +#define IWL_NVM_NUM_CHANNELS_EXT ARRAY_SIZE(iwl_ext_nvm_channels) #define NUM_2GHZ_CHANNELS 14 #define NUM_2GHZ_CHANNELS_EXT 14 #define FIRST_2GHZ_HT_MINUS 5 @@ -301,11 +302,11 @@ static int iwl_init_channel_map(struct device *dev, const struct iwl_cfg *cfg, const u8 *nvm_chan; if (cfg->nvm_type != IWL_NVM_EXT) { - num_of_ch = IWL_NUM_CHANNELS; + num_of_ch = IWL_NVM_NUM_CHANNELS; nvm_chan = &iwl_nvm_channels[0]; num_2ghz_channels = NUM_2GHZ_CHANNELS; } else { - num_of_ch = IWL_NUM_CHANNELS_EXT; + num_of_ch = IWL_NVM_NUM_CHANNELS_EXT; nvm_chan = &iwl_ext_nvm_channels[0]; num_2ghz_channels = NUM_2GHZ_CHANNELS_EXT; } @@ -720,12 +721,12 @@ iwl_parse_nvm_data(struct iwl_trans *trans, const struct iwl_cfg *cfg, if (cfg->nvm_type != IWL_NVM_EXT) data = kzalloc(sizeof(*data) + sizeof(struct ieee80211_channel) * - IWL_NUM_CHANNELS, + IWL_NVM_NUM_CHANNELS, GFP_KERNEL); else data = kzalloc(sizeof(*data) + sizeof(struct ieee80211_channel) * - IWL_NUM_CHANNELS_EXT, + IWL_NVM_NUM_CHANNELS_EXT, GFP_KERNEL); if (!data) return NULL; @@ -842,24 +843,34 @@ static u32 iwl_nvm_get_regdom_bw_flags(const u8 *nvm_chan, return flags; } +struct regdb_ptrs { + struct ieee80211_wmm_rule *rule; + u32 token; +}; + struct ieee80211_regdomain * iwl_parse_nvm_mcc_info(struct device *dev, const struct iwl_cfg *cfg, - int num_of_ch, __le32 *channels, u16 fw_mcc) + int num_of_ch, __le32 *channels, u16 fw_mcc, + u16 geo_info) { int ch_idx; u16 ch_flags; u32 reg_rule_flags, prev_reg_rule_flags = 0; const u8 *nvm_chan = cfg->nvm_type == IWL_NVM_EXT ? iwl_ext_nvm_channels : iwl_nvm_channels; - struct ieee80211_regdomain *regd; - int size_of_regd; + struct ieee80211_regdomain *regd, *copy_rd; + int size_of_regd, regd_to_copy, wmms_to_copy; + int size_of_wmms = 0; struct ieee80211_reg_rule *rule; + struct ieee80211_wmm_rule *wmm_rule, *d_wmm, *s_wmm; + struct regdb_ptrs *regdb_ptrs; enum nl80211_band band; int center_freq, prev_center_freq = 0; - int valid_rules = 0; + int valid_rules = 0, n_wmms = 0; + int i; bool new_rule; int max_num_ch = cfg->nvm_type == IWL_NVM_EXT ? - IWL_NUM_CHANNELS_EXT : IWL_NUM_CHANNELS; + IWL_NVM_NUM_CHANNELS_EXT : IWL_NVM_NUM_CHANNELS; if (WARN_ON_ONCE(num_of_ch > NL80211_MAX_SUPP_REG_RULES)) return ERR_PTR(-EINVAL); @@ -875,10 +886,26 @@ iwl_parse_nvm_mcc_info(struct device *dev, const struct iwl_cfg *cfg, sizeof(struct ieee80211_regdomain) + num_of_ch * sizeof(struct ieee80211_reg_rule); - regd = kzalloc(size_of_regd, GFP_KERNEL); + if (geo_info & GEO_WMM_ETSI_5GHZ_INFO) + size_of_wmms = + num_of_ch * sizeof(struct ieee80211_wmm_rule); + + regd = kzalloc(size_of_regd + size_of_wmms, GFP_KERNEL); if (!regd) return ERR_PTR(-ENOMEM); + regdb_ptrs = kcalloc(num_of_ch, sizeof(*regdb_ptrs), GFP_KERNEL); + if (!regdb_ptrs) { + copy_rd = ERR_PTR(-ENOMEM); + goto out; + } + + /* set alpha2 from FW. */ + regd->alpha2[0] = fw_mcc >> 8; + regd->alpha2[1] = fw_mcc & 0xff; + + wmm_rule = (struct ieee80211_wmm_rule *)((u8 *)regd + size_of_regd); + for (ch_idx = 0; ch_idx < num_of_ch; ch_idx++) { ch_flags = (u16)__le32_to_cpup(channels + ch_idx); band = (ch_idx < NUM_2GHZ_CHANNELS) ? @@ -927,14 +954,66 @@ iwl_parse_nvm_mcc_info(struct device *dev, const struct iwl_cfg *cfg, iwl_nvm_print_channel_flags(dev, IWL_DL_LAR, nvm_chan[ch_idx], ch_flags); + + if (!(geo_info & GEO_WMM_ETSI_5GHZ_INFO) || + band == NL80211_BAND_2GHZ) + continue; + + if (!reg_query_regdb_wmm(regd->alpha2, center_freq, + ®db_ptrs[n_wmms].token, wmm_rule)) { + /* Add only new rules */ + for (i = 0; i < n_wmms; i++) { + if (regdb_ptrs[i].token == + regdb_ptrs[n_wmms].token) { + rule->wmm_rule = regdb_ptrs[i].rule; + break; + } + } + if (i == n_wmms) { + rule->wmm_rule = wmm_rule; + regdb_ptrs[n_wmms++].rule = wmm_rule; + wmm_rule++; + } + } } regd->n_reg_rules = valid_rules; + regd->n_wmm_rules = n_wmms; - /* set alpha2 from FW. */ - regd->alpha2[0] = fw_mcc >> 8; - regd->alpha2[1] = fw_mcc & 0xff; + /* + * Narrow down regdom for unused regulatory rules to prevent hole + * between reg rules to wmm rules. + */ + regd_to_copy = sizeof(struct ieee80211_regdomain) + + valid_rules * sizeof(struct ieee80211_reg_rule); + + wmms_to_copy = sizeof(struct ieee80211_wmm_rule) * n_wmms; + + copy_rd = kzalloc(regd_to_copy + wmms_to_copy, GFP_KERNEL); + if (!copy_rd) { + copy_rd = ERR_PTR(-ENOMEM); + goto out; + } + + memcpy(copy_rd, regd, regd_to_copy); + memcpy((u8 *)copy_rd + regd_to_copy, (u8 *)regd + size_of_regd, + wmms_to_copy); + + d_wmm = (struct ieee80211_wmm_rule *)((u8 *)copy_rd + regd_to_copy); + s_wmm = (struct ieee80211_wmm_rule *)((u8 *)regd + size_of_regd); + + for (i = 0; i < regd->n_reg_rules; i++) { + if (!regd->reg_rules[i].wmm_rule) + continue; + + copy_rd->reg_rules[i].wmm_rule = d_wmm + + (regd->reg_rules[i].wmm_rule - s_wmm) / + sizeof(struct ieee80211_wmm_rule); + } - return regd; +out: + kfree(regdb_ptrs); + kfree(regd); + return copy_rd; } IWL_EXPORT_SYMBOL(iwl_parse_nvm_mcc_info); diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.h b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.h index 306736c7a042..3071a23b7606 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.h +++ b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.h @@ -101,12 +101,14 @@ void iwl_init_sbands(struct device *dev, const struct iwl_cfg *cfg, * * This function parses the regulatory channel data received as a * MCC_UPDATE_CMD command. It returns a newly allocation regulatory domain, - * to be fed into the regulatory core. An ERR_PTR is returned on error. + * to be fed into the regulatory core. In case the geo_info is set handle + * accordingly. An ERR_PTR is returned on error. * If not given to the regulatory core, the user is responsible for freeing * the regdomain returned here with kfree. */ struct ieee80211_regdomain * iwl_parse_nvm_mcc_info(struct device *dev, const struct iwl_cfg *cfg, - int num_of_ch, __le32 *channels, u16 fw_mcc); + int num_of_ch, __le32 *channels, u16 fw_mcc, + u16 geo_info); #endif /* __iwl_nvm_parse_h__ */ diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c index 51b30424575b..90f8c89ea59c 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c @@ -311,7 +311,8 @@ struct ieee80211_regdomain *iwl_mvm_get_regdomain(struct wiphy *wiphy, regd = iwl_parse_nvm_mcc_info(mvm->trans->dev, mvm->cfg, __le32_to_cpu(resp->n_channels), resp->channels, - __le16_to_cpu(resp->mcc)); + __le16_to_cpu(resp->mcc), + __le16_to_cpu(resp->geo_info)); /* Store the return source id */ src_id = resp->source_id; kfree(resp); diff --git a/drivers/net/wireless/mac80211_hwsim.c b/drivers/net/wireless/mac80211_hwsim.c index 96d26cfae90b..4a017a0d71ea 100644 --- a/drivers/net/wireless/mac80211_hwsim.c +++ b/drivers/net/wireless/mac80211_hwsim.c @@ -3236,6 +3236,7 @@ static int hwsim_new_radio_nl(struct sk_buff *msg, struct genl_info *info) GENL_SET_ERR_MSG(info,"MAC is no valid source addr"); NL_SET_BAD_ATTR(info->extack, info->attrs[HWSIM_ATTR_PERM_ADDR]); + kfree(hwname); return -EINVAL; } diff --git a/drivers/net/wireless/realtek/rtlwifi/btcoexist/halbtcoutsrc.c b/drivers/net/wireless/realtek/rtlwifi/btcoexist/halbtcoutsrc.c index 8b6b07a936f5..b026e80940a4 100644 --- a/drivers/net/wireless/realtek/rtlwifi/btcoexist/halbtcoutsrc.c +++ b/drivers/net/wireless/realtek/rtlwifi/btcoexist/halbtcoutsrc.c @@ -158,16 +158,6 @@ static u8 halbtc_get_wifi_central_chnl(struct btc_coexist *btcoexist) static u8 rtl_get_hwpg_single_ant_path(struct rtl_priv *rtlpriv) { - struct rtl_mod_params *mod_params = rtlpriv->cfg->mod_params; - - /* override ant_num / ant_path */ - if (mod_params->ant_sel) { - rtlpriv->btcoexist.btc_info.ant_num = - (mod_params->ant_sel == 1 ? ANT_X2 : ANT_X1); - - rtlpriv->btcoexist.btc_info.single_ant_path = - (mod_params->ant_sel == 1 ? 0 : 1); - } return rtlpriv->btcoexist.btc_info.single_ant_path; } @@ -178,7 +168,6 @@ static u8 rtl_get_hwpg_bt_type(struct rtl_priv *rtlpriv) static u8 rtl_get_hwpg_ant_num(struct rtl_priv *rtlpriv) { - struct rtl_mod_params *mod_params = rtlpriv->cfg->mod_params; u8 num; if (rtlpriv->btcoexist.btc_info.ant_num == ANT_X2) @@ -186,10 +175,6 @@ static u8 rtl_get_hwpg_ant_num(struct rtl_priv *rtlpriv) else num = 1; - /* override ant_num / ant_path */ - if (mod_params->ant_sel) - num = (mod_params->ant_sel == 1 ? ANT_X2 : ANT_X1) + 1; - return num; } diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8723be/hw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8723be/hw.c index e7bbbc95cdb1..b4f3f91b590e 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8723be/hw.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8723be/hw.c @@ -848,6 +848,9 @@ static bool _rtl8723be_init_mac(struct ieee80211_hw *hw) return false; } + if (rtlpriv->cfg->ops->get_btc_status()) + rtlpriv->btcoexist.btc_ops->btc_power_on_setting(rtlpriv); + bytetmp = rtl_read_byte(rtlpriv, REG_MULTI_FUNC_CTRL); rtl_write_byte(rtlpriv, REG_MULTI_FUNC_CTRL, bytetmp | BIT(3)); @@ -2696,21 +2699,21 @@ void rtl8723be_read_bt_coexist_info_from_hwpg(struct ieee80211_hw *hw, rtlpriv->btcoexist.btc_info.bt_type = BT_RTL8723B; rtlpriv->btcoexist.btc_info.ant_num = (value & 0x1); rtlpriv->btcoexist.btc_info.single_ant_path = - (value & 0x40); /*0xc3[6]*/ + (value & 0x40 ? ANT_AUX : ANT_MAIN); /*0xc3[6]*/ } else { rtlpriv->btcoexist.btc_info.btcoexist = 0; rtlpriv->btcoexist.btc_info.bt_type = BT_RTL8723B; rtlpriv->btcoexist.btc_info.ant_num = ANT_X2; - rtlpriv->btcoexist.btc_info.single_ant_path = 0; + rtlpriv->btcoexist.btc_info.single_ant_path = ANT_MAIN; } /* override ant_num / ant_path */ if (mod_params->ant_sel) { rtlpriv->btcoexist.btc_info.ant_num = - (mod_params->ant_sel == 1 ? ANT_X2 : ANT_X1); + (mod_params->ant_sel == 1 ? ANT_X1 : ANT_X2); rtlpriv->btcoexist.btc_info.single_ant_path = - (mod_params->ant_sel == 1 ? 0 : 1); + (mod_params->ant_sel == 1 ? ANT_AUX : ANT_MAIN); } } diff --git a/drivers/net/wireless/realtek/rtlwifi/wifi.h b/drivers/net/wireless/realtek/rtlwifi/wifi.h index d27e33960e77..ce1754054a07 100644 --- a/drivers/net/wireless/realtek/rtlwifi/wifi.h +++ b/drivers/net/wireless/realtek/rtlwifi/wifi.h @@ -2823,6 +2823,11 @@ enum bt_ant_num { ANT_X1 = 1, }; +enum bt_ant_path { + ANT_MAIN = 0, + ANT_AUX = 1, +}; + enum bt_co_type { BT_2WIRE = 0, BT_ISSC_3WIRE = 1, diff --git a/drivers/nvme/host/Kconfig b/drivers/nvme/host/Kconfig index b979cf3bce65..88a8b5916624 100644 --- a/drivers/nvme/host/Kconfig +++ b/drivers/nvme/host/Kconfig @@ -27,7 +27,7 @@ config NVME_FABRICS config NVME_RDMA tristate "NVM Express over Fabrics RDMA host driver" - depends on INFINIBAND && BLOCK + depends on INFINIBAND && INFINIBAND_ADDR_TRANS && BLOCK select NVME_CORE select NVME_FABRICS select SG_POOL diff --git a/drivers/nvme/host/core.c b/drivers/nvme/host/core.c index 9df4f71e58ca..99b857e5a7a9 100644 --- a/drivers/nvme/host/core.c +++ b/drivers/nvme/host/core.c @@ -99,6 +99,7 @@ static struct class *nvme_subsys_class; static void nvme_ns_remove(struct nvme_ns *ns); static int nvme_revalidate_disk(struct gendisk *disk); +static void nvme_put_subsystem(struct nvme_subsystem *subsys); int nvme_reset_ctrl(struct nvme_ctrl *ctrl) { @@ -117,7 +118,8 @@ int nvme_reset_ctrl_sync(struct nvme_ctrl *ctrl) ret = nvme_reset_ctrl(ctrl); if (!ret) { flush_work(&ctrl->reset_work); - if (ctrl->state != NVME_CTRL_LIVE) + if (ctrl->state != NVME_CTRL_LIVE && + ctrl->state != NVME_CTRL_ADMIN_ONLY) ret = -ENETRESET; } @@ -350,6 +352,7 @@ static void nvme_free_ns_head(struct kref *ref) ida_simple_remove(&head->subsys->ns_ida, head->instance); list_del_init(&head->entry); cleanup_srcu_struct(&head->srcu); + nvme_put_subsystem(head->subsys); kfree(head); } @@ -764,6 +767,7 @@ static int nvme_submit_user_cmd(struct request_queue *q, ret = PTR_ERR(meta); goto out_unmap; } + req->cmd_flags |= REQ_INTEGRITY; } } @@ -2860,6 +2864,9 @@ static struct nvme_ns_head *nvme_alloc_ns_head(struct nvme_ctrl *ctrl, goto out_cleanup_srcu; list_add_tail(&head->entry, &ctrl->subsys->nsheads); + + kref_get(&ctrl->subsys->ref); + return head; out_cleanup_srcu: cleanup_srcu_struct(&head->srcu); @@ -2997,31 +3004,7 @@ static void nvme_alloc_ns(struct nvme_ctrl *ctrl, unsigned nsid) if (nvme_init_ns_head(ns, nsid, id)) goto out_free_id; nvme_setup_streams_ns(ctrl, ns); - -#ifdef CONFIG_NVME_MULTIPATH - /* - * If multipathing is enabled we need to always use the subsystem - * instance number for numbering our devices to avoid conflicts - * between subsystems that have multiple controllers and thus use - * the multipath-aware subsystem node and those that have a single - * controller and use the controller node directly. - */ - if (ns->head->disk) { - sprintf(disk_name, "nvme%dc%dn%d", ctrl->subsys->instance, - ctrl->cntlid, ns->head->instance); - flags = GENHD_FL_HIDDEN; - } else { - sprintf(disk_name, "nvme%dn%d", ctrl->subsys->instance, - ns->head->instance); - } -#else - /* - * But without the multipath code enabled, multiple controller per - * subsystems are visible as devices and thus we cannot use the - * subsystem instance. - */ - sprintf(disk_name, "nvme%dn%d", ctrl->instance, ns->head->instance); -#endif + nvme_set_disk_name(disk_name, ns, ctrl, &flags); if ((ctrl->quirks & NVME_QUIRK_LIGHTNVM) && id->vs[0] == 0x1) { if (nvme_nvm_register(ns, disk_name, node)) { diff --git a/drivers/nvme/host/fabrics.c b/drivers/nvme/host/fabrics.c index 124c458806df..7ae732a77fe8 100644 --- a/drivers/nvme/host/fabrics.c +++ b/drivers/nvme/host/fabrics.c @@ -668,6 +668,7 @@ static int nvmf_parse_options(struct nvmf_ctrl_options *opts, ret = -ENOMEM; goto out; } + kfree(opts->transport); opts->transport = p; break; case NVMF_OPT_NQN: @@ -676,6 +677,7 @@ static int nvmf_parse_options(struct nvmf_ctrl_options *opts, ret = -ENOMEM; goto out; } + kfree(opts->subsysnqn); opts->subsysnqn = p; nqnlen = strlen(opts->subsysnqn); if (nqnlen >= NVMF_NQN_SIZE) { @@ -698,6 +700,7 @@ static int nvmf_parse_options(struct nvmf_ctrl_options *opts, ret = -ENOMEM; goto out; } + kfree(opts->traddr); opts->traddr = p; break; case NVMF_OPT_TRSVCID: @@ -706,6 +709,7 @@ static int nvmf_parse_options(struct nvmf_ctrl_options *opts, ret = -ENOMEM; goto out; } + kfree(opts->trsvcid); opts->trsvcid = p; break; case NVMF_OPT_QUEUE_SIZE: @@ -792,6 +796,7 @@ static int nvmf_parse_options(struct nvmf_ctrl_options *opts, ret = -EINVAL; goto out; } + nvmf_host_put(opts->host); opts->host = nvmf_host_add(p); kfree(p); if (!opts->host) { @@ -817,6 +822,7 @@ static int nvmf_parse_options(struct nvmf_ctrl_options *opts, ret = -ENOMEM; goto out; } + kfree(opts->host_traddr); opts->host_traddr = p; break; case NVMF_OPT_HOST_ID: diff --git a/drivers/nvme/host/multipath.c b/drivers/nvme/host/multipath.c index 956e0b8e9c4d..d7b664ae5923 100644 --- a/drivers/nvme/host/multipath.c +++ b/drivers/nvme/host/multipath.c @@ -15,10 +15,32 @@ #include "nvme.h" static bool multipath = true; -module_param(multipath, bool, 0644); +module_param(multipath, bool, 0444); MODULE_PARM_DESC(multipath, "turn on native support for multiple controllers per subsystem"); +/* + * If multipathing is enabled we need to always use the subsystem instance + * number for numbering our devices to avoid conflicts between subsystems that + * have multiple controllers and thus use the multipath-aware subsystem node + * and those that have a single controller and use the controller node + * directly. + */ +void nvme_set_disk_name(char *disk_name, struct nvme_ns *ns, + struct nvme_ctrl *ctrl, int *flags) +{ + if (!multipath) { + sprintf(disk_name, "nvme%dn%d", ctrl->instance, ns->head->instance); + } else if (ns->head->disk) { + sprintf(disk_name, "nvme%dc%dn%d", ctrl->subsys->instance, + ctrl->cntlid, ns->head->instance); + *flags = GENHD_FL_HIDDEN; + } else { + sprintf(disk_name, "nvme%dn%d", ctrl->subsys->instance, + ns->head->instance); + } +} + void nvme_failover_req(struct request *req) { struct nvme_ns *ns = req->q->queuedata; diff --git a/drivers/nvme/host/nvme.h b/drivers/nvme/host/nvme.h index 061fecfd44f5..17d2f7cf3fed 100644 --- a/drivers/nvme/host/nvme.h +++ b/drivers/nvme/host/nvme.h @@ -84,6 +84,11 @@ enum nvme_quirks { * Supports the LighNVM command set if indicated in vs[1]. */ NVME_QUIRK_LIGHTNVM = (1 << 6), + + /* + * Set MEDIUM priority on SQ creation + */ + NVME_QUIRK_MEDIUM_PRIO_SQ = (1 << 7), }; /* @@ -436,6 +441,8 @@ extern const struct attribute_group nvme_ns_id_attr_group; extern const struct block_device_operations nvme_ns_head_ops; #ifdef CONFIG_NVME_MULTIPATH +void nvme_set_disk_name(char *disk_name, struct nvme_ns *ns, + struct nvme_ctrl *ctrl, int *flags); void nvme_failover_req(struct request *req); bool nvme_req_needs_failover(struct request *req, blk_status_t error); void nvme_kick_requeue_lists(struct nvme_ctrl *ctrl); @@ -461,6 +468,16 @@ static inline void nvme_mpath_check_last_path(struct nvme_ns *ns) } #else +/* + * Without the multipath code enabled, multiple controller per subsystems are + * visible as devices and thus we cannot use the subsystem instance. + */ +static inline void nvme_set_disk_name(char *disk_name, struct nvme_ns *ns, + struct nvme_ctrl *ctrl, int *flags) +{ + sprintf(disk_name, "nvme%dn%d", ctrl->instance, ns->head->instance); +} + static inline void nvme_failover_req(struct request *req) { } diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c index fbc71fac6f1e..17a0190bd88f 100644 --- a/drivers/nvme/host/pci.c +++ b/drivers/nvme/host/pci.c @@ -1093,10 +1093,19 @@ static int adapter_alloc_cq(struct nvme_dev *dev, u16 qid, static int adapter_alloc_sq(struct nvme_dev *dev, u16 qid, struct nvme_queue *nvmeq) { + struct nvme_ctrl *ctrl = &dev->ctrl; struct nvme_command c; int flags = NVME_QUEUE_PHYS_CONTIG; /* + * Some drives have a bug that auto-enables WRRU if MEDIUM isn't + * set. Since URGENT priority is zeroes, it makes all queues + * URGENT. + */ + if (ctrl->quirks & NVME_QUIRK_MEDIUM_PRIO_SQ) + flags |= NVME_SQ_PRIO_MEDIUM; + + /* * Note: we (ab)use the fact that the prp fields survive if no data * is attached to the request. */ @@ -2701,7 +2710,8 @@ static const struct pci_device_id nvme_id_table[] = { .driver_data = NVME_QUIRK_STRIPE_SIZE | NVME_QUIRK_DEALLOCATE_ZEROES, }, { PCI_VDEVICE(INTEL, 0xf1a5), /* Intel 600P/P3100 */ - .driver_data = NVME_QUIRK_NO_DEEPEST_PS }, + .driver_data = NVME_QUIRK_NO_DEEPEST_PS | + NVME_QUIRK_MEDIUM_PRIO_SQ }, { PCI_VDEVICE(INTEL, 0x5845), /* Qemu emulated controller */ .driver_data = NVME_QUIRK_IDENTIFY_CNS, }, { PCI_DEVICE(0x1c58, 0x0003), /* HGST adapter */ diff --git a/drivers/nvme/target/Kconfig b/drivers/nvme/target/Kconfig index 5f4f8b16685f..3c7b61ddb0d1 100644 --- a/drivers/nvme/target/Kconfig +++ b/drivers/nvme/target/Kconfig @@ -27,7 +27,7 @@ config NVME_TARGET_LOOP config NVME_TARGET_RDMA tristate "NVMe over Fabrics RDMA target support" - depends on INFINIBAND + depends on INFINIBAND && INFINIBAND_ADDR_TRANS depends on NVME_TARGET select SGL_ALLOC help diff --git a/drivers/nvme/target/loop.c b/drivers/nvme/target/loop.c index 31fdfba556a8..27a8561c0cb9 100644 --- a/drivers/nvme/target/loop.c +++ b/drivers/nvme/target/loop.c @@ -469,6 +469,12 @@ static void nvme_loop_reset_ctrl_work(struct work_struct *work) nvme_stop_ctrl(&ctrl->ctrl); nvme_loop_shutdown_ctrl(ctrl); + if (!nvme_change_ctrl_state(&ctrl->ctrl, NVME_CTRL_CONNECTING)) { + /* state change failure should never happen */ + WARN_ON_ONCE(1); + return; + } + ret = nvme_loop_configure_admin_queue(ctrl); if (ret) goto out_disable; diff --git a/drivers/of/overlay.c b/drivers/of/overlay.c index b35fe88f1851..7baa53e5b1d7 100644 --- a/drivers/of/overlay.c +++ b/drivers/of/overlay.c @@ -102,12 +102,28 @@ static DEFINE_IDR(ovcs_idr); static BLOCKING_NOTIFIER_HEAD(overlay_notify_chain); +/** + * of_overlay_notifier_register() - Register notifier for overlay operations + * @nb: Notifier block to register + * + * Register for notification on overlay operations on device tree nodes. The + * reported actions definied by @of_reconfig_change. The notifier callback + * furthermore receives a pointer to the affected device tree node. + * + * Note that a notifier callback is not supposed to store pointers to a device + * tree node or its content beyond @OF_OVERLAY_POST_REMOVE corresponding to the + * respective node it received. + */ int of_overlay_notifier_register(struct notifier_block *nb) { return blocking_notifier_chain_register(&overlay_notify_chain, nb); } EXPORT_SYMBOL_GPL(of_overlay_notifier_register); +/** + * of_overlay_notifier_register() - Unregister notifier for overlay operations + * @nb: Notifier block to unregister + */ int of_overlay_notifier_unregister(struct notifier_block *nb) { return blocking_notifier_chain_unregister(&overlay_notify_chain, nb); @@ -671,17 +687,13 @@ static void free_overlay_changeset(struct overlay_changeset *ovcs) of_node_put(ovcs->fragments[i].overlay); } kfree(ovcs->fragments); - /* - * TODO - * - * would like to: kfree(ovcs->overlay_tree); - * but can not since drivers may have pointers into this data - * - * would like to: kfree(ovcs->fdt); - * but can not since drivers may have pointers into this data + * There should be no live pointers into ovcs->overlay_tree and + * ovcs->fdt due to the policy that overlay notifiers are not allowed + * to retain pointers into the overlay devicetree. */ - + kfree(ovcs->overlay_tree); + kfree(ovcs->fdt); kfree(ovcs); } diff --git a/drivers/parisc/ccio-dma.c b/drivers/parisc/ccio-dma.c index acba1f56af3e..126cf19e869b 100644 --- a/drivers/parisc/ccio-dma.c +++ b/drivers/parisc/ccio-dma.c @@ -1263,7 +1263,7 @@ static struct parisc_driver ccio_driver __refdata = { * I/O Page Directory, the resource map, and initalizing the * U2/Uturn chip into virtual mode. */ -static void +static void __init ccio_ioc_init(struct ioc *ioc) { int i; diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c index a04197ce767d..dbfe7c4f3776 100644 --- a/drivers/pci/pci.c +++ b/drivers/pci/pci.c @@ -1910,7 +1910,7 @@ void pci_pme_active(struct pci_dev *dev, bool enable) EXPORT_SYMBOL(pci_pme_active); /** - * pci_enable_wake - enable PCI device as wakeup event source + * __pci_enable_wake - enable PCI device as wakeup event source * @dev: PCI device affected * @state: PCI state from which device will issue wakeup events * @enable: True to enable event generation; false to disable @@ -1928,7 +1928,7 @@ EXPORT_SYMBOL(pci_pme_active); * Error code depending on the platform is returned if both the platform and * the native mechanism fail to enable the generation of wake-up events */ -int pci_enable_wake(struct pci_dev *dev, pci_power_t state, bool enable) +static int __pci_enable_wake(struct pci_dev *dev, pci_power_t state, bool enable) { int ret = 0; @@ -1969,6 +1969,23 @@ int pci_enable_wake(struct pci_dev *dev, pci_power_t state, bool enable) return ret; } + +/** + * pci_enable_wake - change wakeup settings for a PCI device + * @pci_dev: Target device + * @state: PCI state from which device will issue wakeup events + * @enable: Whether or not to enable event generation + * + * If @enable is set, check device_may_wakeup() for the device before calling + * __pci_enable_wake() for it. + */ +int pci_enable_wake(struct pci_dev *pci_dev, pci_power_t state, bool enable) +{ + if (enable && !device_may_wakeup(&pci_dev->dev)) + return -EINVAL; + + return __pci_enable_wake(pci_dev, state, enable); +} EXPORT_SYMBOL(pci_enable_wake); /** @@ -1981,9 +1998,9 @@ EXPORT_SYMBOL(pci_enable_wake); * should not be called twice in a row to enable wake-up due to PCI PM vs ACPI * ordering constraints. * - * This function only returns error code if the device is not capable of - * generating PME# from both D3_hot and D3_cold, and the platform is unable to - * enable wake-up power for it. + * This function only returns error code if the device is not allowed to wake + * up the system from sleep or it is not capable of generating PME# from both + * D3_hot and D3_cold and the platform is unable to enable wake-up power for it. */ int pci_wake_from_d3(struct pci_dev *dev, bool enable) { @@ -2114,7 +2131,7 @@ int pci_finish_runtime_suspend(struct pci_dev *dev) dev->runtime_d3cold = target_state == PCI_D3cold; - pci_enable_wake(dev, target_state, pci_dev_run_wake(dev)); + __pci_enable_wake(dev, target_state, pci_dev_run_wake(dev)); error = pci_set_power_state(dev, target_state); @@ -2138,16 +2155,16 @@ bool pci_dev_run_wake(struct pci_dev *dev) { struct pci_bus *bus = dev->bus; - if (device_can_wakeup(&dev->dev)) - return true; - if (!dev->pme_support) return false; /* PME-capable in principle, but not from the target power state */ - if (!pci_pme_capable(dev, pci_target_state(dev, false))) + if (!pci_pme_capable(dev, pci_target_state(dev, true))) return false; + if (device_can_wakeup(&dev->dev)) + return true; + while (bus->parent) { struct pci_dev *bridge = bus->self; diff --git a/drivers/pinctrl/intel/pinctrl-cherryview.c b/drivers/pinctrl/intel/pinctrl-cherryview.c index b1ae1618fefe..fee9225ca559 100644 --- a/drivers/pinctrl/intel/pinctrl-cherryview.c +++ b/drivers/pinctrl/intel/pinctrl-cherryview.c @@ -1622,22 +1622,30 @@ static int chv_gpio_probe(struct chv_pinctrl *pctrl, int irq) if (!need_valid_mask) { irq_base = devm_irq_alloc_descs(pctrl->dev, -1, 0, - chip->ngpio, NUMA_NO_NODE); + community->npins, NUMA_NO_NODE); if (irq_base < 0) { dev_err(pctrl->dev, "Failed to allocate IRQ numbers\n"); return irq_base; } - } else { - irq_base = 0; } - ret = gpiochip_irqchip_add(chip, &chv_gpio_irqchip, irq_base, + ret = gpiochip_irqchip_add(chip, &chv_gpio_irqchip, 0, handle_bad_irq, IRQ_TYPE_NONE); if (ret) { dev_err(pctrl->dev, "failed to add IRQ chip\n"); return ret; } + if (!need_valid_mask) { + for (i = 0; i < community->ngpio_ranges; i++) { + range = &community->gpio_ranges[i]; + + irq_domain_associate_many(chip->irq.domain, irq_base, + range->base, range->npins); + irq_base += range->npins; + } + } + gpiochip_set_chained_irqchip(chip, &chv_gpio_irqchip, irq, chv_gpio_irq_handler); return 0; diff --git a/drivers/pinctrl/intel/pinctrl-sunrisepoint.c b/drivers/pinctrl/intel/pinctrl-sunrisepoint.c index 8870a4100164..fee3435a6f15 100644 --- a/drivers/pinctrl/intel/pinctrl-sunrisepoint.c +++ b/drivers/pinctrl/intel/pinctrl-sunrisepoint.c @@ -36,6 +36,27 @@ .npins = ((e) - (s) + 1), \ } +#define SPTH_GPP(r, s, e, g) \ + { \ + .reg_num = (r), \ + .base = (s), \ + .size = ((e) - (s) + 1), \ + .gpio_base = (g), \ + } + +#define SPTH_COMMUNITY(b, s, e, g) \ + { \ + .barno = (b), \ + .padown_offset = SPT_PAD_OWN, \ + .padcfglock_offset = SPT_PADCFGLOCK, \ + .hostown_offset = SPT_HOSTSW_OWN, \ + .ie_offset = SPT_GPI_IE, \ + .pin_base = (s), \ + .npins = ((e) - (s) + 1), \ + .gpps = (g), \ + .ngpps = ARRAY_SIZE(g), \ + } + /* Sunrisepoint-LP */ static const struct pinctrl_pin_desc sptlp_pins[] = { /* GPP_A */ @@ -531,10 +552,28 @@ static const struct intel_function spth_functions[] = { FUNCTION("i2c2", spth_i2c2_groups), }; +static const struct intel_padgroup spth_community0_gpps[] = { + SPTH_GPP(0, 0, 23, 0), /* GPP_A */ + SPTH_GPP(1, 24, 47, 24), /* GPP_B */ +}; + +static const struct intel_padgroup spth_community1_gpps[] = { + SPTH_GPP(0, 48, 71, 48), /* GPP_C */ + SPTH_GPP(1, 72, 95, 72), /* GPP_D */ + SPTH_GPP(2, 96, 108, 96), /* GPP_E */ + SPTH_GPP(3, 109, 132, 120), /* GPP_F */ + SPTH_GPP(4, 133, 156, 144), /* GPP_G */ + SPTH_GPP(5, 157, 180, 168), /* GPP_H */ +}; + +static const struct intel_padgroup spth_community3_gpps[] = { + SPTH_GPP(0, 181, 191, 192), /* GPP_I */ +}; + static const struct intel_community spth_communities[] = { - SPT_COMMUNITY(0, 0, 47), - SPT_COMMUNITY(1, 48, 180), - SPT_COMMUNITY(2, 181, 191), + SPTH_COMMUNITY(0, 0, 47, spth_community0_gpps), + SPTH_COMMUNITY(1, 48, 180, spth_community1_gpps), + SPTH_COMMUNITY(2, 181, 191, spth_community3_gpps), }; static const struct intel_pinctrl_soc_data spth_soc_data = { diff --git a/drivers/pinctrl/meson/pinctrl-meson-axg.c b/drivers/pinctrl/meson/pinctrl-meson-axg.c index 4b91ff74779b..99a6ceac8e53 100644 --- a/drivers/pinctrl/meson/pinctrl-meson-axg.c +++ b/drivers/pinctrl/meson/pinctrl-meson-axg.c @@ -898,7 +898,7 @@ static struct meson_bank meson_axg_periphs_banks[] = { static struct meson_bank meson_axg_aobus_banks[] = { /* name first last irq pullen pull dir out in */ - BANK("AO", GPIOAO_0, GPIOAO_9, 0, 13, 0, 16, 0, 0, 0, 0, 0, 16, 1, 0), + BANK("AO", GPIOAO_0, GPIOAO_13, 0, 13, 0, 16, 0, 0, 0, 0, 0, 16, 1, 0), }; static struct meson_pmx_bank meson_axg_periphs_pmx_banks[] = { diff --git a/drivers/platform/x86/Kconfig b/drivers/platform/x86/Kconfig index 39d06dd1f63a..bc309c5327ff 100644 --- a/drivers/platform/x86/Kconfig +++ b/drivers/platform/x86/Kconfig @@ -154,7 +154,7 @@ config DELL_LAPTOP depends on ACPI_VIDEO || ACPI_VIDEO = n depends on RFKILL || RFKILL = n depends on SERIO_I8042 - select DELL_SMBIOS + depends on DELL_SMBIOS select POWER_SUPPLY select LEDS_CLASS select NEW_LEDS diff --git a/drivers/platform/x86/asus-wireless.c b/drivers/platform/x86/asus-wireless.c index d4aeac3477f5..f086469ea740 100644 --- a/drivers/platform/x86/asus-wireless.c +++ b/drivers/platform/x86/asus-wireless.c @@ -178,8 +178,10 @@ static int asus_wireless_remove(struct acpi_device *adev) { struct asus_wireless_data *data = acpi_driver_data(adev); - if (data->wq) + if (data->wq) { + devm_led_classdev_unregister(&adev->dev, &data->led); destroy_workqueue(data->wq); + } return 0; } diff --git a/drivers/remoteproc/qcom_q6v5_pil.c b/drivers/remoteproc/qcom_q6v5_pil.c index 8e70a627e0bb..cbbafdcaaecb 100644 --- a/drivers/remoteproc/qcom_q6v5_pil.c +++ b/drivers/remoteproc/qcom_q6v5_pil.c @@ -1083,6 +1083,7 @@ static int q6v5_alloc_memory_region(struct q6v5 *qproc) dev_err(qproc->dev, "unable to resolve mba region\n"); return ret; } + of_node_put(node); qproc->mba_phys = r.start; qproc->mba_size = resource_size(&r); @@ -1100,6 +1101,7 @@ static int q6v5_alloc_memory_region(struct q6v5 *qproc) dev_err(qproc->dev, "unable to resolve mpss region\n"); return ret; } + of_node_put(node); qproc->mpss_phys = qproc->mpss_reloc = r.start; qproc->mpss_size = resource_size(&r); diff --git a/drivers/remoteproc/remoteproc_core.c b/drivers/remoteproc/remoteproc_core.c index 6d9c5832ce47..a9609d971f7f 100644 --- a/drivers/remoteproc/remoteproc_core.c +++ b/drivers/remoteproc/remoteproc_core.c @@ -1163,7 +1163,7 @@ int rproc_trigger_recovery(struct rproc *rproc) if (ret) return ret; - ret = rproc_stop(rproc, false); + ret = rproc_stop(rproc, true); if (ret) goto unlock_mutex; @@ -1316,7 +1316,7 @@ void rproc_shutdown(struct rproc *rproc) if (!atomic_dec_and_test(&rproc->power)) goto out; - ret = rproc_stop(rproc, true); + ret = rproc_stop(rproc, false); if (ret) { atomic_inc(&rproc->power); goto out; diff --git a/drivers/rpmsg/rpmsg_char.c b/drivers/rpmsg/rpmsg_char.c index 64b6de9763ee..1efdf9ff8679 100644 --- a/drivers/rpmsg/rpmsg_char.c +++ b/drivers/rpmsg/rpmsg_char.c @@ -581,4 +581,6 @@ static void rpmsg_chrdev_exit(void) unregister_chrdev_region(rpmsg_major, RPMSG_DEV_MAX); } module_exit(rpmsg_chrdev_exit); + +MODULE_ALIAS("rpmsg:rpmsg_chrdev"); MODULE_LICENSE("GPL v2"); diff --git a/drivers/sbus/char/oradax.c b/drivers/sbus/char/oradax.c index c44d7c7ffc92..1754f55e2fac 100644 --- a/drivers/sbus/char/oradax.c +++ b/drivers/sbus/char/oradax.c @@ -3,7 +3,7 @@ * * This program is free software: you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by - * the Free Software Foundation, either version 3 of the License, or + * the Free Software Foundation, either version 2 of the License, or * (at your option) any later version. * * This program is distributed in the hope that it will be useful, diff --git a/drivers/scsi/isci/port_config.c b/drivers/scsi/isci/port_config.c index edb7be786c65..9e8de1462593 100644 --- a/drivers/scsi/isci/port_config.c +++ b/drivers/scsi/isci/port_config.c @@ -291,7 +291,7 @@ sci_mpc_agent_validate_phy_configuration(struct isci_host *ihost, * Note: We have not moved the current phy_index so we will actually * compare the startting phy with itself. * This is expected and required to add the phy to the port. */ - while (phy_index < SCI_MAX_PHYS) { + for (; phy_index < SCI_MAX_PHYS; phy_index++) { if ((phy_mask & (1 << phy_index)) == 0) continue; sci_phy_get_sas_address(&ihost->phys[phy_index], @@ -311,7 +311,6 @@ sci_mpc_agent_validate_phy_configuration(struct isci_host *ihost, &ihost->phys[phy_index]); assigned_phy_mask |= (1 << phy_index); - phy_index++; } } diff --git a/drivers/scsi/storvsc_drv.c b/drivers/scsi/storvsc_drv.c index 8c51d628b52e..a2ec0bc9e9fa 100644 --- a/drivers/scsi/storvsc_drv.c +++ b/drivers/scsi/storvsc_drv.c @@ -1722,11 +1722,14 @@ static int storvsc_probe(struct hv_device *device, max_targets = STORVSC_MAX_TARGETS; max_channels = STORVSC_MAX_CHANNELS; /* - * On Windows8 and above, we support sub-channels for storage. + * On Windows8 and above, we support sub-channels for storage + * on SCSI and FC controllers. * The number of sub-channels offerred is based on the number of * VCPUs in the guest. */ - max_sub_channels = (num_cpus / storvsc_vcpus_per_sub_channel); + if (!dev_is_ide) + max_sub_channels = + (num_cpus - 1) / storvsc_vcpus_per_sub_channel; } scsi_driver.can_queue = (max_outstanding_req_per_channel * diff --git a/drivers/staging/media/imx/imx-media-csi.c b/drivers/staging/media/imx/imx-media-csi.c index 16cab40156ca..aeab05f682d9 100644 --- a/drivers/staging/media/imx/imx-media-csi.c +++ b/drivers/staging/media/imx/imx-media-csi.c @@ -1799,7 +1799,7 @@ static int imx_csi_probe(struct platform_device *pdev) priv->dev->of_node = pdata->of_node; pinctrl = devm_pinctrl_get_select_default(priv->dev); if (IS_ERR(pinctrl)) { - ret = PTR_ERR(priv->vdev); + ret = PTR_ERR(pinctrl); dev_dbg(priv->dev, "devm_pinctrl_get_select_default() failed: %d\n", ret); if (ret != -ENODEV) diff --git a/drivers/target/target_core_iblock.c b/drivers/target/target_core_iblock.c index 07c814c42648..60429011292a 100644 --- a/drivers/target/target_core_iblock.c +++ b/drivers/target/target_core_iblock.c @@ -427,8 +427,8 @@ iblock_execute_zero_out(struct block_device *bdev, struct se_cmd *cmd) { struct se_device *dev = cmd->se_dev; struct scatterlist *sg = &cmd->t_data_sg[0]; - unsigned char *buf, zero = 0x00, *p = &zero; - int rc, ret; + unsigned char *buf, *not_zero; + int ret; buf = kmap(sg_page(sg)) + sg->offset; if (!buf) @@ -437,10 +437,10 @@ iblock_execute_zero_out(struct block_device *bdev, struct se_cmd *cmd) * Fall back to block_execute_write_same() slow-path if * incoming WRITE_SAME payload does not contain zeros. */ - rc = memcmp(buf, p, cmd->data_length); + not_zero = memchr_inv(buf, 0x00, cmd->data_length); kunmap(sg_page(sg)); - if (rc) + if (not_zero) return TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE; ret = blkdev_issue_zeroout(bdev, diff --git a/drivers/thermal/int340x_thermal/int3403_thermal.c b/drivers/thermal/int340x_thermal/int3403_thermal.c index 8a7f24dd9315..0c19fcd56a0d 100644 --- a/drivers/thermal/int340x_thermal/int3403_thermal.c +++ b/drivers/thermal/int340x_thermal/int3403_thermal.c @@ -194,6 +194,7 @@ static int int3403_cdev_add(struct int3403_priv *priv) return -EFAULT; } + priv->priv = obj; obj->max_state = p->package.count - 1; obj->cdev = thermal_cooling_device_register(acpi_device_bid(priv->adev), @@ -201,8 +202,6 @@ static int int3403_cdev_add(struct int3403_priv *priv) if (IS_ERR(obj->cdev)) result = PTR_ERR(obj->cdev); - priv->priv = obj; - kfree(buf.pointer); /* TODO: add ACPI notification support */ diff --git a/drivers/thermal/samsung/exynos_tmu.c b/drivers/thermal/samsung/exynos_tmu.c index ed805c7c5ace..ac83f721db24 100644 --- a/drivers/thermal/samsung/exynos_tmu.c +++ b/drivers/thermal/samsung/exynos_tmu.c @@ -185,6 +185,7 @@ * @regulator: pointer to the TMU regulator structure. * @reg_conf: pointer to structure to register with core thermal. * @ntrip: number of supported trip points. + * @enabled: current status of TMU device * @tmu_initialize: SoC specific TMU initialization method * @tmu_control: SoC specific TMU control method * @tmu_read: SoC specific TMU temperature read method @@ -205,6 +206,7 @@ struct exynos_tmu_data { struct regulator *regulator; struct thermal_zone_device *tzd; unsigned int ntrip; + bool enabled; int (*tmu_initialize)(struct platform_device *pdev); void (*tmu_control)(struct platform_device *pdev, bool on); @@ -398,6 +400,7 @@ static void exynos_tmu_control(struct platform_device *pdev, bool on) mutex_lock(&data->lock); clk_enable(data->clk); data->tmu_control(pdev, on); + data->enabled = on; clk_disable(data->clk); mutex_unlock(&data->lock); } @@ -889,19 +892,24 @@ static void exynos7_tmu_control(struct platform_device *pdev, bool on) static int exynos_get_temp(void *p, int *temp) { struct exynos_tmu_data *data = p; + int value, ret = 0; - if (!data || !data->tmu_read) + if (!data || !data->tmu_read || !data->enabled) return -EINVAL; mutex_lock(&data->lock); clk_enable(data->clk); - *temp = code_to_temp(data, data->tmu_read(data)) * MCELSIUS; + value = data->tmu_read(data); + if (value < 0) + ret = value; + else + *temp = code_to_temp(data, value) * MCELSIUS; clk_disable(data->clk); mutex_unlock(&data->lock); - return 0; + return ret; } #ifdef CONFIG_THERMAL_EMULATION diff --git a/drivers/usb/core/config.c b/drivers/usb/core/config.c index c821b4b9647e..7b5cb28ffb35 100644 --- a/drivers/usb/core/config.c +++ b/drivers/usb/core/config.c @@ -191,7 +191,9 @@ static const unsigned short full_speed_maxpacket_maxes[4] = { static const unsigned short high_speed_maxpacket_maxes[4] = { [USB_ENDPOINT_XFER_CONTROL] = 64, [USB_ENDPOINT_XFER_ISOC] = 1024, - [USB_ENDPOINT_XFER_BULK] = 512, + + /* Bulk should be 512, but some devices use 1024: we will warn below */ + [USB_ENDPOINT_XFER_BULK] = 1024, [USB_ENDPOINT_XFER_INT] = 1024, }; static const unsigned short super_speed_maxpacket_maxes[4] = { diff --git a/drivers/usb/dwc2/core.h b/drivers/usb/dwc2/core.h index d83be5651f87..a666e0758a99 100644 --- a/drivers/usb/dwc2/core.h +++ b/drivers/usb/dwc2/core.h @@ -985,6 +985,7 @@ struct dwc2_hsotg { /* DWC OTG HW Release versions */ #define DWC2_CORE_REV_2_71a 0x4f54271a +#define DWC2_CORE_REV_2_72a 0x4f54272a #define DWC2_CORE_REV_2_80a 0x4f54280a #define DWC2_CORE_REV_2_90a 0x4f54290a #define DWC2_CORE_REV_2_91a 0x4f54291a @@ -992,6 +993,7 @@ struct dwc2_hsotg { #define DWC2_CORE_REV_2_94a 0x4f54294a #define DWC2_CORE_REV_3_00a 0x4f54300a #define DWC2_CORE_REV_3_10a 0x4f54310a +#define DWC2_CORE_REV_4_00a 0x4f54400a #define DWC2_FS_IOT_REV_1_00a 0x5531100a #define DWC2_HS_IOT_REV_1_00a 0x5532100a diff --git a/drivers/usb/dwc2/gadget.c b/drivers/usb/dwc2/gadget.c index 6c32bf26e48e..83cb5577a52f 100644 --- a/drivers/usb/dwc2/gadget.c +++ b/drivers/usb/dwc2/gadget.c @@ -3928,6 +3928,27 @@ static int dwc2_hsotg_ep_enable(struct usb_ep *ep, if (index && !hs_ep->isochronous) epctrl |= DXEPCTL_SETD0PID; + /* WA for Full speed ISOC IN in DDMA mode. + * By Clear NAK status of EP, core will send ZLP + * to IN token and assert NAK interrupt relying + * on TxFIFO status only + */ + + if (hsotg->gadget.speed == USB_SPEED_FULL && + hs_ep->isochronous && dir_in) { + /* The WA applies only to core versions from 2.72a + * to 4.00a (including both). Also for FS_IOT_1.00a + * and HS_IOT_1.00a. + */ + u32 gsnpsid = dwc2_readl(hsotg->regs + GSNPSID); + + if ((gsnpsid >= DWC2_CORE_REV_2_72a && + gsnpsid <= DWC2_CORE_REV_4_00a) || + gsnpsid == DWC2_FS_IOT_REV_1_00a || + gsnpsid == DWC2_HS_IOT_REV_1_00a) + epctrl |= DXEPCTL_CNAK; + } + dev_dbg(hsotg->dev, "%s: write DxEPCTL=0x%08x\n", __func__, epctrl); diff --git a/drivers/usb/dwc2/hcd.c b/drivers/usb/dwc2/hcd.c index 190f95964000..c51b73b3e048 100644 --- a/drivers/usb/dwc2/hcd.c +++ b/drivers/usb/dwc2/hcd.c @@ -358,9 +358,14 @@ static void dwc2_gusbcfg_init(struct dwc2_hsotg *hsotg) static int dwc2_vbus_supply_init(struct dwc2_hsotg *hsotg) { + int ret; + hsotg->vbus_supply = devm_regulator_get_optional(hsotg->dev, "vbus"); - if (IS_ERR(hsotg->vbus_supply)) - return 0; + if (IS_ERR(hsotg->vbus_supply)) { + ret = PTR_ERR(hsotg->vbus_supply); + hsotg->vbus_supply = NULL; + return ret == -ENODEV ? 0 : ret; + } return regulator_enable(hsotg->vbus_supply); } @@ -4342,9 +4347,7 @@ static int _dwc2_hcd_start(struct usb_hcd *hcd) spin_unlock_irqrestore(&hsotg->lock, flags); - dwc2_vbus_supply_init(hsotg); - - return 0; + return dwc2_vbus_supply_init(hsotg); } /* diff --git a/drivers/usb/dwc2/pci.c b/drivers/usb/dwc2/pci.c index 7f21747007f1..bea2e8ec0369 100644 --- a/drivers/usb/dwc2/pci.c +++ b/drivers/usb/dwc2/pci.c @@ -141,8 +141,10 @@ static int dwc2_pci_probe(struct pci_dev *pci, goto err; glue = devm_kzalloc(dev, sizeof(*glue), GFP_KERNEL); - if (!glue) + if (!glue) { + ret = -ENOMEM; goto err; + } ret = platform_device_add(dwc2); if (ret) { diff --git a/drivers/usb/dwc3/gadget.c b/drivers/usb/dwc3/gadget.c index 8796a5ee9bb9..0dedf8a799f4 100644 --- a/drivers/usb/dwc3/gadget.c +++ b/drivers/usb/dwc3/gadget.c @@ -166,7 +166,7 @@ static void dwc3_ep_inc_deq(struct dwc3_ep *dep) dwc3_ep_inc_trb(&dep->trb_dequeue); } -void dwc3_gadget_del_and_unmap_request(struct dwc3_ep *dep, +static void dwc3_gadget_del_and_unmap_request(struct dwc3_ep *dep, struct dwc3_request *req, int status) { struct dwc3 *dwc = dep->dwc; @@ -1424,7 +1424,7 @@ static int dwc3_gadget_ep_dequeue(struct usb_ep *ep, dwc->lock); if (!r->trb) - goto out1; + goto out0; if (r->num_pending_sgs) { struct dwc3_trb *trb; diff --git a/drivers/usb/gadget/function/f_phonet.c b/drivers/usb/gadget/function/f_phonet.c index 7889bcc0509a..8b72b192c747 100644 --- a/drivers/usb/gadget/function/f_phonet.c +++ b/drivers/usb/gadget/function/f_phonet.c @@ -221,7 +221,7 @@ static void pn_tx_complete(struct usb_ep *ep, struct usb_request *req) netif_wake_queue(dev); } -static int pn_net_xmit(struct sk_buff *skb, struct net_device *dev) +static netdev_tx_t pn_net_xmit(struct sk_buff *skb, struct net_device *dev) { struct phonet_port *port = netdev_priv(dev); struct f_phonet *fp; diff --git a/drivers/usb/host/ehci-mem.c b/drivers/usb/host/ehci-mem.c index 4c6c08b675b5..21307d862af6 100644 --- a/drivers/usb/host/ehci-mem.c +++ b/drivers/usb/host/ehci-mem.c @@ -73,9 +73,10 @@ static struct ehci_qh *ehci_qh_alloc (struct ehci_hcd *ehci, gfp_t flags) if (!qh) goto done; qh->hw = (struct ehci_qh_hw *) - dma_pool_zalloc(ehci->qh_pool, flags, &dma); + dma_pool_alloc(ehci->qh_pool, flags, &dma); if (!qh->hw) goto fail; + memset(qh->hw, 0, sizeof *qh->hw); qh->qh_dma = dma; // INIT_LIST_HEAD (&qh->qh_list); INIT_LIST_HEAD (&qh->qtd_list); diff --git a/drivers/usb/host/ehci-sched.c b/drivers/usb/host/ehci-sched.c index 28e2a338b481..e56db44708bc 100644 --- a/drivers/usb/host/ehci-sched.c +++ b/drivers/usb/host/ehci-sched.c @@ -1287,7 +1287,7 @@ itd_urb_transaction( } else { alloc_itd: spin_unlock_irqrestore(&ehci->lock, flags); - itd = dma_pool_zalloc(ehci->itd_pool, mem_flags, + itd = dma_pool_alloc(ehci->itd_pool, mem_flags, &itd_dma); spin_lock_irqsave(&ehci->lock, flags); if (!itd) { @@ -1297,6 +1297,7 @@ itd_urb_transaction( } } + memset(itd, 0, sizeof(*itd)); itd->itd_dma = itd_dma; itd->frame = NO_FRAME; list_add(&itd->itd_list, &sched->td_list); @@ -2080,7 +2081,7 @@ sitd_urb_transaction( } else { alloc_sitd: spin_unlock_irqrestore(&ehci->lock, flags); - sitd = dma_pool_zalloc(ehci->sitd_pool, mem_flags, + sitd = dma_pool_alloc(ehci->sitd_pool, mem_flags, &sitd_dma); spin_lock_irqsave(&ehci->lock, flags); if (!sitd) { @@ -2090,6 +2091,7 @@ sitd_urb_transaction( } } + memset(sitd, 0, sizeof(*sitd)); sitd->sitd_dma = sitd_dma; sitd->frame = NO_FRAME; list_add(&sitd->sitd_list, &iso_sched->td_list); diff --git a/drivers/usb/host/xhci.c b/drivers/usb/host/xhci.c index 9b27798ecce5..711da3306b14 100644 --- a/drivers/usb/host/xhci.c +++ b/drivers/usb/host/xhci.c @@ -3621,6 +3621,7 @@ static void xhci_free_dev(struct usb_hcd *hcd, struct usb_device *udev) del_timer_sync(&virt_dev->eps[i].stop_cmd_timer); } xhci_debugfs_remove_slot(xhci, udev->slot_id); + virt_dev->udev = NULL; ret = xhci_disable_slot(xhci, udev->slot_id); if (ret) xhci_free_virt_device(xhci, udev->slot_id); diff --git a/drivers/usb/musb/musb_gadget.c b/drivers/usb/musb/musb_gadget.c index e564695c6c8d..71c5835ea9cd 100644 --- a/drivers/usb/musb/musb_gadget.c +++ b/drivers/usb/musb/musb_gadget.c @@ -417,7 +417,6 @@ void musb_g_tx(struct musb *musb, u8 epnum) req = next_request(musb_ep); request = &req->request; - trace_musb_req_tx(req); csr = musb_readw(epio, MUSB_TXCSR); musb_dbg(musb, "<== %s, txcsr %04x", musb_ep->end_point.name, csr); @@ -456,6 +455,8 @@ void musb_g_tx(struct musb *musb, u8 epnum) u8 is_dma = 0; bool short_packet = false; + trace_musb_req_tx(req); + if (dma && (csr & MUSB_TXCSR_DMAENAB)) { is_dma = 1; csr |= MUSB_TXCSR_P_WZC_BITS; diff --git a/drivers/usb/musb/musb_host.c b/drivers/usb/musb/musb_host.c index 4fa372c845e1..e7f99d55922a 100644 --- a/drivers/usb/musb/musb_host.c +++ b/drivers/usb/musb/musb_host.c @@ -990,7 +990,9 @@ static void musb_bulk_nak_timeout(struct musb *musb, struct musb_hw_ep *ep, /* set tx_reinit and schedule the next qh */ ep->tx_reinit = 1; } - musb_start_urb(musb, is_in, next_qh); + + if (next_qh) + musb_start_urb(musb, is_in, next_qh); } } diff --git a/drivers/usb/serial/option.c b/drivers/usb/serial/option.c index c3f252283ab9..2058852a87fa 100644 --- a/drivers/usb/serial/option.c +++ b/drivers/usb/serial/option.c @@ -233,6 +233,8 @@ static void option_instat_callback(struct urb *urb); /* These Quectel products use Qualcomm's vendor ID */ #define QUECTEL_PRODUCT_UC20 0x9003 #define QUECTEL_PRODUCT_UC15 0x9090 +/* These u-blox products use Qualcomm's vendor ID */ +#define UBLOX_PRODUCT_R410M 0x90b2 /* These Yuga products use Qualcomm's vendor ID */ #define YUGA_PRODUCT_CLM920_NC5 0x9625 @@ -1065,6 +1067,9 @@ static const struct usb_device_id option_ids[] = { /* Yuga products use Qualcomm vendor ID */ { USB_DEVICE(QUALCOMM_VENDOR_ID, YUGA_PRODUCT_CLM920_NC5), .driver_info = RSVD(1) | RSVD(4) }, + /* u-blox products using Qualcomm vendor ID */ + { USB_DEVICE(QUALCOMM_VENDOR_ID, UBLOX_PRODUCT_R410M), + .driver_info = RSVD(1) | RSVD(3) }, /* Quectel products using Quectel vendor ID */ { USB_DEVICE(QUECTEL_VENDOR_ID, QUECTEL_PRODUCT_EC21), .driver_info = RSVD(4) }, diff --git a/drivers/usb/serial/visor.c b/drivers/usb/serial/visor.c index f5373ed2cd45..8ddbecc25d89 100644 --- a/drivers/usb/serial/visor.c +++ b/drivers/usb/serial/visor.c @@ -335,47 +335,48 @@ static int palm_os_3_probe(struct usb_serial *serial, goto exit; } - if (retval == sizeof(*connection_info)) { - connection_info = (struct visor_connection_info *) - transfer_buffer; - - num_ports = le16_to_cpu(connection_info->num_ports); - for (i = 0; i < num_ports; ++i) { - switch ( - connection_info->connections[i].port_function_id) { - case VISOR_FUNCTION_GENERIC: - string = "Generic"; - break; - case VISOR_FUNCTION_DEBUGGER: - string = "Debugger"; - break; - case VISOR_FUNCTION_HOTSYNC: - string = "HotSync"; - break; - case VISOR_FUNCTION_CONSOLE: - string = "Console"; - break; - case VISOR_FUNCTION_REMOTE_FILE_SYS: - string = "Remote File System"; - break; - default: - string = "unknown"; - break; - } - dev_info(dev, "%s: port %d, is for %s use\n", - serial->type->description, - connection_info->connections[i].port, string); - } + if (retval != sizeof(*connection_info)) { + dev_err(dev, "Invalid connection information received from device\n"); + retval = -ENODEV; + goto exit; } - /* - * Handle devices that report invalid stuff here. - */ + + connection_info = (struct visor_connection_info *)transfer_buffer; + + num_ports = le16_to_cpu(connection_info->num_ports); + + /* Handle devices that report invalid stuff here. */ if (num_ports == 0 || num_ports > 2) { dev_warn(dev, "%s: No valid connect info available\n", serial->type->description); num_ports = 2; } + for (i = 0; i < num_ports; ++i) { + switch (connection_info->connections[i].port_function_id) { + case VISOR_FUNCTION_GENERIC: + string = "Generic"; + break; + case VISOR_FUNCTION_DEBUGGER: + string = "Debugger"; + break; + case VISOR_FUNCTION_HOTSYNC: + string = "HotSync"; + break; + case VISOR_FUNCTION_CONSOLE: + string = "Console"; + break; + case VISOR_FUNCTION_REMOTE_FILE_SYS: + string = "Remote File System"; + break; + default: + string = "unknown"; + break; + } + dev_info(dev, "%s: port %d, is for %s use\n", + serial->type->description, + connection_info->connections[i].port, string); + } dev_info(dev, "%s: Number of ports: %d\n", serial->type->description, num_ports); diff --git a/drivers/usb/typec/tcpm.c b/drivers/usb/typec/tcpm.c index 677d12138dbd..ded49e3bf2b0 100644 --- a/drivers/usb/typec/tcpm.c +++ b/drivers/usb/typec/tcpm.c @@ -3725,6 +3725,7 @@ void tcpm_unregister_port(struct tcpm_port *port) for (i = 0; i < ARRAY_SIZE(port->port_altmode); i++) typec_unregister_altmode(port->port_altmode[i]); typec_unregister_port(port->typec_port); + usb_role_switch_put(port->role_sw); tcpm_debugfs_exit(port); destroy_workqueue(port->wq); } diff --git a/drivers/usb/typec/tps6598x.c b/drivers/usb/typec/tps6598x.c index 8b8406867c02..4b4c8d271b27 100644 --- a/drivers/usb/typec/tps6598x.c +++ b/drivers/usb/typec/tps6598x.c @@ -73,6 +73,7 @@ struct tps6598x { struct device *dev; struct regmap *regmap; struct mutex lock; /* device lock */ + u8 i2c_protocol:1; struct typec_port *port; struct typec_partner *partner; @@ -80,19 +81,39 @@ struct tps6598x { struct typec_capability typec_cap; }; +static int +tps6598x_block_read(struct tps6598x *tps, u8 reg, void *val, size_t len) +{ + u8 data[len + 1]; + int ret; + + if (!tps->i2c_protocol) + return regmap_raw_read(tps->regmap, reg, val, len); + + ret = regmap_raw_read(tps->regmap, reg, data, sizeof(data)); + if (ret) + return ret; + + if (data[0] < len) + return -EIO; + + memcpy(val, &data[1], len); + return 0; +} + static inline int tps6598x_read16(struct tps6598x *tps, u8 reg, u16 *val) { - return regmap_raw_read(tps->regmap, reg, val, sizeof(u16)); + return tps6598x_block_read(tps, reg, val, sizeof(u16)); } static inline int tps6598x_read32(struct tps6598x *tps, u8 reg, u32 *val) { - return regmap_raw_read(tps->regmap, reg, val, sizeof(u32)); + return tps6598x_block_read(tps, reg, val, sizeof(u32)); } static inline int tps6598x_read64(struct tps6598x *tps, u8 reg, u64 *val) { - return regmap_raw_read(tps->regmap, reg, val, sizeof(u64)); + return tps6598x_block_read(tps, reg, val, sizeof(u64)); } static inline int tps6598x_write16(struct tps6598x *tps, u8 reg, u16 val) @@ -121,8 +142,8 @@ static int tps6598x_read_partner_identity(struct tps6598x *tps) struct tps6598x_rx_identity_reg id; int ret; - ret = regmap_raw_read(tps->regmap, TPS_REG_RX_IDENTITY_SOP, - &id, sizeof(id)); + ret = tps6598x_block_read(tps, TPS_REG_RX_IDENTITY_SOP, + &id, sizeof(id)); if (ret) return ret; @@ -224,13 +245,13 @@ static int tps6598x_exec_cmd(struct tps6598x *tps, const char *cmd, } while (val); if (out_len) { - ret = regmap_raw_read(tps->regmap, TPS_REG_DATA1, - out_data, out_len); + ret = tps6598x_block_read(tps, TPS_REG_DATA1, + out_data, out_len); if (ret) return ret; val = out_data[0]; } else { - ret = regmap_read(tps->regmap, TPS_REG_DATA1, &val); + ret = tps6598x_block_read(tps, TPS_REG_DATA1, &val, sizeof(u8)); if (ret) return ret; } @@ -385,6 +406,16 @@ static int tps6598x_probe(struct i2c_client *client) if (!vid) return -ENODEV; + /* + * Checking can the adapter handle SMBus protocol. If it can not, the + * driver needs to take care of block reads separately. + * + * FIXME: Testing with I2C_FUNC_I2C. regmap-i2c uses I2C protocol + * unconditionally if the adapter has I2C_FUNC_I2C set. + */ + if (i2c_check_functionality(client->adapter, I2C_FUNC_I2C)) + tps->i2c_protocol = true; + ret = tps6598x_read32(tps, TPS_REG_STATUS, &status); if (ret < 0) return ret; diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index e2f16b68fcbf..51b5e2da708c 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -3142,7 +3142,11 @@ static noinline int check_delayed_ref(struct btrfs_root *root, struct rb_node *node; int ret = 0; + spin_lock(&root->fs_info->trans_lock); cur_trans = root->fs_info->running_transaction; + if (cur_trans) + refcount_inc(&cur_trans->use_count); + spin_unlock(&root->fs_info->trans_lock); if (!cur_trans) return 0; @@ -3151,6 +3155,7 @@ static noinline int check_delayed_ref(struct btrfs_root *root, head = btrfs_find_delayed_ref_head(delayed_refs, bytenr); if (!head) { spin_unlock(&delayed_refs->lock); + btrfs_put_transaction(cur_trans); return 0; } @@ -3167,6 +3172,7 @@ static noinline int check_delayed_ref(struct btrfs_root *root, mutex_lock(&head->mutex); mutex_unlock(&head->mutex); btrfs_put_delayed_ref_head(head); + btrfs_put_transaction(cur_trans); return -EAGAIN; } spin_unlock(&delayed_refs->lock); @@ -3199,6 +3205,7 @@ static noinline int check_delayed_ref(struct btrfs_root *root, } spin_unlock(&head->lock); mutex_unlock(&head->mutex); + btrfs_put_transaction(cur_trans); return ret; } diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c index 00b7d3231821..b041b945a7ae 100644 --- a/fs/btrfs/relocation.c +++ b/fs/btrfs/relocation.c @@ -1841,7 +1841,7 @@ again: old_bytenr = btrfs_node_blockptr(parent, slot); blocksize = fs_info->nodesize; old_ptr_gen = btrfs_node_ptr_generation(parent, slot); - btrfs_node_key_to_cpu(parent, &key, slot); + btrfs_node_key_to_cpu(parent, &first_key, slot); if (level <= max_level) { eb = path->nodes[level]; diff --git a/fs/btrfs/send.c b/fs/btrfs/send.c index 221e5cdb060b..c0074d2d7d6d 100644 --- a/fs/btrfs/send.c +++ b/fs/btrfs/send.c @@ -5236,6 +5236,10 @@ static int send_write_or_clone(struct send_ctx *sctx, len = btrfs_file_extent_num_bytes(path->nodes[0], ei); } + if (offset >= sctx->cur_inode_size) { + ret = 0; + goto out; + } if (offset + len > sctx->cur_inode_size) len = sctx->cur_inode_size - offset; if (len == 0) { diff --git a/fs/ceph/file.c b/fs/ceph/file.c index f85040d73e3d..cf0e45b10121 100644 --- a/fs/ceph/file.c +++ b/fs/ceph/file.c @@ -70,69 +70,104 @@ static __le32 ceph_flags_sys2wire(u32 flags) */ /* - * Calculate the length sum of direct io vectors that can - * be combined into one page vector. + * How many pages to get in one call to iov_iter_get_pages(). This + * determines the size of the on-stack array used as a buffer. */ -static size_t dio_get_pagev_size(const struct iov_iter *it) +#define ITER_GET_BVECS_PAGES 64 + +static ssize_t __iter_get_bvecs(struct iov_iter *iter, size_t maxsize, + struct bio_vec *bvecs) { - const struct iovec *iov = it->iov; - const struct iovec *iovend = iov + it->nr_segs; - size_t size; - - size = iov->iov_len - it->iov_offset; - /* - * An iov can be page vectored when both the current tail - * and the next base are page aligned. - */ - while (PAGE_ALIGNED((iov->iov_base + iov->iov_len)) && - (++iov < iovend && PAGE_ALIGNED((iov->iov_base)))) { - size += iov->iov_len; - } - dout("dio_get_pagevlen len = %zu\n", size); - return size; + size_t size = 0; + int bvec_idx = 0; + + if (maxsize > iov_iter_count(iter)) + maxsize = iov_iter_count(iter); + + while (size < maxsize) { + struct page *pages[ITER_GET_BVECS_PAGES]; + ssize_t bytes; + size_t start; + int idx = 0; + + bytes = iov_iter_get_pages(iter, pages, maxsize - size, + ITER_GET_BVECS_PAGES, &start); + if (bytes < 0) + return size ?: bytes; + + iov_iter_advance(iter, bytes); + size += bytes; + + for ( ; bytes; idx++, bvec_idx++) { + struct bio_vec bv = { + .bv_page = pages[idx], + .bv_len = min_t(int, bytes, PAGE_SIZE - start), + .bv_offset = start, + }; + + bvecs[bvec_idx] = bv; + bytes -= bv.bv_len; + start = 0; + } + } + + return size; } /* - * Allocate a page vector based on (@it, @nbytes). - * The return value is the tuple describing a page vector, - * that is (@pages, @page_align, @num_pages). + * iov_iter_get_pages() only considers one iov_iter segment, no matter + * what maxsize or maxpages are given. For ITER_BVEC that is a single + * page. + * + * Attempt to get up to @maxsize bytes worth of pages from @iter. + * Return the number of bytes in the created bio_vec array, or an error. */ -static struct page ** -dio_get_pages_alloc(const struct iov_iter *it, size_t nbytes, - size_t *page_align, int *num_pages) +static ssize_t iter_get_bvecs_alloc(struct iov_iter *iter, size_t maxsize, + struct bio_vec **bvecs, int *num_bvecs) { - struct iov_iter tmp_it = *it; - size_t align; - struct page **pages; - int ret = 0, idx, npages; + struct bio_vec *bv; + size_t orig_count = iov_iter_count(iter); + ssize_t bytes; + int npages; - align = (unsigned long)(it->iov->iov_base + it->iov_offset) & - (PAGE_SIZE - 1); - npages = calc_pages_for(align, nbytes); - pages = kvmalloc(sizeof(*pages) * npages, GFP_KERNEL); - if (!pages) - return ERR_PTR(-ENOMEM); + iov_iter_truncate(iter, maxsize); + npages = iov_iter_npages(iter, INT_MAX); + iov_iter_reexpand(iter, orig_count); - for (idx = 0; idx < npages; ) { - size_t start; - ret = iov_iter_get_pages(&tmp_it, pages + idx, nbytes, - npages - idx, &start); - if (ret < 0) - goto fail; + /* + * __iter_get_bvecs() may populate only part of the array -- zero it + * out. + */ + bv = kvmalloc_array(npages, sizeof(*bv), GFP_KERNEL | __GFP_ZERO); + if (!bv) + return -ENOMEM; - iov_iter_advance(&tmp_it, ret); - nbytes -= ret; - idx += (ret + start + PAGE_SIZE - 1) / PAGE_SIZE; + bytes = __iter_get_bvecs(iter, maxsize, bv); + if (bytes < 0) { + /* + * No pages were pinned -- just free the array. + */ + kvfree(bv); + return bytes; } - BUG_ON(nbytes != 0); - *num_pages = npages; - *page_align = align; - dout("dio_get_pages_alloc: got %d pages align %zu\n", npages, align); - return pages; -fail: - ceph_put_page_vector(pages, idx, false); - return ERR_PTR(ret); + *bvecs = bv; + *num_bvecs = npages; + return bytes; +} + +static void put_bvecs(struct bio_vec *bvecs, int num_bvecs, bool should_dirty) +{ + int i; + + for (i = 0; i < num_bvecs; i++) { + if (bvecs[i].bv_page) { + if (should_dirty) + set_page_dirty_lock(bvecs[i].bv_page); + put_page(bvecs[i].bv_page); + } + } + kvfree(bvecs); } /* @@ -746,11 +781,12 @@ static void ceph_aio_complete_req(struct ceph_osd_request *req) struct inode *inode = req->r_inode; struct ceph_aio_request *aio_req = req->r_priv; struct ceph_osd_data *osd_data = osd_req_op_extent_osd_data(req, 0); - int num_pages = calc_pages_for((u64)osd_data->alignment, - osd_data->length); - dout("ceph_aio_complete_req %p rc %d bytes %llu\n", - inode, rc, osd_data->length); + BUG_ON(osd_data->type != CEPH_OSD_DATA_TYPE_BVECS); + BUG_ON(!osd_data->num_bvecs); + + dout("ceph_aio_complete_req %p rc %d bytes %u\n", + inode, rc, osd_data->bvec_pos.iter.bi_size); if (rc == -EOLDSNAPC) { struct ceph_aio_work *aio_work; @@ -768,9 +804,10 @@ static void ceph_aio_complete_req(struct ceph_osd_request *req) } else if (!aio_req->write) { if (rc == -ENOENT) rc = 0; - if (rc >= 0 && osd_data->length > rc) { - int zoff = osd_data->alignment + rc; - int zlen = osd_data->length - rc; + if (rc >= 0 && osd_data->bvec_pos.iter.bi_size > rc) { + struct iov_iter i; + int zlen = osd_data->bvec_pos.iter.bi_size - rc; + /* * If read is satisfied by single OSD request, * it can pass EOF. Otherwise read is within @@ -785,13 +822,16 @@ static void ceph_aio_complete_req(struct ceph_osd_request *req) aio_req->total_len = rc + zlen; } - if (zlen > 0) - ceph_zero_page_vector_range(zoff, zlen, - osd_data->pages); + iov_iter_bvec(&i, ITER_BVEC, osd_data->bvec_pos.bvecs, + osd_data->num_bvecs, + osd_data->bvec_pos.iter.bi_size); + iov_iter_advance(&i, rc); + iov_iter_zero(zlen, &i); } } - ceph_put_page_vector(osd_data->pages, num_pages, aio_req->should_dirty); + put_bvecs(osd_data->bvec_pos.bvecs, osd_data->num_bvecs, + aio_req->should_dirty); ceph_osdc_put_request(req); if (rc < 0) @@ -879,7 +919,7 @@ ceph_direct_read_write(struct kiocb *iocb, struct iov_iter *iter, struct ceph_fs_client *fsc = ceph_inode_to_client(inode); struct ceph_vino vino; struct ceph_osd_request *req; - struct page **pages; + struct bio_vec *bvecs; struct ceph_aio_request *aio_req = NULL; int num_pages = 0; int flags; @@ -914,10 +954,14 @@ ceph_direct_read_write(struct kiocb *iocb, struct iov_iter *iter, } while (iov_iter_count(iter) > 0) { - u64 size = dio_get_pagev_size(iter); - size_t start = 0; + u64 size = iov_iter_count(iter); ssize_t len; + if (write) + size = min_t(u64, size, fsc->mount_options->wsize); + else + size = min_t(u64, size, fsc->mount_options->rsize); + vino = ceph_vino(inode); req = ceph_osdc_new_request(&fsc->client->osdc, &ci->i_layout, vino, pos, &size, 0, @@ -933,18 +977,14 @@ ceph_direct_read_write(struct kiocb *iocb, struct iov_iter *iter, break; } - if (write) - size = min_t(u64, size, fsc->mount_options->wsize); - else - size = min_t(u64, size, fsc->mount_options->rsize); - - len = size; - pages = dio_get_pages_alloc(iter, len, &start, &num_pages); - if (IS_ERR(pages)) { + len = iter_get_bvecs_alloc(iter, size, &bvecs, &num_pages); + if (len < 0) { ceph_osdc_put_request(req); - ret = PTR_ERR(pages); + ret = len; break; } + if (len != size) + osd_req_op_extent_update(req, 0, len); /* * To simplify error handling, allow AIO when IO within i_size @@ -977,8 +1017,7 @@ ceph_direct_read_write(struct kiocb *iocb, struct iov_iter *iter, req->r_mtime = mtime; } - osd_req_op_extent_osd_data_pages(req, 0, pages, len, start, - false, false); + osd_req_op_extent_osd_data_bvecs(req, 0, bvecs, num_pages, len); if (aio_req) { aio_req->total_len += len; @@ -991,7 +1030,6 @@ ceph_direct_read_write(struct kiocb *iocb, struct iov_iter *iter, list_add_tail(&req->r_unsafe_item, &aio_req->osd_reqs); pos += len; - iov_iter_advance(iter, len); continue; } @@ -1004,25 +1042,26 @@ ceph_direct_read_write(struct kiocb *iocb, struct iov_iter *iter, if (ret == -ENOENT) ret = 0; if (ret >= 0 && ret < len && pos + ret < size) { + struct iov_iter i; int zlen = min_t(size_t, len - ret, size - pos - ret); - ceph_zero_page_vector_range(start + ret, zlen, - pages); + + iov_iter_bvec(&i, ITER_BVEC, bvecs, num_pages, + len); + iov_iter_advance(&i, ret); + iov_iter_zero(zlen, &i); ret += zlen; } if (ret >= 0) len = ret; } - ceph_put_page_vector(pages, num_pages, should_dirty); - + put_bvecs(bvecs, num_pages, should_dirty); ceph_osdc_put_request(req); if (ret < 0) break; pos += len; - iov_iter_advance(iter, len); - if (!write && pos >= size) break; diff --git a/fs/cifs/Kconfig b/fs/cifs/Kconfig index 741749a98614..5f132d59dfc2 100644 --- a/fs/cifs/Kconfig +++ b/fs/cifs/Kconfig @@ -197,7 +197,7 @@ config CIFS_SMB311 config CIFS_SMB_DIRECT bool "SMB Direct support (Experimental)" - depends on CIFS=m && INFINIBAND || CIFS=y && INFINIBAND=y + depends on CIFS=m && INFINIBAND && INFINIBAND_ADDR_TRANS || CIFS=y && INFINIBAND=y && INFINIBAND_ADDR_TRANS=y help Enables SMB Direct experimental support for SMB 3.0, 3.02 and 3.1.1. SMB Direct allows transferring SMB packets over RDMA. If unsure, diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index f715609b13f3..5a5a0158cc8f 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -1047,6 +1047,18 @@ out: return rc; } +/* + * Directory operations under CIFS/SMB2/SMB3 are synchronous, so fsync() + * is a dummy operation. + */ +static int cifs_dir_fsync(struct file *file, loff_t start, loff_t end, int datasync) +{ + cifs_dbg(FYI, "Sync directory - name: %pD datasync: 0x%x\n", + file, datasync); + + return 0; +} + static ssize_t cifs_copy_file_range(struct file *src_file, loff_t off, struct file *dst_file, loff_t destoff, size_t len, unsigned int flags) @@ -1181,6 +1193,7 @@ const struct file_operations cifs_dir_ops = { .copy_file_range = cifs_copy_file_range, .clone_file_range = cifs_clone_file_range, .llseek = generic_file_llseek, + .fsync = cifs_dir_fsync, }; static void diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index a5aa158d535a..7a10a5d0731f 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -1977,14 +1977,6 @@ cifs_parse_mount_options(const char *mountdata, const char *devname, goto cifs_parse_mount_err; } -#ifdef CONFIG_CIFS_SMB_DIRECT - if (vol->rdma && vol->sign) { - cifs_dbg(VFS, "Currently SMB direct doesn't support signing." - " This is being fixed\n"); - goto cifs_parse_mount_err; - } -#endif - #ifndef CONFIG_KEYS /* Muliuser mounts require CONFIG_KEYS support */ if (vol->multiuser) { diff --git a/fs/cifs/smb2ops.c b/fs/cifs/smb2ops.c index b76b85881dcc..9c6d95ffca97 100644 --- a/fs/cifs/smb2ops.c +++ b/fs/cifs/smb2ops.c @@ -589,9 +589,15 @@ smb2_query_eas(const unsigned int xid, struct cifs_tcon *tcon, SMB2_close(xid, tcon, fid.persistent_fid, fid.volatile_fid); + /* + * If ea_name is NULL (listxattr) and there are no EAs, return 0 as it's + * not an error. Otherwise, the specified ea_name was not found. + */ if (!rc) rc = move_smb2_ea_to_cifs(ea_data, buf_size, smb2_data, SMB2_MAX_EA_BUF, ea_name); + else if (!ea_name && rc == -ENODATA) + rc = 0; kfree(smb2_data); return rc; diff --git a/fs/cifs/smb2pdu.c b/fs/cifs/smb2pdu.c index 60db51bae0e3..0f48741a0130 100644 --- a/fs/cifs/smb2pdu.c +++ b/fs/cifs/smb2pdu.c @@ -730,19 +730,14 @@ neg_exit: int smb3_validate_negotiate(const unsigned int xid, struct cifs_tcon *tcon) { - int rc = 0; - struct validate_negotiate_info_req vneg_inbuf; + int rc; + struct validate_negotiate_info_req *pneg_inbuf; struct validate_negotiate_info_rsp *pneg_rsp = NULL; u32 rsplen; u32 inbuflen; /* max of 4 dialects */ cifs_dbg(FYI, "validate negotiate\n"); -#ifdef CONFIG_CIFS_SMB_DIRECT - if (tcon->ses->server->rdma) - return 0; -#endif - /* In SMB3.11 preauth integrity supersedes validate negotiate */ if (tcon->ses->server->dialect == SMB311_PROT_ID) return 0; @@ -765,63 +760,69 @@ int smb3_validate_negotiate(const unsigned int xid, struct cifs_tcon *tcon) if (tcon->ses->session_flags & SMB2_SESSION_FLAG_IS_NULL) cifs_dbg(VFS, "Unexpected null user (anonymous) auth flag sent by server\n"); - vneg_inbuf.Capabilities = + pneg_inbuf = kmalloc(sizeof(*pneg_inbuf), GFP_NOFS); + if (!pneg_inbuf) + return -ENOMEM; + + pneg_inbuf->Capabilities = cpu_to_le32(tcon->ses->server->vals->req_capabilities); - memcpy(vneg_inbuf.Guid, tcon->ses->server->client_guid, + memcpy(pneg_inbuf->Guid, tcon->ses->server->client_guid, SMB2_CLIENT_GUID_SIZE); if (tcon->ses->sign) - vneg_inbuf.SecurityMode = + pneg_inbuf->SecurityMode = cpu_to_le16(SMB2_NEGOTIATE_SIGNING_REQUIRED); else if (global_secflags & CIFSSEC_MAY_SIGN) - vneg_inbuf.SecurityMode = + pneg_inbuf->SecurityMode = cpu_to_le16(SMB2_NEGOTIATE_SIGNING_ENABLED); else - vneg_inbuf.SecurityMode = 0; + pneg_inbuf->SecurityMode = 0; if (strcmp(tcon->ses->server->vals->version_string, SMB3ANY_VERSION_STRING) == 0) { - vneg_inbuf.Dialects[0] = cpu_to_le16(SMB30_PROT_ID); - vneg_inbuf.Dialects[1] = cpu_to_le16(SMB302_PROT_ID); - vneg_inbuf.DialectCount = cpu_to_le16(2); + pneg_inbuf->Dialects[0] = cpu_to_le16(SMB30_PROT_ID); + pneg_inbuf->Dialects[1] = cpu_to_le16(SMB302_PROT_ID); + pneg_inbuf->DialectCount = cpu_to_le16(2); /* structure is big enough for 3 dialects, sending only 2 */ - inbuflen = sizeof(struct validate_negotiate_info_req) - 2; + inbuflen = sizeof(*pneg_inbuf) - + sizeof(pneg_inbuf->Dialects[0]); } else if (strcmp(tcon->ses->server->vals->version_string, SMBDEFAULT_VERSION_STRING) == 0) { - vneg_inbuf.Dialects[0] = cpu_to_le16(SMB21_PROT_ID); - vneg_inbuf.Dialects[1] = cpu_to_le16(SMB30_PROT_ID); - vneg_inbuf.Dialects[2] = cpu_to_le16(SMB302_PROT_ID); - vneg_inbuf.DialectCount = cpu_to_le16(3); + pneg_inbuf->Dialects[0] = cpu_to_le16(SMB21_PROT_ID); + pneg_inbuf->Dialects[1] = cpu_to_le16(SMB30_PROT_ID); + pneg_inbuf->Dialects[2] = cpu_to_le16(SMB302_PROT_ID); + pneg_inbuf->DialectCount = cpu_to_le16(3); /* structure is big enough for 3 dialects */ - inbuflen = sizeof(struct validate_negotiate_info_req); + inbuflen = sizeof(*pneg_inbuf); } else { /* otherwise specific dialect was requested */ - vneg_inbuf.Dialects[0] = + pneg_inbuf->Dialects[0] = cpu_to_le16(tcon->ses->server->vals->protocol_id); - vneg_inbuf.DialectCount = cpu_to_le16(1); + pneg_inbuf->DialectCount = cpu_to_le16(1); /* structure is big enough for 3 dialects, sending only 1 */ - inbuflen = sizeof(struct validate_negotiate_info_req) - 4; + inbuflen = sizeof(*pneg_inbuf) - + sizeof(pneg_inbuf->Dialects[0]) * 2; } rc = SMB2_ioctl(xid, tcon, NO_FILE_ID, NO_FILE_ID, FSCTL_VALIDATE_NEGOTIATE_INFO, true /* is_fsctl */, - (char *)&vneg_inbuf, sizeof(struct validate_negotiate_info_req), - (char **)&pneg_rsp, &rsplen); + (char *)pneg_inbuf, inbuflen, (char **)&pneg_rsp, &rsplen); if (rc != 0) { cifs_dbg(VFS, "validate protocol negotiate failed: %d\n", rc); - return -EIO; + rc = -EIO; + goto out_free_inbuf; } - if (rsplen != sizeof(struct validate_negotiate_info_rsp)) { + rc = -EIO; + if (rsplen != sizeof(*pneg_rsp)) { cifs_dbg(VFS, "invalid protocol negotiate response size: %d\n", rsplen); /* relax check since Mac returns max bufsize allowed on ioctl */ - if ((rsplen > CIFSMaxBufSize) - || (rsplen < sizeof(struct validate_negotiate_info_rsp))) - goto err_rsp_free; + if (rsplen > CIFSMaxBufSize || rsplen < sizeof(*pneg_rsp)) + goto out_free_rsp; } /* check validate negotiate info response matches what we got earlier */ @@ -838,15 +839,17 @@ int smb3_validate_negotiate(const unsigned int xid, struct cifs_tcon *tcon) goto vneg_out; /* validate negotiate successful */ + rc = 0; cifs_dbg(FYI, "validate negotiate info successful\n"); - kfree(pneg_rsp); - return 0; + goto out_free_rsp; vneg_out: cifs_dbg(VFS, "protocol revalidation - security settings mismatch\n"); -err_rsp_free: +out_free_rsp: kfree(pneg_rsp); - return -EIO; +out_free_inbuf: + kfree(pneg_inbuf); + return rc; } enum securityEnum diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index 47d7c151fcba..471d863958bc 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -1961,7 +1961,7 @@ void wb_workfn(struct work_struct *work) } if (!list_empty(&wb->work_list)) - mod_delayed_work(bdi_wq, &wb->dwork, 0); + wb_wakeup(wb); else if (wb_has_dirty_io(wb) && dirty_writeback_interval) wb_wakeup_delayed(wb); diff --git a/fs/ocfs2/refcounttree.c b/fs/ocfs2/refcounttree.c index 01c6b3894406..7869622af22a 100644 --- a/fs/ocfs2/refcounttree.c +++ b/fs/ocfs2/refcounttree.c @@ -4250,10 +4250,11 @@ out: static int ocfs2_reflink(struct dentry *old_dentry, struct inode *dir, struct dentry *new_dentry, bool preserve) { - int error; + int error, had_lock; struct inode *inode = d_inode(old_dentry); struct buffer_head *old_bh = NULL; struct inode *new_orphan_inode = NULL; + struct ocfs2_lock_holder oh; if (!ocfs2_refcount_tree(OCFS2_SB(inode->i_sb))) return -EOPNOTSUPP; @@ -4295,6 +4296,14 @@ static int ocfs2_reflink(struct dentry *old_dentry, struct inode *dir, goto out; } + had_lock = ocfs2_inode_lock_tracker(new_orphan_inode, NULL, 1, + &oh); + if (had_lock < 0) { + error = had_lock; + mlog_errno(error); + goto out; + } + /* If the security isn't preserved, we need to re-initialize them. */ if (!preserve) { error = ocfs2_init_security_and_acl(dir, new_orphan_inode, @@ -4302,14 +4311,15 @@ static int ocfs2_reflink(struct dentry *old_dentry, struct inode *dir, if (error) mlog_errno(error); } -out: if (!error) { error = ocfs2_mv_orphaned_inode_to_new(dir, new_orphan_inode, new_dentry); if (error) mlog_errno(error); } + ocfs2_inode_unlock_tracker(new_orphan_inode, 1, &oh, had_lock); +out: if (new_orphan_inode) { /* * We need to open_unlock the inode no matter whether we diff --git a/fs/proc/kcore.c b/fs/proc/kcore.c index d1e82761de81..e64ecb9f2720 100644 --- a/fs/proc/kcore.c +++ b/fs/proc/kcore.c @@ -209,25 +209,34 @@ kclist_add_private(unsigned long pfn, unsigned long nr_pages, void *arg) { struct list_head *head = (struct list_head *)arg; struct kcore_list *ent; + struct page *p; + + if (!pfn_valid(pfn)) + return 1; + + p = pfn_to_page(pfn); + if (!memmap_valid_within(pfn, p, page_zone(p))) + return 1; ent = kmalloc(sizeof(*ent), GFP_KERNEL); if (!ent) return -ENOMEM; - ent->addr = (unsigned long)__va((pfn << PAGE_SHIFT)); + ent->addr = (unsigned long)page_to_virt(p); ent->size = nr_pages << PAGE_SHIFT; - /* Sanity check: Can happen in 32bit arch...maybe */ - if (ent->addr < (unsigned long) __va(0)) + if (!virt_addr_valid(ent->addr)) goto free_out; /* cut not-mapped area. ....from ppc-32 code. */ if (ULONG_MAX - ent->addr < ent->size) ent->size = ULONG_MAX - ent->addr; - /* cut when vmalloc() area is higher than direct-map area */ - if (VMALLOC_START > (unsigned long)__va(0)) { - if (ent->addr > VMALLOC_START) - goto free_out; + /* + * We've already checked virt_addr_valid so we know this address + * is a valid pointer, therefore we can check against it to determine + * if we need to trim + */ + if (VMALLOC_START > ent->addr) { if (VMALLOC_START - ent->addr < ent->size) ent->size = VMALLOC_START - ent->addr; } diff --git a/fs/xfs/libxfs/xfs_attr.c b/fs/xfs/libxfs/xfs_attr.c index ce4a34a2751d..35a124400d60 100644 --- a/fs/xfs/libxfs/xfs_attr.c +++ b/fs/xfs/libxfs/xfs_attr.c @@ -511,7 +511,14 @@ xfs_attr_shortform_addname(xfs_da_args_t *args) if (args->flags & ATTR_CREATE) return retval; retval = xfs_attr_shortform_remove(args); - ASSERT(retval == 0); + if (retval) + return retval; + /* + * Since we have removed the old attr, clear ATTR_REPLACE so + * that the leaf format add routine won't trip over the attr + * not being around. + */ + args->flags &= ~ATTR_REPLACE; } if (args->namelen >= XFS_ATTR_SF_ENTSIZE_MAX || diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index 6a7c2f03ea11..040eeda8426f 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -725,12 +725,16 @@ xfs_bmap_extents_to_btree( *logflagsp = 0; if ((error = xfs_alloc_vextent(&args))) { xfs_iroot_realloc(ip, -1, whichfork); + ASSERT(ifp->if_broot == NULL); + XFS_IFORK_FMT_SET(ip, whichfork, XFS_DINODE_FMT_EXTENTS); xfs_btree_del_cursor(cur, XFS_BTREE_ERROR); return error; } if (WARN_ON_ONCE(args.fsbno == NULLFSBLOCK)) { xfs_iroot_realloc(ip, -1, whichfork); + ASSERT(ifp->if_broot == NULL); + XFS_IFORK_FMT_SET(ip, whichfork, XFS_DINODE_FMT_EXTENTS); xfs_btree_del_cursor(cur, XFS_BTREE_ERROR); return -ENOSPC; } diff --git a/fs/xfs/libxfs/xfs_inode_buf.c b/fs/xfs/libxfs/xfs_inode_buf.c index ef68b1de006a..1201107eabc6 100644 --- a/fs/xfs/libxfs/xfs_inode_buf.c +++ b/fs/xfs/libxfs/xfs_inode_buf.c @@ -466,6 +466,8 @@ xfs_dinode_verify( return __this_address; if (di_size > XFS_DFORK_DSIZE(dip, mp)) return __this_address; + if (dip->di_nextents) + return __this_address; /* fall through */ case XFS_DINODE_FMT_EXTENTS: case XFS_DINODE_FMT_BTREE: @@ -484,12 +486,31 @@ xfs_dinode_verify( if (XFS_DFORK_Q(dip)) { switch (dip->di_aformat) { case XFS_DINODE_FMT_LOCAL: + if (dip->di_anextents) + return __this_address; + /* fall through */ case XFS_DINODE_FMT_EXTENTS: case XFS_DINODE_FMT_BTREE: break; default: return __this_address; } + } else { + /* + * If there is no fork offset, this may be a freshly-made inode + * in a new disk cluster, in which case di_aformat is zeroed. + * Otherwise, such an inode must be in EXTENTS format; this goes + * for freed inodes as well. + */ + switch (dip->di_aformat) { + case 0: + case XFS_DINODE_FMT_EXTENTS: + break; + default: + return __this_address; + } + if (dip->di_anextents) + return __this_address; } /* only version 3 or greater inodes are extensively verified here */ diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c index 299aee4b7b0b..e70fb8ccecea 100644 --- a/fs/xfs/xfs_file.c +++ b/fs/xfs/xfs_file.c @@ -778,22 +778,26 @@ xfs_file_fallocate( if (error) goto out_unlock; } else if (mode & FALLOC_FL_INSERT_RANGE) { - unsigned int blksize_mask = i_blocksize(inode) - 1; + unsigned int blksize_mask = i_blocksize(inode) - 1; + loff_t isize = i_size_read(inode); - new_size = i_size_read(inode) + len; if (offset & blksize_mask || len & blksize_mask) { error = -EINVAL; goto out_unlock; } - /* check the new inode size does not wrap through zero */ - if (new_size > inode->i_sb->s_maxbytes) { + /* + * New inode size must not exceed ->s_maxbytes, accounting for + * possible signed overflow. + */ + if (inode->i_sb->s_maxbytes - isize < len) { error = -EFBIG; goto out_unlock; } + new_size = isize + len; /* Offset should be less than i_size */ - if (offset >= i_size_read(inode)) { + if (offset >= isize) { error = -EINVAL; goto out_unlock; } @@ -876,8 +880,18 @@ xfs_file_dedupe_range( struct file *dst_file, u64 dst_loff) { + struct inode *srci = file_inode(src_file); + u64 max_dedupe; int error; + /* + * Since we have to read all these pages in to compare them, cut + * it off at MAX_RW_COUNT/2 rounded down to the nearest block. + * That means we won't do more than MAX_RW_COUNT IO per request. + */ + max_dedupe = (MAX_RW_COUNT >> 1) & ~(i_blocksize(srci) - 1); + if (len > max_dedupe) + len = max_dedupe; error = xfs_reflink_remap_range(src_file, loff, dst_file, dst_loff, len, true); if (error) diff --git a/include/drm/amd_asic_type.h b/include/drm/amd_asic_type.h index 695bde7eb055..dd63d08cc54e 100644 --- a/include/drm/amd_asic_type.h +++ b/include/drm/amd_asic_type.h @@ -47,6 +47,7 @@ enum amd_asic_type { CHIP_VEGAM, CHIP_VEGA10, CHIP_VEGA12, + CHIP_VEGA20, CHIP_RAVEN, CHIP_LAST, }; diff --git a/include/drm/gpu_scheduler.h b/include/drm/gpu_scheduler.h index 52380067a43f..dec655894d08 100644 --- a/include/drm/gpu_scheduler.h +++ b/include/drm/gpu_scheduler.h @@ -56,7 +56,6 @@ struct drm_sched_entity { spinlock_t rq_lock; struct drm_gpu_scheduler *sched; - spinlock_t queue_lock; struct spsc_queue job_queue; atomic_t fence_seq; diff --git a/include/dt-bindings/clock/stm32mp1-clks.h b/include/dt-bindings/clock/stm32mp1-clks.h index 86e3ec662ef4..90ec780bfc68 100644 --- a/include/dt-bindings/clock/stm32mp1-clks.h +++ b/include/dt-bindings/clock/stm32mp1-clks.h @@ -76,7 +76,7 @@ #define I2C6 63 #define USART1 64 #define RTCAPB 65 -#define TZC 66 +#define TZC1 66 #define TZPC 67 #define IWDG1 68 #define BSEC 69 @@ -123,6 +123,7 @@ #define CRC1 110 #define USBH 111 #define ETHSTP 112 +#define TZC2 113 /* Kernel clocks */ #define SDMMC1_K 118 @@ -228,7 +229,6 @@ #define CK_MCO2 212 /* TRACE & DEBUG clocks */ -#define DBG 213 #define CK_DBG 214 #define CK_TRACE 215 diff --git a/include/kvm/arm_vgic.h b/include/kvm/arm_vgic.h index 24f03941ada8..e7efe12a81bd 100644 --- a/include/kvm/arm_vgic.h +++ b/include/kvm/arm_vgic.h @@ -131,6 +131,7 @@ struct vgic_irq { u32 mpidr; /* GICv3 target VCPU */ }; u8 source; /* GICv2 SGIs only */ + u8 active_source; /* GICv2 SGIs only */ u8 priority; enum vgic_irq_config config; /* Level or edge */ diff --git a/include/linux/bpf.h b/include/linux/bpf.h index 486e65e3db26..469b20e1dd7e 100644 --- a/include/linux/bpf.h +++ b/include/linux/bpf.h @@ -31,6 +31,7 @@ struct bpf_map_ops { void (*map_release)(struct bpf_map *map, struct file *map_file); void (*map_free)(struct bpf_map *map); int (*map_get_next_key)(struct bpf_map *map, void *key, void *next_key); + void (*map_release_uref)(struct bpf_map *map); /* funcs callable from userspace and from eBPF programs */ void *(*map_lookup_elem)(struct bpf_map *map, void *key); @@ -351,6 +352,7 @@ int bpf_prog_array_copy(struct bpf_prog_array __rcu *old_array, struct bpf_prog **_prog, *__prog; \ struct bpf_prog_array *_array; \ u32 _ret = 1; \ + preempt_disable(); \ rcu_read_lock(); \ _array = rcu_dereference(array); \ if (unlikely(check_non_null && !_array))\ @@ -362,6 +364,7 @@ int bpf_prog_array_copy(struct bpf_prog_array __rcu *old_array, } \ _out: \ rcu_read_unlock(); \ + preempt_enable_no_resched(); \ _ret; \ }) @@ -434,7 +437,6 @@ int bpf_stackmap_copy(struct bpf_map *map, void *key, void *value); int bpf_fd_array_map_update_elem(struct bpf_map *map, struct file *map_file, void *key, void *value, u64 map_flags); int bpf_fd_array_map_lookup_elem(struct bpf_map *map, void *key, u32 *value); -void bpf_fd_array_map_clear(struct bpf_map *map); int bpf_fd_htab_map_update_elem(struct bpf_map *map, struct file *map_file, void *key, void *value, u64 map_flags); int bpf_fd_htab_map_lookup_elem(struct bpf_map *map, void *key, u32 *value); diff --git a/include/linux/brcmphy.h b/include/linux/brcmphy.h index d3339dd48b1a..b324e01ccf2d 100644 --- a/include/linux/brcmphy.h +++ b/include/linux/brcmphy.h @@ -25,6 +25,7 @@ #define PHY_ID_BCM54612E 0x03625e60 #define PHY_ID_BCM54616S 0x03625d10 #define PHY_ID_BCM57780 0x03625d90 +#define PHY_ID_BCM89610 0x03625cd0 #define PHY_ID_BCM7250 0xae025280 #define PHY_ID_BCM7260 0xae025190 diff --git a/include/linux/ceph/osd_client.h b/include/linux/ceph/osd_client.h index 528ccc943cee..96bb32285989 100644 --- a/include/linux/ceph/osd_client.h +++ b/include/linux/ceph/osd_client.h @@ -77,7 +77,10 @@ struct ceph_osd_data { u32 bio_length; }; #endif /* CONFIG_BLOCK */ - struct ceph_bvec_iter bvec_pos; + struct { + struct ceph_bvec_iter bvec_pos; + u32 num_bvecs; + }; }; }; @@ -412,6 +415,10 @@ void osd_req_op_extent_osd_data_bio(struct ceph_osd_request *osd_req, struct ceph_bio_iter *bio_pos, u32 bio_length); #endif /* CONFIG_BLOCK */ +void osd_req_op_extent_osd_data_bvecs(struct ceph_osd_request *osd_req, + unsigned int which, + struct bio_vec *bvecs, u32 num_bvecs, + u32 bytes); void osd_req_op_extent_osd_data_bvec_pos(struct ceph_osd_request *osd_req, unsigned int which, struct ceph_bvec_iter *bvec_pos); @@ -426,7 +433,8 @@ extern void osd_req_op_cls_request_data_pages(struct ceph_osd_request *, bool own_pages); void osd_req_op_cls_request_data_bvecs(struct ceph_osd_request *osd_req, unsigned int which, - struct bio_vec *bvecs, u32 bytes); + struct bio_vec *bvecs, u32 num_bvecs, + u32 bytes); extern void osd_req_op_cls_response_data_pages(struct ceph_osd_request *, unsigned int which, struct page **pages, u64 length, diff --git a/include/linux/clk-provider.h b/include/linux/clk-provider.h index 210a890008f9..1d25e149c1c5 100644 --- a/include/linux/clk-provider.h +++ b/include/linux/clk-provider.h @@ -765,6 +765,9 @@ int __clk_mux_determine_rate(struct clk_hw *hw, int __clk_determine_rate(struct clk_hw *core, struct clk_rate_request *req); int __clk_mux_determine_rate_closest(struct clk_hw *hw, struct clk_rate_request *req); +int clk_mux_determine_rate_flags(struct clk_hw *hw, + struct clk_rate_request *req, + unsigned long flags); void clk_hw_reparent(struct clk_hw *hw, struct clk_hw *new_parent); void clk_hw_set_rate_range(struct clk_hw *hw, unsigned long min_rate, unsigned long max_rate); diff --git a/include/linux/genhd.h b/include/linux/genhd.h index c826b0b5232a..6cb8a5789668 100644 --- a/include/linux/genhd.h +++ b/include/linux/genhd.h @@ -368,7 +368,9 @@ static inline void free_part_stats(struct hd_struct *part) part_stat_add(cpu, gendiskp, field, -subnd) void part_in_flight(struct request_queue *q, struct hd_struct *part, - unsigned int inflight[2]); + unsigned int inflight[2]); +void part_in_flight_rw(struct request_queue *q, struct hd_struct *part, + unsigned int inflight[2]); void part_dec_in_flight(struct request_queue *q, struct hd_struct *part, int rw); void part_inc_in_flight(struct request_queue *q, struct hd_struct *part, diff --git a/include/linux/host1x.h b/include/linux/host1x.h index ddf7f9ca86cc..89110d896d72 100644 --- a/include/linux/host1x.h +++ b/include/linux/host1x.h @@ -192,13 +192,6 @@ struct host1x_reloc { unsigned long shift; }; -struct host1x_waitchk { - struct host1x_bo *bo; - u32 offset; - u32 syncpt_id; - u32 thresh; -}; - struct host1x_job { /* When refcount goes to zero, job can be freed */ struct kref ref; @@ -209,19 +202,15 @@ struct host1x_job { /* Channel where job is submitted to */ struct host1x_channel *channel; - u32 client; + /* client where the job originated */ + struct host1x_client *client; /* Gathers and their memory */ struct host1x_job_gather *gathers; unsigned int num_gathers; - /* Wait checks to be processed at submit time */ - struct host1x_waitchk *waitchk; - unsigned int num_waitchk; - u32 waitchk_mask; - /* Array of handles to be pinned & unpinned */ - struct host1x_reloc *relocarray; + struct host1x_reloc *relocs; unsigned int num_relocs; struct host1x_job_unpin_data *unpins; unsigned int num_unpins; @@ -261,10 +250,9 @@ struct host1x_job { }; struct host1x_job *host1x_job_alloc(struct host1x_channel *ch, - u32 num_cmdbufs, u32 num_relocs, - u32 num_waitchks); -void host1x_job_add_gather(struct host1x_job *job, struct host1x_bo *mem_id, - u32 words, u32 offset); + u32 num_cmdbufs, u32 num_relocs); +void host1x_job_add_gather(struct host1x_job *job, struct host1x_bo *bo, + unsigned int words, unsigned int offset); struct host1x_job *host1x_job_get(struct host1x_job *job); void host1x_job_put(struct host1x_job *job); int host1x_job_pin(struct host1x_job *job, struct device *dev); diff --git a/include/linux/kthread.h b/include/linux/kthread.h index c1961761311d..2803264c512f 100644 --- a/include/linux/kthread.h +++ b/include/linux/kthread.h @@ -62,6 +62,7 @@ void *kthread_probe_data(struct task_struct *k); int kthread_park(struct task_struct *k); void kthread_unpark(struct task_struct *k); void kthread_parkme(void); +void kthread_park_complete(struct task_struct *k); int kthreadd(void *unused); extern struct task_struct *kthreadd_task; diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h index 767d193c269a..2a156c5dfadd 100644 --- a/include/linux/mlx5/driver.h +++ b/include/linux/mlx5/driver.h @@ -1284,25 +1284,19 @@ enum { }; static inline const struct cpumask * -mlx5_get_vector_affinity(struct mlx5_core_dev *dev, int vector) +mlx5_get_vector_affinity_hint(struct mlx5_core_dev *dev, int vector) { - const struct cpumask *mask; struct irq_desc *desc; unsigned int irq; int eqn; int err; - err = mlx5_vector2eqn(dev, MLX5_EQ_VEC_COMP_BASE + vector, &eqn, &irq); + err = mlx5_vector2eqn(dev, vector, &eqn, &irq); if (err) return NULL; desc = irq_to_desc(irq); -#ifdef CONFIG_GENERIC_IRQ_EFFECTIVE_AFF_MASK - mask = irq_data_get_effective_affinity_mask(&desc->irq_data); -#else - mask = desc->irq_common_data.affinity; -#endif - return mask; + return desc->affinity_hint; } #endif /* MLX5_DRIVER_H */ diff --git a/include/linux/oom.h b/include/linux/oom.h index 5bad038ac012..6adac113e96d 100644 --- a/include/linux/oom.h +++ b/include/linux/oom.h @@ -95,6 +95,8 @@ static inline int check_stable_address_space(struct mm_struct *mm) return 0; } +void __oom_reap_task_mm(struct mm_struct *mm); + extern unsigned long oom_badness(struct task_struct *p, struct mem_cgroup *memcg, const nodemask_t *nodemask, unsigned long totalpages); diff --git a/include/linux/platform_data/tda9950.h b/include/linux/platform_data/tda9950.h new file mode 100644 index 000000000000..c65efd461102 --- /dev/null +++ b/include/linux/platform_data/tda9950.h @@ -0,0 +1,16 @@ +#ifndef LINUX_PLATFORM_DATA_TDA9950_H +#define LINUX_PLATFORM_DATA_TDA9950_H + +struct device; + +struct tda9950_glue { + struct device *parent; + unsigned long irq_flags; + void *data; + int (*init)(void *); + void (*exit)(void *); + int (*open)(void *); + void (*release)(void *); +}; + +#endif diff --git a/include/linux/rbtree_augmented.h b/include/linux/rbtree_augmented.h index 6bfd2b581f75..af8a61be2d8d 100644 --- a/include/linux/rbtree_augmented.h +++ b/include/linux/rbtree_augmented.h @@ -26,6 +26,7 @@ #include <linux/compiler.h> #include <linux/rbtree.h> +#include <linux/rcupdate.h> /* * Please note - only struct rb_augment_callbacks and the prototypes for diff --git a/include/linux/rbtree_latch.h b/include/linux/rbtree_latch.h index ece43e882b56..7d012faa509a 100644 --- a/include/linux/rbtree_latch.h +++ b/include/linux/rbtree_latch.h @@ -35,6 +35,7 @@ #include <linux/rbtree.h> #include <linux/seqlock.h> +#include <linux/rcupdate.h> struct latch_tree_node { struct rb_node node[2]; diff --git a/include/linux/remoteproc.h b/include/linux/remoteproc.h index d09a9c7af109..dfdaede9139e 100644 --- a/include/linux/remoteproc.h +++ b/include/linux/remoteproc.h @@ -569,7 +569,7 @@ static inline struct rproc *vdev_to_rproc(struct virtio_device *vdev) void rproc_add_subdev(struct rproc *rproc, struct rproc_subdev *subdev, int (*probe)(struct rproc_subdev *subdev), - void (*remove)(struct rproc_subdev *subdev, bool graceful)); + void (*remove)(struct rproc_subdev *subdev, bool crashed)); void rproc_remove_subdev(struct rproc *rproc, struct rproc_subdev *subdev); diff --git a/include/linux/sched.h b/include/linux/sched.h index b3d697f3b573..c2413703f45d 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -112,17 +112,36 @@ struct task_group; #ifdef CONFIG_DEBUG_ATOMIC_SLEEP +/* + * Special states are those that do not use the normal wait-loop pattern. See + * the comment with set_special_state(). + */ +#define is_special_task_state(state) \ + ((state) & (__TASK_STOPPED | __TASK_TRACED | TASK_DEAD)) + #define __set_current_state(state_value) \ do { \ + WARN_ON_ONCE(is_special_task_state(state_value));\ current->task_state_change = _THIS_IP_; \ current->state = (state_value); \ } while (0) + #define set_current_state(state_value) \ do { \ + WARN_ON_ONCE(is_special_task_state(state_value));\ current->task_state_change = _THIS_IP_; \ smp_store_mb(current->state, (state_value)); \ } while (0) +#define set_special_state(state_value) \ + do { \ + unsigned long flags; /* may shadow */ \ + WARN_ON_ONCE(!is_special_task_state(state_value)); \ + raw_spin_lock_irqsave(¤t->pi_lock, flags); \ + current->task_state_change = _THIS_IP_; \ + current->state = (state_value); \ + raw_spin_unlock_irqrestore(¤t->pi_lock, flags); \ + } while (0) #else /* * set_current_state() includes a barrier so that the write of current->state @@ -144,8 +163,8 @@ struct task_group; * * The above is typically ordered against the wakeup, which does: * - * need_sleep = false; - * wake_up_state(p, TASK_UNINTERRUPTIBLE); + * need_sleep = false; + * wake_up_state(p, TASK_UNINTERRUPTIBLE); * * Where wake_up_state() (and all other wakeup primitives) imply enough * barriers to order the store of the variable against wakeup. @@ -154,12 +173,33 @@ struct task_group; * once it observes the TASK_UNINTERRUPTIBLE store the waking CPU can issue a * TASK_RUNNING store which can collide with __set_current_state(TASK_RUNNING). * - * This is obviously fine, since they both store the exact same value. + * However, with slightly different timing the wakeup TASK_RUNNING store can + * also collide with the TASK_UNINTERRUPTIBLE store. Loosing that store is not + * a problem either because that will result in one extra go around the loop + * and our @cond test will save the day. * * Also see the comments of try_to_wake_up(). */ -#define __set_current_state(state_value) do { current->state = (state_value); } while (0) -#define set_current_state(state_value) smp_store_mb(current->state, (state_value)) +#define __set_current_state(state_value) \ + current->state = (state_value) + +#define set_current_state(state_value) \ + smp_store_mb(current->state, (state_value)) + +/* + * set_special_state() should be used for those states when the blocking task + * can not use the regular condition based wait-loop. In that case we must + * serialize against wakeups such that any possible in-flight TASK_RUNNING stores + * will not collide with our state change. + */ +#define set_special_state(state_value) \ + do { \ + unsigned long flags; /* may shadow */ \ + raw_spin_lock_irqsave(¤t->pi_lock, flags); \ + current->state = (state_value); \ + raw_spin_unlock_irqrestore(¤t->pi_lock, flags); \ + } while (0) + #endif /* Task command name length: */ diff --git a/include/linux/sched/signal.h b/include/linux/sched/signal.h index a7ce74c74e49..113d1ad1ced7 100644 --- a/include/linux/sched/signal.h +++ b/include/linux/sched/signal.h @@ -280,7 +280,7 @@ static inline void kernel_signal_stop(void) { spin_lock_irq(¤t->sighand->siglock); if (current->jobctl & JOBCTL_STOP_DEQUEUED) - __set_current_state(TASK_STOPPED); + set_special_state(TASK_STOPPED); spin_unlock_irq(¤t->sighand->siglock); schedule(); diff --git a/include/linux/usb/composite.h b/include/linux/usb/composite.h index 4b6b9283fa7b..8675e145ea8b 100644 --- a/include/linux/usb/composite.h +++ b/include/linux/usb/composite.h @@ -52,7 +52,7 @@ #define USB_GADGET_DELAYED_STATUS 0x7fff /* Impossibly large value */ /* big enough to hold our biggest descriptor */ -#define USB_COMP_EP0_BUFSIZ 1024 +#define USB_COMP_EP0_BUFSIZ 4096 /* OS feature descriptor length <= 4kB */ #define USB_COMP_EP0_OS_DESC_BUFSIZ 4096 diff --git a/include/linux/wait_bit.h b/include/linux/wait_bit.h index 9318b2166439..2b0072fa5e92 100644 --- a/include/linux/wait_bit.h +++ b/include/linux/wait_bit.h @@ -305,4 +305,21 @@ do { \ __ret; \ }) +/** + * clear_and_wake_up_bit - clear a bit and wake up anyone waiting on that bit + * + * @bit: the bit of the word being waited on + * @word: the word being waited on, a kernel virtual address + * + * You can use this helper if bitflags are manipulated atomically rather than + * non-atomically under a lock. + */ +static inline void clear_and_wake_up_bit(int bit, void *word) +{ + clear_bit_unlock(bit, word); + /* See wake_up_bit() for which memory barrier you need to use. */ + smp_mb__after_atomic(); + wake_up_bit(word, bit); +} + #endif /* _LINUX_WAIT_BIT_H */ diff --git a/include/media/i2c/tvp7002.h b/include/media/i2c/tvp7002.h index 5ee007c1cead..cb213c136089 100644 --- a/include/media/i2c/tvp7002.h +++ b/include/media/i2c/tvp7002.h @@ -5,7 +5,7 @@ * Author: Santiago Nunez-Corrales <santiago.nunez@ridgerun.com> * * This code is partially based upon the TVP5150 driver - * written by Mauro Carvalho Chehab (mchehab@infradead.org), + * written by Mauro Carvalho Chehab <mchehab@kernel.org>, * the TVP514x driver written by Vaibhav Hiremath <hvaibhav@ti.com> * and the TVP7002 driver in the TI LSP 2.10.00.14 * diff --git a/include/media/videobuf-core.h b/include/media/videobuf-core.h index 0bda0adc744f..60a664febba0 100644 --- a/include/media/videobuf-core.h +++ b/include/media/videobuf-core.h @@ -1,11 +1,11 @@ /* * generic helper functions for handling video4linux capture buffers * - * (c) 2007 Mauro Carvalho Chehab, <mchehab@infradead.org> + * (c) 2007 Mauro Carvalho Chehab, <mchehab@kernel.org> * * Highly based on video-buf written originally by: * (c) 2001,02 Gerd Knorr <kraxel@bytesex.org> - * (c) 2006 Mauro Carvalho Chehab, <mchehab@infradead.org> + * (c) 2006 Mauro Carvalho Chehab, <mchehab@kernel.org> * (c) 2006 Ted Walther and John Sokol * * This program is free software; you can redistribute it and/or modify diff --git a/include/media/videobuf-dma-sg.h b/include/media/videobuf-dma-sg.h index d8b27854e3bf..01bd142b979d 100644 --- a/include/media/videobuf-dma-sg.h +++ b/include/media/videobuf-dma-sg.h @@ -6,11 +6,11 @@ * into PAGE_SIZE chunks). They also assume the driver does not need * to touch the video data. * - * (c) 2007 Mauro Carvalho Chehab, <mchehab@infradead.org> + * (c) 2007 Mauro Carvalho Chehab, <mchehab@kernel.org> * * Highly based on video-buf written originally by: * (c) 2001,02 Gerd Knorr <kraxel@bytesex.org> - * (c) 2006 Mauro Carvalho Chehab, <mchehab@infradead.org> + * (c) 2006 Mauro Carvalho Chehab, <mchehab@kernel.org> * (c) 2006 Ted Walther and John Sokol * * This program is free software; you can redistribute it and/or modify diff --git a/include/media/videobuf-vmalloc.h b/include/media/videobuf-vmalloc.h index 486a97efdb56..36c6a4ad3504 100644 --- a/include/media/videobuf-vmalloc.h +++ b/include/media/videobuf-vmalloc.h @@ -6,7 +6,7 @@ * into PAGE_SIZE chunks). They also assume the driver does not need * to touch the video data. * - * (c) 2007 Mauro Carvalho Chehab, <mchehab@infradead.org> + * (c) 2007 Mauro Carvalho Chehab, <mchehab@kernel.org> * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by diff --git a/include/net/bonding.h b/include/net/bonding.h index f801fc940b29..b52235158836 100644 --- a/include/net/bonding.h +++ b/include/net/bonding.h @@ -198,6 +198,7 @@ struct bonding { struct slave __rcu *primary_slave; struct bond_up_slave __rcu *slave_arr; /* Array of usable slaves */ bool force_primary; + u32 nest_level; s32 slave_cnt; /* never change this value outside the attach/detach wrappers */ int (*recv_probe)(const struct sk_buff *, struct bonding *, struct slave *); diff --git a/include/net/flow_dissector.h b/include/net/flow_dissector.h index 9a074776f70b..d1fcf2442a42 100644 --- a/include/net/flow_dissector.h +++ b/include/net/flow_dissector.h @@ -251,7 +251,7 @@ extern struct flow_dissector flow_keys_buf_dissector; * This structure is used to hold a digest of the full flow keys. This is a * larger "hash" of a flow to allow definitively matching specific flows where * the 32 bit skb->hash is not large enough. The size is limited to 16 bytes so - * that it can by used in CB of skb (see sch_choke for an example). + * that it can be used in CB of skb (see sch_choke for an example). */ #define FLOW_KEYS_DIGEST_LEN 16 struct flow_keys_digest { diff --git a/include/net/mac80211.h b/include/net/mac80211.h index d2279b2d61aa..b2f3a0c018e7 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -2080,7 +2080,7 @@ struct ieee80211_txq { * virtual interface might not be given air time for the transmission of * the frame, as it is not synced with the AP/P2P GO yet, and thus the * deauthentication frame might not be transmitted. - > + * * @IEEE80211_HW_DOESNT_SUPPORT_QOS_NDP: The driver (or firmware) doesn't * support QoS NDP for AP probing - that's most likely a driver bug. * diff --git a/include/net/tls.h b/include/net/tls.h index 3da8e13a6d96..b400d0bb7448 100644 --- a/include/net/tls.h +++ b/include/net/tls.h @@ -148,6 +148,7 @@ struct tls_context { struct scatterlist *partially_sent_record; u16 partially_sent_offset; unsigned long flags; + bool in_tcp_sendpages; u16 pending_open_record_frags; int (*push_pending_record)(struct sock *sk, int flags); diff --git a/include/net/xfrm.h b/include/net/xfrm.h index a872379b69da..45e75c36b738 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -375,6 +375,7 @@ struct xfrm_input_afinfo { int xfrm_input_register_afinfo(const struct xfrm_input_afinfo *afinfo); int xfrm_input_unregister_afinfo(const struct xfrm_input_afinfo *afinfo); +void xfrm_flush_gc(void); void xfrm_state_delete_tunnel(struct xfrm_state *x); struct xfrm_type { diff --git a/include/trace/events/host1x.h b/include/trace/events/host1x.h index 63116362543c..a37ef73092e5 100644 --- a/include/trace/events/host1x.h +++ b/include/trace/events/host1x.h @@ -115,16 +115,15 @@ TRACE_EVENT(host1x_cdma_push_gather, ); TRACE_EVENT(host1x_channel_submit, - TP_PROTO(const char *name, u32 cmdbufs, u32 relocs, u32 waitchks, - u32 syncpt_id, u32 syncpt_incrs), + TP_PROTO(const char *name, u32 cmdbufs, u32 relocs, u32 syncpt_id, + u32 syncpt_incrs), - TP_ARGS(name, cmdbufs, relocs, waitchks, syncpt_id, syncpt_incrs), + TP_ARGS(name, cmdbufs, relocs, syncpt_id, syncpt_incrs), TP_STRUCT__entry( __field(const char *, name) __field(u32, cmdbufs) __field(u32, relocs) - __field(u32, waitchks) __field(u32, syncpt_id) __field(u32, syncpt_incrs) ), @@ -133,15 +132,14 @@ TRACE_EVENT(host1x_channel_submit, __entry->name = name; __entry->cmdbufs = cmdbufs; __entry->relocs = relocs; - __entry->waitchks = waitchks; __entry->syncpt_id = syncpt_id; __entry->syncpt_incrs = syncpt_incrs; ), - TP_printk("name=%s, cmdbufs=%u, relocs=%u, waitchks=%d," - "syncpt_id=%u, syncpt_incrs=%u", - __entry->name, __entry->cmdbufs, __entry->relocs, __entry->waitchks, - __entry->syncpt_id, __entry->syncpt_incrs) + TP_printk("name=%s, cmdbufs=%u, relocs=%u, syncpt_id=%u, " + "syncpt_incrs=%u", + __entry->name, __entry->cmdbufs, __entry->relocs, + __entry->syncpt_id, __entry->syncpt_incrs) ); TRACE_EVENT(host1x_channel_submitted, diff --git a/include/trace/events/initcall.h b/include/trace/events/initcall.h index 8d6cf10d27c9..eb903c3f195f 100644 --- a/include/trace/events/initcall.h +++ b/include/trace/events/initcall.h @@ -31,7 +31,11 @@ TRACE_EVENT(initcall_start, TP_ARGS(func), TP_STRUCT__entry( - __field(initcall_t, func) + /* + * Use field_struct to avoid is_signed_type() + * comparison of a function pointer + */ + __field_struct(initcall_t, func) ), TP_fast_assign( @@ -48,8 +52,12 @@ TRACE_EVENT(initcall_finish, TP_ARGS(func, ret), TP_STRUCT__entry( - __field(initcall_t, func) - __field(int, ret) + /* + * Use field_struct to avoid is_signed_type() + * comparison of a function pointer + */ + __field_struct(initcall_t, func) + __field(int, ret) ), TP_fast_assign( diff --git a/include/trace/events/rxrpc.h b/include/trace/events/rxrpc.h index 9e96c2fe2793..077e664ac9a2 100644 --- a/include/trace/events/rxrpc.h +++ b/include/trace/events/rxrpc.h @@ -15,6 +15,7 @@ #define _TRACE_RXRPC_H #include <linux/tracepoint.h> +#include <linux/errqueue.h> /* * Define enums for tracing information. @@ -210,6 +211,20 @@ enum rxrpc_congest_change { rxrpc_cong_saw_nack, }; +enum rxrpc_tx_fail_trace { + rxrpc_tx_fail_call_abort, + rxrpc_tx_fail_call_ack, + rxrpc_tx_fail_call_data_frag, + rxrpc_tx_fail_call_data_nofrag, + rxrpc_tx_fail_call_final_resend, + rxrpc_tx_fail_conn_abort, + rxrpc_tx_fail_conn_challenge, + rxrpc_tx_fail_conn_response, + rxrpc_tx_fail_reject, + rxrpc_tx_fail_version_keepalive, + rxrpc_tx_fail_version_reply, +}; + #endif /* end __RXRPC_DECLARE_TRACE_ENUMS_ONCE_ONLY */ /* @@ -437,6 +452,19 @@ enum rxrpc_congest_change { EM(RXRPC_CALL_LOCAL_ERROR, "LocalError") \ E_(RXRPC_CALL_NETWORK_ERROR, "NetError") +#define rxrpc_tx_fail_traces \ + EM(rxrpc_tx_fail_call_abort, "CallAbort") \ + EM(rxrpc_tx_fail_call_ack, "CallAck") \ + EM(rxrpc_tx_fail_call_data_frag, "CallDataFrag") \ + EM(rxrpc_tx_fail_call_data_nofrag, "CallDataNofrag") \ + EM(rxrpc_tx_fail_call_final_resend, "CallFinalResend") \ + EM(rxrpc_tx_fail_conn_abort, "ConnAbort") \ + EM(rxrpc_tx_fail_conn_challenge, "ConnChall") \ + EM(rxrpc_tx_fail_conn_response, "ConnResp") \ + EM(rxrpc_tx_fail_reject, "Reject") \ + EM(rxrpc_tx_fail_version_keepalive, "VerKeepalive") \ + E_(rxrpc_tx_fail_version_reply, "VerReply") + /* * Export enum symbols via userspace. */ @@ -460,6 +488,7 @@ rxrpc_propose_ack_traces; rxrpc_propose_ack_outcomes; rxrpc_congest_modes; rxrpc_congest_changes; +rxrpc_tx_fail_traces; /* * Now redefine the EM() and E_() macros to map the enums to the strings that @@ -1374,6 +1403,62 @@ TRACE_EVENT(rxrpc_resend, __entry->anno) ); +TRACE_EVENT(rxrpc_rx_icmp, + TP_PROTO(struct rxrpc_peer *peer, struct sock_extended_err *ee, + struct sockaddr_rxrpc *srx), + + TP_ARGS(peer, ee, srx), + + TP_STRUCT__entry( + __field(unsigned int, peer ) + __field_struct(struct sock_extended_err, ee ) + __field_struct(struct sockaddr_rxrpc, srx ) + ), + + TP_fast_assign( + __entry->peer = peer->debug_id; + memcpy(&__entry->ee, ee, sizeof(__entry->ee)); + memcpy(&__entry->srx, srx, sizeof(__entry->srx)); + ), + + TP_printk("P=%08x o=%u t=%u c=%u i=%u d=%u e=%d %pISp", + __entry->peer, + __entry->ee.ee_origin, + __entry->ee.ee_type, + __entry->ee.ee_code, + __entry->ee.ee_info, + __entry->ee.ee_data, + __entry->ee.ee_errno, + &__entry->srx.transport) + ); + +TRACE_EVENT(rxrpc_tx_fail, + TP_PROTO(unsigned int debug_id, rxrpc_serial_t serial, int ret, + enum rxrpc_tx_fail_trace what), + + TP_ARGS(debug_id, serial, ret, what), + + TP_STRUCT__entry( + __field(unsigned int, debug_id ) + __field(rxrpc_serial_t, serial ) + __field(int, ret ) + __field(enum rxrpc_tx_fail_trace, what ) + ), + + TP_fast_assign( + __entry->debug_id = debug_id; + __entry->serial = serial; + __entry->ret = ret; + __entry->what = what; + ), + + TP_printk("c=%08x r=%x ret=%d %s", + __entry->debug_id, + __entry->serial, + __entry->ret, + __print_symbolic(__entry->what, rxrpc_tx_fail_traces)) + ); + #endif /* _TRACE_RXRPC_H */ /* This part must be outside protection */ diff --git a/include/trace/events/sunrpc.h b/include/trace/events/sunrpc.h index 335d87242439..bbb08a3ef5cc 100644 --- a/include/trace/events/sunrpc.h +++ b/include/trace/events/sunrpc.h @@ -224,6 +224,8 @@ TRACE_EVENT(rpc_stats_latency, TP_ARGS(task, backlog, rtt, execute), TP_STRUCT__entry( + __field(unsigned int, task_id) + __field(unsigned int, client_id) __field(u32, xid) __field(int, version) __string(progname, task->tk_client->cl_program->name) @@ -231,13 +233,11 @@ TRACE_EVENT(rpc_stats_latency, __field(unsigned long, backlog) __field(unsigned long, rtt) __field(unsigned long, execute) - __string(addr, - task->tk_xprt->address_strings[RPC_DISPLAY_ADDR]) - __string(port, - task->tk_xprt->address_strings[RPC_DISPLAY_PORT]) ), TP_fast_assign( + __entry->client_id = task->tk_client->cl_clid; + __entry->task_id = task->tk_pid; __entry->xid = be32_to_cpu(task->tk_rqstp->rq_xid); __entry->version = task->tk_client->cl_vers; __assign_str(progname, task->tk_client->cl_program->name) @@ -245,14 +245,10 @@ TRACE_EVENT(rpc_stats_latency, __entry->backlog = ktime_to_us(backlog); __entry->rtt = ktime_to_us(rtt); __entry->execute = ktime_to_us(execute); - __assign_str(addr, - task->tk_xprt->address_strings[RPC_DISPLAY_ADDR]); - __assign_str(port, - task->tk_xprt->address_strings[RPC_DISPLAY_PORT]); ), - TP_printk("peer=[%s]:%s xid=0x%08x %sv%d %s backlog=%lu rtt=%lu execute=%lu", - __get_str(addr), __get_str(port), __entry->xid, + TP_printk("task:%u@%d xid=0x%08x %sv%d %s backlog=%lu rtt=%lu execute=%lu", + __entry->task_id, __entry->client_id, __entry->xid, __get_str(progname), __entry->version, __get_str(procname), __entry->backlog, __entry->rtt, __entry->execute) ); diff --git a/include/uapi/drm/tegra_drm.h b/include/uapi/drm/tegra_drm.h index d954f8c33321..c4df3c3668b3 100644 --- a/include/uapi/drm/tegra_drm.h +++ b/include/uapi/drm/tegra_drm.h @@ -32,143 +32,615 @@ extern "C" { #define DRM_TEGRA_GEM_CREATE_TILED (1 << 0) #define DRM_TEGRA_GEM_CREATE_BOTTOM_UP (1 << 1) +/** + * struct drm_tegra_gem_create - parameters for the GEM object creation IOCTL + */ struct drm_tegra_gem_create { + /** + * @size: + * + * The size, in bytes, of the buffer object to be created. + */ __u64 size; + + /** + * @flags: + * + * A bitmask of flags that influence the creation of GEM objects: + * + * DRM_TEGRA_GEM_CREATE_TILED + * Use the 16x16 tiling format for this buffer. + * + * DRM_TEGRA_GEM_CREATE_BOTTOM_UP + * The buffer has a bottom-up layout. + */ __u32 flags; + + /** + * @handle: + * + * The handle of the created GEM object. Set by the kernel upon + * successful completion of the IOCTL. + */ __u32 handle; }; +/** + * struct drm_tegra_gem_mmap - parameters for the GEM mmap IOCTL + */ struct drm_tegra_gem_mmap { + /** + * @handle: + * + * Handle of the GEM object to obtain an mmap offset for. + */ __u32 handle; + + /** + * @pad: + * + * Structure padding that may be used in the future. Must be 0. + */ __u32 pad; + + /** + * @offset: + * + * The mmap offset for the given GEM object. Set by the kernel upon + * successful completion of the IOCTL. + */ __u64 offset; }; +/** + * struct drm_tegra_syncpt_read - parameters for the read syncpoint IOCTL + */ struct drm_tegra_syncpt_read { + /** + * @id: + * + * ID of the syncpoint to read the current value from. + */ __u32 id; + + /** + * @value: + * + * The current syncpoint value. Set by the kernel upon successful + * completion of the IOCTL. + */ __u32 value; }; +/** + * struct drm_tegra_syncpt_incr - parameters for the increment syncpoint IOCTL + */ struct drm_tegra_syncpt_incr { + /** + * @id: + * + * ID of the syncpoint to increment. + */ __u32 id; + + /** + * @pad: + * + * Structure padding that may be used in the future. Must be 0. + */ __u32 pad; }; +/** + * struct drm_tegra_syncpt_wait - parameters for the wait syncpoint IOCTL + */ struct drm_tegra_syncpt_wait { + /** + * @id: + * + * ID of the syncpoint to wait on. + */ __u32 id; + + /** + * @thresh: + * + * Threshold value for which to wait. + */ __u32 thresh; + + /** + * @timeout: + * + * Timeout, in milliseconds, to wait. + */ __u32 timeout; + + /** + * @value: + * + * The new syncpoint value after the wait. Set by the kernel upon + * successful completion of the IOCTL. + */ __u32 value; }; #define DRM_TEGRA_NO_TIMEOUT (0xffffffff) +/** + * struct drm_tegra_open_channel - parameters for the open channel IOCTL + */ struct drm_tegra_open_channel { + /** + * @client: + * + * The client ID for this channel. + */ __u32 client; + + /** + * @pad: + * + * Structure padding that may be used in the future. Must be 0. + */ __u32 pad; + + /** + * @context: + * + * The application context of this channel. Set by the kernel upon + * successful completion of the IOCTL. This context needs to be passed + * to the DRM_TEGRA_CHANNEL_CLOSE or the DRM_TEGRA_SUBMIT IOCTLs. + */ __u64 context; }; +/** + * struct drm_tegra_close_channel - parameters for the close channel IOCTL + */ struct drm_tegra_close_channel { + /** + * @context: + * + * The application context of this channel. This is obtained from the + * DRM_TEGRA_OPEN_CHANNEL IOCTL. + */ __u64 context; }; +/** + * struct drm_tegra_get_syncpt - parameters for the get syncpoint IOCTL + */ struct drm_tegra_get_syncpt { + /** + * @context: + * + * The application context identifying the channel for which to obtain + * the syncpoint ID. + */ __u64 context; + + /** + * @index: + * + * Index of the client syncpoint for which to obtain the ID. + */ __u32 index; + + /** + * @id: + * + * The ID of the given syncpoint. Set by the kernel upon successful + * completion of the IOCTL. + */ __u32 id; }; +/** + * struct drm_tegra_get_syncpt_base - parameters for the get wait base IOCTL + */ struct drm_tegra_get_syncpt_base { + /** + * @context: + * + * The application context identifying for which channel to obtain the + * wait base. + */ __u64 context; + + /** + * @syncpt: + * + * ID of the syncpoint for which to obtain the wait base. + */ __u32 syncpt; + + /** + * @id: + * + * The ID of the wait base corresponding to the client syncpoint. Set + * by the kernel upon successful completion of the IOCTL. + */ __u32 id; }; +/** + * struct drm_tegra_syncpt - syncpoint increment operation + */ struct drm_tegra_syncpt { + /** + * @id: + * + * ID of the syncpoint to operate on. + */ __u32 id; + + /** + * @incrs: + * + * Number of increments to perform for the syncpoint. + */ __u32 incrs; }; +/** + * struct drm_tegra_cmdbuf - structure describing a command buffer + */ struct drm_tegra_cmdbuf { + /** + * @handle: + * + * Handle to a GEM object containing the command buffer. + */ __u32 handle; + + /** + * @offset: + * + * Offset, in bytes, into the GEM object identified by @handle at + * which the command buffer starts. + */ __u32 offset; + + /** + * @words: + * + * Number of 32-bit words in this command buffer. + */ __u32 words; + + /** + * @pad: + * + * Structure padding that may be used in the future. Must be 0. + */ __u32 pad; }; +/** + * struct drm_tegra_reloc - GEM object relocation structure + */ struct drm_tegra_reloc { struct { + /** + * @cmdbuf.handle: + * + * Handle to the GEM object containing the command buffer for + * which to perform this GEM object relocation. + */ __u32 handle; + + /** + * @cmdbuf.offset: + * + * Offset, in bytes, into the command buffer at which to + * insert the relocated address. + */ __u32 offset; } cmdbuf; struct { + /** + * @target.handle: + * + * Handle to the GEM object to be relocated. + */ __u32 handle; + + /** + * @target.offset: + * + * Offset, in bytes, into the target GEM object at which the + * relocated data starts. + */ __u32 offset; } target; + + /** + * @shift: + * + * The number of bits by which to shift relocated addresses. + */ __u32 shift; + + /** + * @pad: + * + * Structure padding that may be used in the future. Must be 0. + */ __u32 pad; }; +/** + * struct drm_tegra_waitchk - wait check structure + */ struct drm_tegra_waitchk { + /** + * @handle: + * + * Handle to the GEM object containing a command stream on which to + * perform the wait check. + */ __u32 handle; + + /** + * @offset: + * + * Offset, in bytes, of the location in the command stream to perform + * the wait check on. + */ __u32 offset; + + /** + * @syncpt: + * + * ID of the syncpoint to wait check. + */ __u32 syncpt; + + /** + * @thresh: + * + * Threshold value for which to check. + */ __u32 thresh; }; +/** + * struct drm_tegra_submit - job submission structure + */ struct drm_tegra_submit { + /** + * @context: + * + * The application context identifying the channel to use for the + * execution of this job. + */ __u64 context; + + /** + * @num_syncpts: + * + * The number of syncpoints operated on by this job. This defines the + * length of the array pointed to by @syncpts. + */ __u32 num_syncpts; + + /** + * @num_cmdbufs: + * + * The number of command buffers to execute as part of this job. This + * defines the length of the array pointed to by @cmdbufs. + */ __u32 num_cmdbufs; + + /** + * @num_relocs: + * + * The number of relocations to perform before executing this job. + * This defines the length of the array pointed to by @relocs. + */ __u32 num_relocs; + + /** + * @num_waitchks: + * + * The number of wait checks to perform as part of this job. This + * defines the length of the array pointed to by @waitchks. + */ __u32 num_waitchks; + + /** + * @waitchk_mask: + * + * Bitmask of valid wait checks. + */ __u32 waitchk_mask; + + /** + * @timeout: + * + * Timeout, in milliseconds, before this job is cancelled. + */ __u32 timeout; + + /** + * @syncpts: + * + * A pointer to an array of &struct drm_tegra_syncpt structures that + * specify the syncpoint operations performed as part of this job. + * The number of elements in the array must be equal to the value + * given by @num_syncpts. + */ __u64 syncpts; + + /** + * @cmdbufs: + * + * A pointer to an array of &struct drm_tegra_cmdbuf structures that + * define the command buffers to execute as part of this job. The + * number of elements in the array must be equal to the value given + * by @num_syncpts. + */ __u64 cmdbufs; + + /** + * @relocs: + * + * A pointer to an array of &struct drm_tegra_reloc structures that + * specify the relocations that need to be performed before executing + * this job. The number of elements in the array must be equal to the + * value given by @num_relocs. + */ __u64 relocs; + + /** + * @waitchks: + * + * A pointer to an array of &struct drm_tegra_waitchk structures that + * specify the wait checks to be performed while executing this job. + * The number of elements in the array must be equal to the value + * given by @num_waitchks. + */ __u64 waitchks; - __u32 fence; /* Return value */ - __u32 reserved[5]; /* future expansion */ + /** + * @fence: + * + * The threshold of the syncpoint associated with this job after it + * has been completed. Set by the kernel upon successful completion of + * the IOCTL. This can be used with the DRM_TEGRA_SYNCPT_WAIT IOCTL to + * wait for this job to be finished. + */ + __u32 fence; + + /** + * @reserved: + * + * This field is reserved for future use. Must be 0. + */ + __u32 reserved[5]; }; #define DRM_TEGRA_GEM_TILING_MODE_PITCH 0 #define DRM_TEGRA_GEM_TILING_MODE_TILED 1 #define DRM_TEGRA_GEM_TILING_MODE_BLOCK 2 +/** + * struct drm_tegra_gem_set_tiling - parameters for the set tiling IOCTL + */ struct drm_tegra_gem_set_tiling { - /* input */ + /** + * @handle: + * + * Handle to the GEM object for which to set the tiling parameters. + */ __u32 handle; + + /** + * @mode: + * + * The tiling mode to set. Must be one of: + * + * DRM_TEGRA_GEM_TILING_MODE_PITCH + * pitch linear format + * + * DRM_TEGRA_GEM_TILING_MODE_TILED + * 16x16 tiling format + * + * DRM_TEGRA_GEM_TILING_MODE_BLOCK + * 16Bx2 tiling format + */ __u32 mode; + + /** + * @value: + * + * The value to set for the tiling mode parameter. + */ __u32 value; + + /** + * @pad: + * + * Structure padding that may be used in the future. Must be 0. + */ __u32 pad; }; +/** + * struct drm_tegra_gem_get_tiling - parameters for the get tiling IOCTL + */ struct drm_tegra_gem_get_tiling { - /* input */ + /** + * @handle: + * + * Handle to the GEM object for which to query the tiling parameters. + */ __u32 handle; - /* output */ + + /** + * @mode: + * + * The tiling mode currently associated with the GEM object. Set by + * the kernel upon successful completion of the IOCTL. + */ __u32 mode; + + /** + * @value: + * + * The tiling mode parameter currently associated with the GEM object. + * Set by the kernel upon successful completion of the IOCTL. + */ __u32 value; + + /** + * @pad: + * + * Structure padding that may be used in the future. Must be 0. + */ __u32 pad; }; #define DRM_TEGRA_GEM_BOTTOM_UP (1 << 0) #define DRM_TEGRA_GEM_FLAGS (DRM_TEGRA_GEM_BOTTOM_UP) +/** + * struct drm_tegra_gem_set_flags - parameters for the set flags IOCTL + */ struct drm_tegra_gem_set_flags { - /* input */ + /** + * @handle: + * + * Handle to the GEM object for which to set the flags. + */ __u32 handle; - /* output */ + + /** + * @flags: + * + * The flags to set for the GEM object. + */ __u32 flags; }; +/** + * struct drm_tegra_gem_get_flags - parameters for the get flags IOCTL + */ struct drm_tegra_gem_get_flags { - /* input */ + /** + * @handle: + * + * Handle to the GEM object for which to query the flags. + */ __u32 handle; - /* output */ + + /** + * @flags: + * + * The flags currently associated with the GEM object. Set by the + * kernel upon successful completion of the IOCTL. + */ __u32 flags; }; @@ -193,7 +665,7 @@ struct drm_tegra_gem_get_flags { #define DRM_IOCTL_TEGRA_SYNCPT_INCR DRM_IOWR(DRM_COMMAND_BASE + DRM_TEGRA_SYNCPT_INCR, struct drm_tegra_syncpt_incr) #define DRM_IOCTL_TEGRA_SYNCPT_WAIT DRM_IOWR(DRM_COMMAND_BASE + DRM_TEGRA_SYNCPT_WAIT, struct drm_tegra_syncpt_wait) #define DRM_IOCTL_TEGRA_OPEN_CHANNEL DRM_IOWR(DRM_COMMAND_BASE + DRM_TEGRA_OPEN_CHANNEL, struct drm_tegra_open_channel) -#define DRM_IOCTL_TEGRA_CLOSE_CHANNEL DRM_IOWR(DRM_COMMAND_BASE + DRM_TEGRA_CLOSE_CHANNEL, struct drm_tegra_open_channel) +#define DRM_IOCTL_TEGRA_CLOSE_CHANNEL DRM_IOWR(DRM_COMMAND_BASE + DRM_TEGRA_CLOSE_CHANNEL, struct drm_tegra_close_channel) #define DRM_IOCTL_TEGRA_GET_SYNCPT DRM_IOWR(DRM_COMMAND_BASE + DRM_TEGRA_GET_SYNCPT, struct drm_tegra_get_syncpt) #define DRM_IOCTL_TEGRA_SUBMIT DRM_IOWR(DRM_COMMAND_BASE + DRM_TEGRA_SUBMIT, struct drm_tegra_submit) #define DRM_IOCTL_TEGRA_GET_SYNCPT_BASE DRM_IOWR(DRM_COMMAND_BASE + DRM_TEGRA_GET_SYNCPT_BASE, struct drm_tegra_get_syncpt_base) diff --git a/include/uapi/linux/if_infiniband.h b/include/uapi/linux/if_infiniband.h index 050b92dcf8cf..0fc33bf30e45 100644 --- a/include/uapi/linux/if_infiniband.h +++ b/include/uapi/linux/if_infiniband.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR BSD-3-Clause) */ +/* SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR BSD-2-Clause) */ /* * This software is available to you under a choice of one of two * licenses. You may choose to be licensed under the terms of the GNU diff --git a/include/uapi/linux/nl80211.h b/include/uapi/linux/nl80211.h index 15daf5e2638d..9c3630146cec 100644 --- a/include/uapi/linux/nl80211.h +++ b/include/uapi/linux/nl80211.h @@ -2698,6 +2698,8 @@ enum nl80211_attrs { #define NL80211_ATTR_KEYS NL80211_ATTR_KEYS #define NL80211_ATTR_FEATURE_FLAGS NL80211_ATTR_FEATURE_FLAGS +#define NL80211_WIPHY_NAME_MAXLEN 128 + #define NL80211_MAX_SUPP_RATES 32 #define NL80211_MAX_SUPP_HT_RATES 77 #define NL80211_MAX_SUPP_REG_RULES 64 diff --git a/include/uapi/linux/rds.h b/include/uapi/linux/rds.h index a66b213de3d7..20c6bd0b0007 100644 --- a/include/uapi/linux/rds.h +++ b/include/uapi/linux/rds.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR BSD-2-Clause) */ +/* SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR Linux-OpenIB) */ /* * Copyright (c) 2008 Oracle. All rights reserved. * diff --git a/include/uapi/linux/tls.h b/include/uapi/linux/tls.h index c6633e97eca4..ff02287495ac 100644 --- a/include/uapi/linux/tls.h +++ b/include/uapi/linux/tls.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR BSD-2-Clause) */ +/* SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR Linux-OpenIB) */ /* * Copyright (c) 2016-2017, Mellanox Technologies. All rights reserved. * diff --git a/include/uapi/rdma/cxgb3-abi.h b/include/uapi/rdma/cxgb3-abi.h index 9acb4b7a6246..85aed672f43e 100644 --- a/include/uapi/rdma/cxgb3-abi.h +++ b/include/uapi/rdma/cxgb3-abi.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR BSD-2-Clause) */ +/* SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR Linux-OpenIB) */ /* * Copyright (c) 2006 Chelsio, Inc. All rights reserved. * diff --git a/include/uapi/rdma/cxgb4-abi.h b/include/uapi/rdma/cxgb4-abi.h index 1fefd0140c26..a159ba8dcf8f 100644 --- a/include/uapi/rdma/cxgb4-abi.h +++ b/include/uapi/rdma/cxgb4-abi.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR BSD-2-Clause) */ +/* SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR Linux-OpenIB) */ /* * Copyright (c) 2009-2010 Chelsio, Inc. All rights reserved. * diff --git a/include/uapi/rdma/hns-abi.h b/include/uapi/rdma/hns-abi.h index 7092c8de4bd8..78613b609fa8 100644 --- a/include/uapi/rdma/hns-abi.h +++ b/include/uapi/rdma/hns-abi.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR BSD-2-Clause) */ +/* SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR Linux-OpenIB) */ /* * Copyright (c) 2016 Hisilicon Limited. * diff --git a/include/uapi/rdma/ib_user_cm.h b/include/uapi/rdma/ib_user_cm.h index 4a8f9562f7cd..e2709bb8cb18 100644 --- a/include/uapi/rdma/ib_user_cm.h +++ b/include/uapi/rdma/ib_user_cm.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR BSD-2-Clause) */ +/* SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR Linux-OpenIB) */ /* * Copyright (c) 2005 Topspin Communications. All rights reserved. * Copyright (c) 2005 Intel Corporation. All rights reserved. diff --git a/include/uapi/rdma/ib_user_ioctl_verbs.h b/include/uapi/rdma/ib_user_ioctl_verbs.h index 04e46ea517d3..625545d862d7 100644 --- a/include/uapi/rdma/ib_user_ioctl_verbs.h +++ b/include/uapi/rdma/ib_user_ioctl_verbs.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR BSD-2-Clause) */ +/* SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR Linux-OpenIB) */ /* * Copyright (c) 2017-2018, Mellanox Technologies inc. All rights reserved. * diff --git a/include/uapi/rdma/ib_user_mad.h b/include/uapi/rdma/ib_user_mad.h index ef92118dad97..90c0cf228020 100644 --- a/include/uapi/rdma/ib_user_mad.h +++ b/include/uapi/rdma/ib_user_mad.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR BSD-2-Clause) */ +/* SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR Linux-OpenIB) */ /* * Copyright (c) 2004 Topspin Communications. All rights reserved. * Copyright (c) 2005 Voltaire, Inc. All rights reserved. diff --git a/include/uapi/rdma/ib_user_sa.h b/include/uapi/rdma/ib_user_sa.h index 0d2607f0cd20..435155d6e1c6 100644 --- a/include/uapi/rdma/ib_user_sa.h +++ b/include/uapi/rdma/ib_user_sa.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR BSD-2-Clause) */ +/* SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR Linux-OpenIB) */ /* * Copyright (c) 2005 Intel Corporation. All rights reserved. * diff --git a/include/uapi/rdma/ib_user_verbs.h b/include/uapi/rdma/ib_user_verbs.h index 9be07394fdbe..6aeb03315b0b 100644 --- a/include/uapi/rdma/ib_user_verbs.h +++ b/include/uapi/rdma/ib_user_verbs.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR BSD-2-Clause) */ +/* SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR Linux-OpenIB) */ /* * Copyright (c) 2005 Topspin Communications. All rights reserved. * Copyright (c) 2005, 2006 Cisco Systems. All rights reserved. diff --git a/include/uapi/rdma/mlx4-abi.h b/include/uapi/rdma/mlx4-abi.h index 04f64bc4045f..f74557528175 100644 --- a/include/uapi/rdma/mlx4-abi.h +++ b/include/uapi/rdma/mlx4-abi.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR BSD-2-Clause) */ +/* SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR Linux-OpenIB) */ /* * Copyright (c) 2007 Cisco Systems, Inc. All rights reserved. * Copyright (c) 2007, 2008 Mellanox Technologies. All rights reserved. diff --git a/include/uapi/rdma/mlx5-abi.h b/include/uapi/rdma/mlx5-abi.h index cb4a02c4a1ce..fdaf00e20649 100644 --- a/include/uapi/rdma/mlx5-abi.h +++ b/include/uapi/rdma/mlx5-abi.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR BSD-2-Clause) */ +/* SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR Linux-OpenIB) */ /* * Copyright (c) 2013-2015, Mellanox Technologies. All rights reserved. * diff --git a/include/uapi/rdma/mthca-abi.h b/include/uapi/rdma/mthca-abi.h index ac756cd9e807..91b12e1a6f43 100644 --- a/include/uapi/rdma/mthca-abi.h +++ b/include/uapi/rdma/mthca-abi.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR BSD-2-Clause) */ +/* SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR Linux-OpenIB) */ /* * Copyright (c) 2005 Topspin Communications. All rights reserved. * Copyright (c) 2005, 2006 Cisco Systems. All rights reserved. diff --git a/include/uapi/rdma/nes-abi.h b/include/uapi/rdma/nes-abi.h index 35bfd4015d07..f80495baa969 100644 --- a/include/uapi/rdma/nes-abi.h +++ b/include/uapi/rdma/nes-abi.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR BSD-2-Clause) */ +/* SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR Linux-OpenIB) */ /* * Copyright (c) 2006 - 2011 Intel Corporation. All rights reserved. * Copyright (c) 2005 Topspin Communications. All rights reserved. diff --git a/include/uapi/rdma/qedr-abi.h b/include/uapi/rdma/qedr-abi.h index 8ba098900e9a..24c658b3c790 100644 --- a/include/uapi/rdma/qedr-abi.h +++ b/include/uapi/rdma/qedr-abi.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR BSD-2-Clause) */ +/* SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR Linux-OpenIB) */ /* QLogic qedr NIC Driver * Copyright (c) 2015-2016 QLogic Corporation * diff --git a/include/uapi/rdma/rdma_user_cm.h b/include/uapi/rdma/rdma_user_cm.h index e1269024af47..0d1e78ebad05 100644 --- a/include/uapi/rdma/rdma_user_cm.h +++ b/include/uapi/rdma/rdma_user_cm.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR BSD-2-Clause) */ +/* SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR Linux-OpenIB) */ /* * Copyright (c) 2005-2006 Intel Corporation. All rights reserved. * diff --git a/include/uapi/rdma/rdma_user_ioctl.h b/include/uapi/rdma/rdma_user_ioctl.h index d223f4164a0f..d92d2721b28c 100644 --- a/include/uapi/rdma/rdma_user_ioctl.h +++ b/include/uapi/rdma/rdma_user_ioctl.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR BSD-2-Clause) */ +/* SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR Linux-OpenIB) */ /* * Copyright (c) 2016 Mellanox Technologies, LTD. All rights reserved. * diff --git a/include/uapi/rdma/rdma_user_rxe.h b/include/uapi/rdma/rdma_user_rxe.h index 1f8a9e7daea4..44ef6a3b7afc 100644 --- a/include/uapi/rdma/rdma_user_rxe.h +++ b/include/uapi/rdma/rdma_user_rxe.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR BSD-2-Clause) */ +/* SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR Linux-OpenIB) */ /* * Copyright (c) 2016 Mellanox Technologies Ltd. All rights reserved. * diff --git a/init/main.c b/init/main.c index b795aa341a3a..fd37315835b4 100644 --- a/init/main.c +++ b/init/main.c @@ -423,7 +423,7 @@ static noinline void __ref rest_init(void) /* * Enable might_sleep() and smp_processor_id() checks. - * They cannot be enabled earlier because with CONFIG_PRREMPT=y + * They cannot be enabled earlier because with CONFIG_PREEMPT=y * kernel_thread() would trigger might_sleep() splats. With * CONFIG_PREEMPT_VOLUNTARY=y the init task might have scheduled * already, but it's stuck on the kthreadd_done completion. @@ -1034,6 +1034,13 @@ __setup("rodata=", set_debug_rodata); static void mark_readonly(void) { if (rodata_enabled) { + /* + * load_module() results in W+X mappings, which are cleaned up + * with call_rcu_sched(). Let's make sure that queued work is + * flushed so that we don't hit false positives looking for + * insecure pages which are W+X. + */ + rcu_barrier_sched(); mark_rodata_ro(); rodata_test(); } else diff --git a/kernel/bpf/arraymap.c b/kernel/bpf/arraymap.c index 14750e7c5ee4..027107f4be53 100644 --- a/kernel/bpf/arraymap.c +++ b/kernel/bpf/arraymap.c @@ -476,7 +476,7 @@ static u32 prog_fd_array_sys_lookup_elem(void *ptr) } /* decrement refcnt of all bpf_progs that are stored in this map */ -void bpf_fd_array_map_clear(struct bpf_map *map) +static void bpf_fd_array_map_clear(struct bpf_map *map) { struct bpf_array *array = container_of(map, struct bpf_array, map); int i; @@ -495,6 +495,7 @@ const struct bpf_map_ops prog_array_map_ops = { .map_fd_get_ptr = prog_fd_array_get_ptr, .map_fd_put_ptr = prog_fd_array_put_ptr, .map_fd_sys_lookup_elem = prog_fd_array_sys_lookup_elem, + .map_release_uref = bpf_fd_array_map_clear, }; static struct bpf_event_entry *bpf_event_entry_gen(struct file *perf_file, diff --git a/kernel/bpf/sockmap.c b/kernel/bpf/sockmap.c index a3b21385e947..098eca568c2b 100644 --- a/kernel/bpf/sockmap.c +++ b/kernel/bpf/sockmap.c @@ -43,6 +43,7 @@ #include <net/tcp.h> #include <linux/ptr_ring.h> #include <net/inet_common.h> +#include <linux/sched/signal.h> #define SOCK_CREATE_FLAG_MASK \ (BPF_F_NUMA_NODE | BPF_F_RDONLY | BPF_F_WRONLY) @@ -325,6 +326,9 @@ retry: if (ret > 0) { if (apply) apply_bytes -= ret; + + sg->offset += ret; + sg->length -= ret; size -= ret; offset += ret; if (uncharge) @@ -332,8 +336,6 @@ retry: goto retry; } - sg->length = size; - sg->offset = offset; return ret; } @@ -391,7 +393,8 @@ static void return_mem_sg(struct sock *sk, int bytes, struct sk_msg_buff *md) } while (i != md->sg_end); } -static void free_bytes_sg(struct sock *sk, int bytes, struct sk_msg_buff *md) +static void free_bytes_sg(struct sock *sk, int bytes, + struct sk_msg_buff *md, bool charge) { struct scatterlist *sg = md->sg_data; int i = md->sg_start, free; @@ -401,11 +404,13 @@ static void free_bytes_sg(struct sock *sk, int bytes, struct sk_msg_buff *md) if (bytes < free) { sg[i].length -= bytes; sg[i].offset += bytes; - sk_mem_uncharge(sk, bytes); + if (charge) + sk_mem_uncharge(sk, bytes); break; } - sk_mem_uncharge(sk, sg[i].length); + if (charge) + sk_mem_uncharge(sk, sg[i].length); put_page(sg_page(&sg[i])); bytes -= sg[i].length; sg[i].length = 0; @@ -416,6 +421,7 @@ static void free_bytes_sg(struct sock *sk, int bytes, struct sk_msg_buff *md) if (i == MAX_SKB_FRAGS) i = 0; } + md->sg_start = i; } static int free_sg(struct sock *sk, int start, struct sk_msg_buff *md) @@ -523,8 +529,6 @@ static int bpf_tcp_ingress(struct sock *sk, int apply_bytes, i = md->sg_start; do { - r->sg_data[i] = md->sg_data[i]; - size = (apply && apply_bytes < md->sg_data[i].length) ? apply_bytes : md->sg_data[i].length; @@ -535,6 +539,7 @@ static int bpf_tcp_ingress(struct sock *sk, int apply_bytes, } sk_mem_charge(sk, size); + r->sg_data[i] = md->sg_data[i]; r->sg_data[i].length = size; md->sg_data[i].length -= size; md->sg_data[i].offset += size; @@ -575,10 +580,10 @@ static int bpf_tcp_sendmsg_do_redirect(struct sock *sk, int send, struct sk_msg_buff *md, int flags) { + bool ingress = !!(md->flags & BPF_F_INGRESS); struct smap_psock *psock; struct scatterlist *sg; - int i, err, free = 0; - bool ingress = !!(md->flags & BPF_F_INGRESS); + int err = 0; sg = md->sg_data; @@ -606,16 +611,8 @@ static int bpf_tcp_sendmsg_do_redirect(struct sock *sk, int send, out_rcu: rcu_read_unlock(); out: - i = md->sg_start; - while (sg[i].length) { - free += sg[i].length; - put_page(sg_page(&sg[i])); - sg[i].length = 0; - i++; - if (i == MAX_SKB_FRAGS) - i = 0; - } - return free; + free_bytes_sg(NULL, send, md, false); + return err; } static inline void bpf_md_init(struct smap_psock *psock) @@ -700,19 +697,26 @@ more_data: err = bpf_tcp_sendmsg_do_redirect(redir, send, m, flags); lock_sock(sk); + if (unlikely(err < 0)) { + free_start_sg(sk, m); + psock->sg_size = 0; + if (!cork) + *copied -= send; + } else { + psock->sg_size -= send; + } + if (cork) { free_start_sg(sk, m); + psock->sg_size = 0; kfree(m); m = NULL; + err = 0; } - if (unlikely(err)) - *copied -= err; - else - psock->sg_size -= send; break; case __SK_DROP: default: - free_bytes_sg(sk, send, m); + free_bytes_sg(sk, send, m, true); apply_bytes_dec(psock, send); *copied -= send; psock->sg_size -= send; @@ -732,6 +736,26 @@ out_err: return err; } +static int bpf_wait_data(struct sock *sk, + struct smap_psock *psk, int flags, + long timeo, int *err) +{ + int rc; + + DEFINE_WAIT_FUNC(wait, woken_wake_function); + + add_wait_queue(sk_sleep(sk), &wait); + sk_set_bit(SOCKWQ_ASYNC_WAITDATA, sk); + rc = sk_wait_event(sk, &timeo, + !list_empty(&psk->ingress) || + !skb_queue_empty(&sk->sk_receive_queue), + &wait); + sk_clear_bit(SOCKWQ_ASYNC_WAITDATA, sk); + remove_wait_queue(sk_sleep(sk), &wait); + + return rc; +} + static int bpf_tcp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, int nonblock, int flags, int *addr_len) { @@ -755,6 +779,7 @@ static int bpf_tcp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, return tcp_recvmsg(sk, msg, len, nonblock, flags, addr_len); lock_sock(sk); +bytes_ready: while (copied != len) { struct scatterlist *sg; struct sk_msg_buff *md; @@ -809,6 +834,28 @@ static int bpf_tcp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, } } + if (!copied) { + long timeo; + int data; + int err = 0; + + timeo = sock_rcvtimeo(sk, nonblock); + data = bpf_wait_data(sk, psock, flags, timeo, &err); + + if (data) { + if (!skb_queue_empty(&sk->sk_receive_queue)) { + release_sock(sk); + smap_release_sock(psock, sk); + copied = tcp_recvmsg(sk, msg, len, nonblock, flags, addr_len); + return copied; + } + goto bytes_ready; + } + + if (err) + copied = err; + } + release_sock(sk); smap_release_sock(psock, sk); return copied; @@ -1831,7 +1878,7 @@ static int sock_map_update_elem(struct bpf_map *map, return err; } -static void sock_map_release(struct bpf_map *map, struct file *map_file) +static void sock_map_release(struct bpf_map *map) { struct bpf_stab *stab = container_of(map, struct bpf_stab, map); struct bpf_prog *orig; @@ -1855,7 +1902,7 @@ const struct bpf_map_ops sock_map_ops = { .map_get_next_key = sock_map_get_next_key, .map_update_elem = sock_map_update_elem, .map_delete_elem = sock_map_delete_elem, - .map_release = sock_map_release, + .map_release_uref = sock_map_release, }; BPF_CALL_4(bpf_sock_map_update, struct bpf_sock_ops_kern *, bpf_sock, diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c index 4ca46df19c9a..016ef9025827 100644 --- a/kernel/bpf/syscall.c +++ b/kernel/bpf/syscall.c @@ -26,6 +26,7 @@ #include <linux/cred.h> #include <linux/timekeeping.h> #include <linux/ctype.h> +#include <linux/nospec.h> #define IS_FD_ARRAY(map) ((map)->map_type == BPF_MAP_TYPE_PROG_ARRAY || \ (map)->map_type == BPF_MAP_TYPE_PERF_EVENT_ARRAY || \ @@ -102,12 +103,14 @@ const struct bpf_map_ops bpf_map_offload_ops = { static struct bpf_map *find_and_alloc_map(union bpf_attr *attr) { const struct bpf_map_ops *ops; + u32 type = attr->map_type; struct bpf_map *map; int err; - if (attr->map_type >= ARRAY_SIZE(bpf_map_types)) + if (type >= ARRAY_SIZE(bpf_map_types)) return ERR_PTR(-EINVAL); - ops = bpf_map_types[attr->map_type]; + type = array_index_nospec(type, ARRAY_SIZE(bpf_map_types)); + ops = bpf_map_types[type]; if (!ops) return ERR_PTR(-EINVAL); @@ -122,7 +125,7 @@ static struct bpf_map *find_and_alloc_map(union bpf_attr *attr) if (IS_ERR(map)) return map; map->ops = ops; - map->map_type = attr->map_type; + map->map_type = type; return map; } @@ -257,8 +260,8 @@ static void bpf_map_free_deferred(struct work_struct *work) static void bpf_map_put_uref(struct bpf_map *map) { if (atomic_dec_and_test(&map->usercnt)) { - if (map->map_type == BPF_MAP_TYPE_PROG_ARRAY) - bpf_fd_array_map_clear(map); + if (map->ops->map_release_uref) + map->ops->map_release_uref(map); } } @@ -871,11 +874,17 @@ static const struct bpf_prog_ops * const bpf_prog_types[] = { static int find_prog_type(enum bpf_prog_type type, struct bpf_prog *prog) { - if (type >= ARRAY_SIZE(bpf_prog_types) || !bpf_prog_types[type]) + const struct bpf_prog_ops *ops; + + if (type >= ARRAY_SIZE(bpf_prog_types)) + return -EINVAL; + type = array_index_nospec(type, ARRAY_SIZE(bpf_prog_types)); + ops = bpf_prog_types[type]; + if (!ops) return -EINVAL; if (!bpf_prog_is_dev_bound(prog->aux)) - prog->aux->ops = bpf_prog_types[type]; + prog->aux->ops = ops; else prog->aux->ops = &bpf_offload_prog_ops; prog->type = type; diff --git a/kernel/compat.c b/kernel/compat.c index 6d21894806b4..92d8c98c0f57 100644 --- a/kernel/compat.c +++ b/kernel/compat.c @@ -34,6 +34,7 @@ int compat_get_timex(struct timex *txc, const struct compat_timex __user *utp) { struct compat_timex tx32; + memset(txc, 0, sizeof(struct timex)); if (copy_from_user(&tx32, utp, sizeof(struct compat_timex))) return -EFAULT; diff --git a/kernel/events/ring_buffer.c b/kernel/events/ring_buffer.c index 6c6b3c48db71..1d8ca9ea9979 100644 --- a/kernel/events/ring_buffer.c +++ b/kernel/events/ring_buffer.c @@ -14,6 +14,7 @@ #include <linux/slab.h> #include <linux/circ_buf.h> #include <linux/poll.h> +#include <linux/nospec.h> #include "internal.h" @@ -867,8 +868,10 @@ perf_mmap_to_page(struct ring_buffer *rb, unsigned long pgoff) return NULL; /* AUX space */ - if (pgoff >= rb->aux_pgoff) - return virt_to_page(rb->aux_pages[pgoff - rb->aux_pgoff]); + if (pgoff >= rb->aux_pgoff) { + int aux_pgoff = array_index_nospec(pgoff - rb->aux_pgoff, rb->aux_nr_pages); + return virt_to_page(rb->aux_pages[aux_pgoff]); + } } return __perf_mmap_to_page(rb, pgoff); diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c index ce6848e46e94..1725b902983f 100644 --- a/kernel/events/uprobes.c +++ b/kernel/events/uprobes.c @@ -491,7 +491,7 @@ static struct uprobe *alloc_uprobe(struct inode *inode, loff_t offset) if (!uprobe) return NULL; - uprobe->inode = igrab(inode); + uprobe->inode = inode; uprobe->offset = offset; init_rwsem(&uprobe->register_rwsem); init_rwsem(&uprobe->consumer_rwsem); @@ -502,7 +502,6 @@ static struct uprobe *alloc_uprobe(struct inode *inode, loff_t offset) if (cur_uprobe) { kfree(uprobe); uprobe = cur_uprobe; - iput(inode); } return uprobe; @@ -701,7 +700,6 @@ static void delete_uprobe(struct uprobe *uprobe) rb_erase(&uprobe->rb_node, &uprobes_tree); spin_unlock(&uprobes_treelock); RB_CLEAR_NODE(&uprobe->rb_node); /* for uprobe_is_active() */ - iput(uprobe->inode); put_uprobe(uprobe); } @@ -873,7 +871,8 @@ static void __uprobe_unregister(struct uprobe *uprobe, struct uprobe_consumer *u * tuple). Creation refcount stops uprobe_unregister from freeing the * @uprobe even before the register operation is complete. Creation * refcount is released when the last @uc for the @uprobe - * unregisters. + * unregisters. Caller of uprobe_register() is required to keep @inode + * (and the containing mount) referenced. * * Return errno if it cannot successully install probes * else return 0 (success) diff --git a/kernel/kthread.c b/kernel/kthread.c index cd50e99202b0..2017a39ab490 100644 --- a/kernel/kthread.c +++ b/kernel/kthread.c @@ -55,7 +55,6 @@ enum KTHREAD_BITS { KTHREAD_IS_PER_CPU = 0, KTHREAD_SHOULD_STOP, KTHREAD_SHOULD_PARK, - KTHREAD_IS_PARKED, }; static inline void set_kthread_struct(void *kthread) @@ -177,14 +176,12 @@ void *kthread_probe_data(struct task_struct *task) static void __kthread_parkme(struct kthread *self) { - __set_current_state(TASK_PARKED); - while (test_bit(KTHREAD_SHOULD_PARK, &self->flags)) { - if (!test_and_set_bit(KTHREAD_IS_PARKED, &self->flags)) - complete(&self->parked); + for (;;) { + set_current_state(TASK_PARKED); + if (!test_bit(KTHREAD_SHOULD_PARK, &self->flags)) + break; schedule(); - __set_current_state(TASK_PARKED); } - clear_bit(KTHREAD_IS_PARKED, &self->flags); __set_current_state(TASK_RUNNING); } @@ -194,6 +191,11 @@ void kthread_parkme(void) } EXPORT_SYMBOL_GPL(kthread_parkme); +void kthread_park_complete(struct task_struct *k) +{ + complete(&to_kthread(k)->parked); +} + static int kthread(void *_create) { /* Copy data: it's on kthread's stack */ @@ -450,22 +452,15 @@ void kthread_unpark(struct task_struct *k) { struct kthread *kthread = to_kthread(k); - clear_bit(KTHREAD_SHOULD_PARK, &kthread->flags); /* - * We clear the IS_PARKED bit here as we don't wait - * until the task has left the park code. So if we'd - * park before that happens we'd see the IS_PARKED bit - * which might be about to be cleared. + * Newly created kthread was parked when the CPU was offline. + * The binding was lost and we need to set it again. */ - if (test_and_clear_bit(KTHREAD_IS_PARKED, &kthread->flags)) { - /* - * Newly created kthread was parked when the CPU was offline. - * The binding was lost and we need to set it again. - */ - if (test_bit(KTHREAD_IS_PER_CPU, &kthread->flags)) - __kthread_bind(k, kthread->cpu, TASK_PARKED); - wake_up_state(k, TASK_PARKED); - } + if (test_bit(KTHREAD_IS_PER_CPU, &kthread->flags)) + __kthread_bind(k, kthread->cpu, TASK_PARKED); + + clear_bit(KTHREAD_SHOULD_PARK, &kthread->flags); + wake_up_state(k, TASK_PARKED); } EXPORT_SYMBOL_GPL(kthread_unpark); @@ -488,12 +483,13 @@ int kthread_park(struct task_struct *k) if (WARN_ON(k->flags & PF_EXITING)) return -ENOSYS; - if (!test_bit(KTHREAD_IS_PARKED, &kthread->flags)) { - set_bit(KTHREAD_SHOULD_PARK, &kthread->flags); - if (k != current) { - wake_up_process(k); - wait_for_completion(&kthread->parked); - } + if (WARN_ON_ONCE(test_bit(KTHREAD_SHOULD_PARK, &kthread->flags))) + return -EBUSY; + + set_bit(KTHREAD_SHOULD_PARK, &kthread->flags); + if (k != current) { + wake_up_process(k); + wait_for_completion(&kthread->parked); } return 0; diff --git a/kernel/module.c b/kernel/module.c index ce8066b88178..c9bea7f2b43e 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -3517,6 +3517,11 @@ static noinline int do_init_module(struct module *mod) * walking this with preempt disabled. In all the failure paths, we * call synchronize_sched(), but we don't want to slow down the success * path, so use actual RCU here. + * Note that module_alloc() on most architectures creates W+X page + * mappings which won't be cleaned up until do_free_init() runs. Any + * code such as mark_rodata_ro() which depends on those mappings to + * be cleaned up needs to sync with the queued work - ie + * rcu_barrier_sched() */ call_rcu_sched(&freeinit->rcu, do_free_init); mutex_unlock(&module_mutex); diff --git a/kernel/sched/autogroup.c b/kernel/sched/autogroup.c index 6be6c575b6cd..2d4ff5353ded 100644 --- a/kernel/sched/autogroup.c +++ b/kernel/sched/autogroup.c @@ -2,6 +2,7 @@ /* * Auto-group scheduling implementation: */ +#include <linux/nospec.h> #include "sched.h" unsigned int __read_mostly sysctl_sched_autogroup_enabled = 1; @@ -209,7 +210,7 @@ int proc_sched_autogroup_set_nice(struct task_struct *p, int nice) static unsigned long next = INITIAL_JIFFIES; struct autogroup *ag; unsigned long shares; - int err; + int err, idx; if (nice < MIN_NICE || nice > MAX_NICE) return -EINVAL; @@ -227,7 +228,9 @@ int proc_sched_autogroup_set_nice(struct task_struct *p, int nice) next = HZ / 10 + jiffies; ag = autogroup_task_get(p); - shares = scale_load(sched_prio_to_weight[nice + 20]); + + idx = array_index_nospec(nice + 20, 40); + shares = scale_load(sched_prio_to_weight[idx]); down_write(&ag->lock); err = sched_group_set_shares(ag->tg, shares); diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 5e10aaeebfcc..092f7c4de903 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -7,6 +7,9 @@ */ #include "sched.h" +#include <linux/kthread.h> +#include <linux/nospec.h> + #include <asm/switch_to.h> #include <asm/tlb.h> @@ -2718,20 +2721,28 @@ static struct rq *finish_task_switch(struct task_struct *prev) membarrier_mm_sync_core_before_usermode(mm); mmdrop(mm); } - if (unlikely(prev_state == TASK_DEAD)) { - if (prev->sched_class->task_dead) - prev->sched_class->task_dead(prev); + if (unlikely(prev_state & (TASK_DEAD|TASK_PARKED))) { + switch (prev_state) { + case TASK_DEAD: + if (prev->sched_class->task_dead) + prev->sched_class->task_dead(prev); - /* - * Remove function-return probe instances associated with this - * task and put them back on the free list. - */ - kprobe_flush_task(prev); + /* + * Remove function-return probe instances associated with this + * task and put them back on the free list. + */ + kprobe_flush_task(prev); + + /* Task is done with its stack. */ + put_task_stack(prev); - /* Task is done with its stack. */ - put_task_stack(prev); + put_task_struct(prev); + break; - put_task_struct(prev); + case TASK_PARKED: + kthread_park_complete(prev); + break; + } } tick_nohz_task_switch(); @@ -3498,23 +3509,8 @@ static void __sched notrace __schedule(bool preempt) void __noreturn do_task_dead(void) { - /* - * The setting of TASK_RUNNING by try_to_wake_up() may be delayed - * when the following two conditions become true. - * - There is race condition of mmap_sem (It is acquired by - * exit_mm()), and - * - SMI occurs before setting TASK_RUNINNG. - * (or hypervisor of virtual machine switches to other guest) - * As a result, we may become TASK_RUNNING after becoming TASK_DEAD - * - * To avoid it, we have to wait for releasing tsk->pi_lock which - * is held by try_to_wake_up() - */ - raw_spin_lock_irq(¤t->pi_lock); - raw_spin_unlock_irq(¤t->pi_lock); - /* Causes final put_task_struct in finish_task_switch(): */ - __set_current_state(TASK_DEAD); + set_special_state(TASK_DEAD); /* Tell freezer to ignore us: */ current->flags |= PF_NOFREEZE; @@ -6928,11 +6924,15 @@ static int cpu_weight_nice_write_s64(struct cgroup_subsys_state *css, struct cftype *cft, s64 nice) { unsigned long weight; + int idx; if (nice < MIN_NICE || nice > MAX_NICE) return -ERANGE; - weight = sched_prio_to_weight[NICE_TO_PRIO(nice) - MAX_RT_PRIO]; + idx = NICE_TO_PRIO(nice) - MAX_RT_PRIO; + idx = array_index_nospec(idx, 40); + weight = sched_prio_to_weight[idx]; + return sched_group_set_shares(css_tg(css), scale_load(weight)); } #endif diff --git a/kernel/sched/cpufreq_schedutil.c b/kernel/sched/cpufreq_schedutil.c index d2c6083304b4..e13df951aca7 100644 --- a/kernel/sched/cpufreq_schedutil.c +++ b/kernel/sched/cpufreq_schedutil.c @@ -305,7 +305,8 @@ static void sugov_update_single(struct update_util_data *hook, u64 time, * Do not reduce the frequency if the CPU has not been idle * recently, as the reduction is likely to be premature then. */ - if (busy && next_f < sg_policy->next_freq) { + if (busy && next_f < sg_policy->next_freq && + sg_policy->next_freq != UINT_MAX) { next_f = sg_policy->next_freq; /* Reset cached freq as next_freq has changed */ @@ -396,19 +397,6 @@ static void sugov_irq_work(struct irq_work *irq_work) sg_policy = container_of(irq_work, struct sugov_policy, irq_work); - /* - * For RT tasks, the schedutil governor shoots the frequency to maximum. - * Special care must be taken to ensure that this kthread doesn't result - * in the same behavior. - * - * This is (mostly) guaranteed by the work_in_progress flag. The flag is - * updated only at the end of the sugov_work() function and before that - * the schedutil governor rejects all other frequency scaling requests. - * - * There is a very rare case though, where the RT thread yields right - * after the work_in_progress flag is cleared. The effects of that are - * neglected for now. - */ kthread_queue_work(&sg_policy->worker, &sg_policy->work); } diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 54dc31e7ab9b..79f574dba096 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -1854,7 +1854,6 @@ static int task_numa_migrate(struct task_struct *p) static void numa_migrate_preferred(struct task_struct *p) { unsigned long interval = HZ; - unsigned long numa_migrate_retry; /* This task has no NUMA fault statistics yet */ if (unlikely(p->numa_preferred_nid == -1 || !p->numa_faults)) @@ -1862,18 +1861,7 @@ static void numa_migrate_preferred(struct task_struct *p) /* Periodically retry migrating the task to the preferred node */ interval = min(interval, msecs_to_jiffies(p->numa_scan_period) / 16); - numa_migrate_retry = jiffies + interval; - - /* - * Check that the new retry threshold is after the current one. If - * the retry is in the future, it implies that wake_affine has - * temporarily asked NUMA balancing to backoff from placement. - */ - if (numa_migrate_retry > p->numa_migrate_retry) - return; - - /* Safe to try placing the task on the preferred node */ - p->numa_migrate_retry = numa_migrate_retry; + p->numa_migrate_retry = jiffies + interval; /* Success if task is already running on preferred CPU */ if (task_node(p) == p->numa_preferred_nid) @@ -5922,48 +5910,6 @@ wake_affine_weight(struct sched_domain *sd, struct task_struct *p, return this_eff_load < prev_eff_load ? this_cpu : nr_cpumask_bits; } -#ifdef CONFIG_NUMA_BALANCING -static void -update_wa_numa_placement(struct task_struct *p, int prev_cpu, int target) -{ - unsigned long interval; - - if (!static_branch_likely(&sched_numa_balancing)) - return; - - /* If balancing has no preference then continue gathering data */ - if (p->numa_preferred_nid == -1) - return; - - /* - * If the wakeup is not affecting locality then it is neutral from - * the perspective of NUMA balacing so continue gathering data. - */ - if (cpu_to_node(prev_cpu) == cpu_to_node(target)) - return; - - /* - * Temporarily prevent NUMA balancing trying to place waker/wakee after - * wakee has been moved by wake_affine. This will potentially allow - * related tasks to converge and update their data placement. The - * 4 * numa_scan_period is to allow the two-pass filter to migrate - * hot data to the wakers node. - */ - interval = max(sysctl_numa_balancing_scan_delay, - p->numa_scan_period << 2); - p->numa_migrate_retry = jiffies + msecs_to_jiffies(interval); - - interval = max(sysctl_numa_balancing_scan_delay, - current->numa_scan_period << 2); - current->numa_migrate_retry = jiffies + msecs_to_jiffies(interval); -} -#else -static void -update_wa_numa_placement(struct task_struct *p, int prev_cpu, int target) -{ -} -#endif - static int wake_affine(struct sched_domain *sd, struct task_struct *p, int this_cpu, int prev_cpu, int sync) { @@ -5979,7 +5925,6 @@ static int wake_affine(struct sched_domain *sd, struct task_struct *p, if (target == nr_cpumask_bits) return prev_cpu; - update_wa_numa_placement(p, prev_cpu, target); schedstat_inc(sd->ttwu_move_affine); schedstat_inc(p->se.statistics.nr_wakeups_affine); return target; @@ -9847,6 +9792,7 @@ static int idle_balance(struct rq *this_rq, struct rq_flags *rf) if (curr_cost > this_rq->max_idle_balance_cost) this_rq->max_idle_balance_cost = curr_cost; +out: /* * While browsing the domains, we released the rq lock, a task could * have been enqueued in the meantime. Since we're not going idle, @@ -9855,7 +9801,6 @@ static int idle_balance(struct rq *this_rq, struct rq_flags *rf) if (this_rq->cfs.h_nr_running && !pulled_task) pulled_task = 1; -out: /* Move the next balance forward */ if (time_after(this_rq->next_balance, next_balance)) this_rq->next_balance = next_balance; diff --git a/kernel/signal.c b/kernel/signal.c index d4ccea599692..9c33163a6165 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -1961,14 +1961,27 @@ static void ptrace_stop(int exit_code, int why, int clear_code, siginfo_t *info) return; } + set_special_state(TASK_TRACED); + /* * We're committing to trapping. TRACED should be visible before * TRAPPING is cleared; otherwise, the tracer might fail do_wait(). * Also, transition to TRACED and updates to ->jobctl should be * atomic with respect to siglock and should be done after the arch * hook as siglock is released and regrabbed across it. + * + * TRACER TRACEE + * + * ptrace_attach() + * [L] wait_on_bit(JOBCTL_TRAPPING) [S] set_special_state(TRACED) + * do_wait() + * set_current_state() smp_wmb(); + * ptrace_do_wait() + * wait_task_stopped() + * task_stopped_code() + * [L] task_is_traced() [S] task_clear_jobctl_trapping(); */ - set_current_state(TASK_TRACED); + smp_wmb(); current->last_siginfo = info; current->exit_code = exit_code; @@ -2176,7 +2189,7 @@ static bool do_signal_stop(int signr) if (task_participate_group_stop(current)) notify = CLD_STOPPED; - __set_current_state(TASK_STOPPED); + set_special_state(TASK_STOPPED); spin_unlock_irq(¤t->sighand->siglock); /* diff --git a/kernel/stop_machine.c b/kernel/stop_machine.c index b7591261652d..64c0291b579c 100644 --- a/kernel/stop_machine.c +++ b/kernel/stop_machine.c @@ -21,6 +21,7 @@ #include <linux/smpboot.h> #include <linux/atomic.h> #include <linux/nmi.h> +#include <linux/sched/wake_q.h> /* * Structure to determine completion condition and record errors. May @@ -65,27 +66,31 @@ static void cpu_stop_signal_done(struct cpu_stop_done *done) } static void __cpu_stop_queue_work(struct cpu_stopper *stopper, - struct cpu_stop_work *work) + struct cpu_stop_work *work, + struct wake_q_head *wakeq) { list_add_tail(&work->list, &stopper->works); - wake_up_process(stopper->thread); + wake_q_add(wakeq, stopper->thread); } /* queue @work to @stopper. if offline, @work is completed immediately */ static bool cpu_stop_queue_work(unsigned int cpu, struct cpu_stop_work *work) { struct cpu_stopper *stopper = &per_cpu(cpu_stopper, cpu); + DEFINE_WAKE_Q(wakeq); unsigned long flags; bool enabled; spin_lock_irqsave(&stopper->lock, flags); enabled = stopper->enabled; if (enabled) - __cpu_stop_queue_work(stopper, work); + __cpu_stop_queue_work(stopper, work, &wakeq); else if (work->done) cpu_stop_signal_done(work->done); spin_unlock_irqrestore(&stopper->lock, flags); + wake_up_q(&wakeq); + return enabled; } @@ -229,6 +234,7 @@ static int cpu_stop_queue_two_works(int cpu1, struct cpu_stop_work *work1, { struct cpu_stopper *stopper1 = per_cpu_ptr(&cpu_stopper, cpu1); struct cpu_stopper *stopper2 = per_cpu_ptr(&cpu_stopper, cpu2); + DEFINE_WAKE_Q(wakeq); int err; retry: spin_lock_irq(&stopper1->lock); @@ -252,8 +258,8 @@ retry: goto unlock; err = 0; - __cpu_stop_queue_work(stopper1, work1); - __cpu_stop_queue_work(stopper2, work2); + __cpu_stop_queue_work(stopper1, work1, &wakeq); + __cpu_stop_queue_work(stopper2, work2, &wakeq); unlock: spin_unlock(&stopper2->lock); spin_unlock_irq(&stopper1->lock); @@ -263,6 +269,9 @@ unlock: cpu_relax(); goto retry; } + + wake_up_q(&wakeq); + return err; } /** diff --git a/kernel/time/clocksource.c b/kernel/time/clocksource.c index 0e974cface0b..84f37420fcf5 100644 --- a/kernel/time/clocksource.c +++ b/kernel/time/clocksource.c @@ -119,6 +119,16 @@ static DEFINE_SPINLOCK(watchdog_lock); static int watchdog_running; static atomic_t watchdog_reset_pending; +static void inline clocksource_watchdog_lock(unsigned long *flags) +{ + spin_lock_irqsave(&watchdog_lock, *flags); +} + +static void inline clocksource_watchdog_unlock(unsigned long *flags) +{ + spin_unlock_irqrestore(&watchdog_lock, *flags); +} + static int clocksource_watchdog_kthread(void *data); static void __clocksource_change_rating(struct clocksource *cs, int rating); @@ -142,9 +152,19 @@ static void __clocksource_unstable(struct clocksource *cs) cs->flags &= ~(CLOCK_SOURCE_VALID_FOR_HRES | CLOCK_SOURCE_WATCHDOG); cs->flags |= CLOCK_SOURCE_UNSTABLE; + /* + * If the clocksource is registered clocksource_watchdog_kthread() will + * re-rate and re-select. + */ + if (list_empty(&cs->list)) { + cs->rating = 0; + return; + } + if (cs->mark_unstable) cs->mark_unstable(cs); + /* kick clocksource_watchdog_kthread() */ if (finished_booting) schedule_work(&watchdog_work); } @@ -153,10 +173,8 @@ static void __clocksource_unstable(struct clocksource *cs) * clocksource_mark_unstable - mark clocksource unstable via watchdog * @cs: clocksource to be marked unstable * - * This function is called instead of clocksource_change_rating from - * cpu hotplug code to avoid a deadlock between the clocksource mutex - * and the cpu hotplug mutex. It defers the update of the clocksource - * to the watchdog thread. + * This function is called by the x86 TSC code to mark clocksources as unstable; + * it defers demotion and re-selection to a kthread. */ void clocksource_mark_unstable(struct clocksource *cs) { @@ -164,7 +182,7 @@ void clocksource_mark_unstable(struct clocksource *cs) spin_lock_irqsave(&watchdog_lock, flags); if (!(cs->flags & CLOCK_SOURCE_UNSTABLE)) { - if (list_empty(&cs->wd_list)) + if (!list_empty(&cs->list) && list_empty(&cs->wd_list)) list_add(&cs->wd_list, &watchdog_list); __clocksource_unstable(cs); } @@ -319,9 +337,8 @@ static void clocksource_resume_watchdog(void) static void clocksource_enqueue_watchdog(struct clocksource *cs) { - unsigned long flags; + INIT_LIST_HEAD(&cs->wd_list); - spin_lock_irqsave(&watchdog_lock, flags); if (cs->flags & CLOCK_SOURCE_MUST_VERIFY) { /* cs is a clocksource to be watched. */ list_add(&cs->wd_list, &watchdog_list); @@ -331,7 +348,6 @@ static void clocksource_enqueue_watchdog(struct clocksource *cs) if (cs->flags & CLOCK_SOURCE_IS_CONTINUOUS) cs->flags |= CLOCK_SOURCE_VALID_FOR_HRES; } - spin_unlock_irqrestore(&watchdog_lock, flags); } static void clocksource_select_watchdog(bool fallback) @@ -373,9 +389,6 @@ static void clocksource_select_watchdog(bool fallback) static void clocksource_dequeue_watchdog(struct clocksource *cs) { - unsigned long flags; - - spin_lock_irqsave(&watchdog_lock, flags); if (cs != watchdog) { if (cs->flags & CLOCK_SOURCE_MUST_VERIFY) { /* cs is a watched clocksource. */ @@ -384,21 +397,19 @@ static void clocksource_dequeue_watchdog(struct clocksource *cs) clocksource_stop_watchdog(); } } - spin_unlock_irqrestore(&watchdog_lock, flags); } static int __clocksource_watchdog_kthread(void) { struct clocksource *cs, *tmp; unsigned long flags; - LIST_HEAD(unstable); int select = 0; spin_lock_irqsave(&watchdog_lock, flags); list_for_each_entry_safe(cs, tmp, &watchdog_list, wd_list) { if (cs->flags & CLOCK_SOURCE_UNSTABLE) { list_del_init(&cs->wd_list); - list_add(&cs->wd_list, &unstable); + __clocksource_change_rating(cs, 0); select = 1; } if (cs->flags & CLOCK_SOURCE_RESELECT) { @@ -410,11 +421,6 @@ static int __clocksource_watchdog_kthread(void) clocksource_stop_watchdog(); spin_unlock_irqrestore(&watchdog_lock, flags); - /* Needs to be done outside of watchdog lock */ - list_for_each_entry_safe(cs, tmp, &unstable, wd_list) { - list_del_init(&cs->wd_list); - __clocksource_change_rating(cs, 0); - } return select; } @@ -447,6 +453,9 @@ static inline int __clocksource_watchdog_kthread(void) { return 0; } static bool clocksource_is_watchdog(struct clocksource *cs) { return false; } void clocksource_mark_unstable(struct clocksource *cs) { } +static void inline clocksource_watchdog_lock(unsigned long *flags) { } +static void inline clocksource_watchdog_unlock(unsigned long *flags) { } + #endif /* CONFIG_CLOCKSOURCE_WATCHDOG */ /** @@ -779,14 +788,19 @@ EXPORT_SYMBOL_GPL(__clocksource_update_freq_scale); */ int __clocksource_register_scale(struct clocksource *cs, u32 scale, u32 freq) { + unsigned long flags; /* Initialize mult/shift and max_idle_ns */ __clocksource_update_freq_scale(cs, scale, freq); /* Add clocksource to the clocksource list */ mutex_lock(&clocksource_mutex); + + clocksource_watchdog_lock(&flags); clocksource_enqueue(cs); clocksource_enqueue_watchdog(cs); + clocksource_watchdog_unlock(&flags); + clocksource_select(); clocksource_select_watchdog(false); mutex_unlock(&clocksource_mutex); @@ -808,8 +822,13 @@ static void __clocksource_change_rating(struct clocksource *cs, int rating) */ void clocksource_change_rating(struct clocksource *cs, int rating) { + unsigned long flags; + mutex_lock(&clocksource_mutex); + clocksource_watchdog_lock(&flags); __clocksource_change_rating(cs, rating); + clocksource_watchdog_unlock(&flags); + clocksource_select(); clocksource_select_watchdog(false); mutex_unlock(&clocksource_mutex); @@ -821,6 +840,8 @@ EXPORT_SYMBOL(clocksource_change_rating); */ static int clocksource_unbind(struct clocksource *cs) { + unsigned long flags; + if (clocksource_is_watchdog(cs)) { /* Select and try to install a replacement watchdog. */ clocksource_select_watchdog(true); @@ -834,8 +855,12 @@ static int clocksource_unbind(struct clocksource *cs) if (curr_clocksource == cs) return -EBUSY; } + + clocksource_watchdog_lock(&flags); clocksource_dequeue_watchdog(cs); list_del_init(&cs->list); + clocksource_watchdog_unlock(&flags); + return 0; } diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 16bbf062018f..8d83bcf9ef69 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -5514,10 +5514,10 @@ static __init int ftrace_init_dyn_tracefs(struct dentry *d_tracer) ftrace_create_filter_files(&global_ops, d_tracer); #ifdef CONFIG_FUNCTION_GRAPH_TRACER - trace_create_file("set_graph_function", 0444, d_tracer, + trace_create_file("set_graph_function", 0644, d_tracer, NULL, &ftrace_graph_fops); - trace_create_file("set_graph_notrace", 0444, d_tracer, + trace_create_file("set_graph_notrace", 0644, d_tracer, NULL, &ftrace_graph_notrace_fops); #endif /* CONFIG_FUNCTION_GRAPH_TRACER */ diff --git a/kernel/trace/trace_events_filter.c b/kernel/trace/trace_events_filter.c index 1f951b3df60c..7d306b74230f 100644 --- a/kernel/trace/trace_events_filter.c +++ b/kernel/trace/trace_events_filter.c @@ -762,6 +762,9 @@ static int regex_match_full(char *str, struct regex *r, int len) static int regex_match_front(char *str, struct regex *r, int len) { + if (len < r->len) + return 0; + if (strncmp(str, r->pattern, r->len) == 0) return 1; return 0; diff --git a/kernel/trace/trace_events_hist.c b/kernel/trace/trace_events_hist.c index 0d7b3ffbecc2..b9061ed59bbd 100644 --- a/kernel/trace/trace_events_hist.c +++ b/kernel/trace/trace_events_hist.c @@ -2466,6 +2466,7 @@ parse_field(struct hist_trigger_data *hist_data, struct trace_event_file *file, else if (strcmp(modifier, "usecs") == 0) *flags |= HIST_FIELD_FL_TIMESTAMP_USECS; else { + hist_err("Invalid field modifier: ", modifier); field = ERR_PTR(-EINVAL); goto out; } @@ -2481,6 +2482,7 @@ parse_field(struct hist_trigger_data *hist_data, struct trace_event_file *file, else { field = trace_find_event_field(file->event_call, field_name); if (!field || !field->size) { + hist_err("Couldn't find field: ", field_name); field = ERR_PTR(-EINVAL); goto out; } @@ -4913,6 +4915,16 @@ static void hist_field_print(struct seq_file *m, struct hist_field *hist_field) seq_printf(m, "%s", field_name); } else if (hist_field->flags & HIST_FIELD_FL_TIMESTAMP) seq_puts(m, "common_timestamp"); + + if (hist_field->flags) { + if (!(hist_field->flags & HIST_FIELD_FL_VAR_REF) && + !(hist_field->flags & HIST_FIELD_FL_EXPR)) { + const char *flags = get_hist_field_flags(hist_field); + + if (flags) + seq_printf(m, ".%s", flags); + } + } } static int event_hist_trigger_print(struct seq_file *m, diff --git a/kernel/trace/trace_stack.c b/kernel/trace/trace_stack.c index 3c7bfc4bf5e9..4237eba4ef20 100644 --- a/kernel/trace/trace_stack.c +++ b/kernel/trace/trace_stack.c @@ -472,7 +472,7 @@ static __init int stack_trace_init(void) NULL, &stack_trace_fops); #ifdef CONFIG_DYNAMIC_FTRACE - trace_create_file("stack_trace_filter", 0444, d_tracer, + trace_create_file("stack_trace_filter", 0644, d_tracer, &trace_ops, &stack_trace_filter_fops); #endif diff --git a/kernel/trace/trace_uprobe.c b/kernel/trace/trace_uprobe.c index 34fd0e0ec51d..ac892878dbe6 100644 --- a/kernel/trace/trace_uprobe.c +++ b/kernel/trace/trace_uprobe.c @@ -55,6 +55,7 @@ struct trace_uprobe { struct list_head list; struct trace_uprobe_filter filter; struct uprobe_consumer consumer; + struct path path; struct inode *inode; char *filename; unsigned long offset; @@ -289,7 +290,7 @@ static void free_trace_uprobe(struct trace_uprobe *tu) for (i = 0; i < tu->tp.nr_args; i++) traceprobe_free_probe_arg(&tu->tp.args[i]); - iput(tu->inode); + path_put(&tu->path); kfree(tu->tp.call.class->system); kfree(tu->tp.call.name); kfree(tu->filename); @@ -363,7 +364,6 @@ end: static int create_trace_uprobe(int argc, char **argv) { struct trace_uprobe *tu; - struct inode *inode; char *arg, *event, *group, *filename; char buf[MAX_EVENT_NAME_LEN]; struct path path; @@ -371,7 +371,6 @@ static int create_trace_uprobe(int argc, char **argv) bool is_delete, is_return; int i, ret; - inode = NULL; ret = 0; is_delete = false; is_return = false; @@ -437,21 +436,16 @@ static int create_trace_uprobe(int argc, char **argv) } /* Find the last occurrence, in case the path contains ':' too. */ arg = strrchr(argv[1], ':'); - if (!arg) { - ret = -EINVAL; - goto fail_address_parse; - } + if (!arg) + return -EINVAL; *arg++ = '\0'; filename = argv[1]; ret = kern_path(filename, LOOKUP_FOLLOW, &path); if (ret) - goto fail_address_parse; - - inode = igrab(d_real_inode(path.dentry)); - path_put(&path); + return ret; - if (!inode || !S_ISREG(inode->i_mode)) { + if (!d_is_reg(path.dentry)) { ret = -EINVAL; goto fail_address_parse; } @@ -490,7 +484,7 @@ static int create_trace_uprobe(int argc, char **argv) goto fail_address_parse; } tu->offset = offset; - tu->inode = inode; + tu->path = path; tu->filename = kstrdup(filename, GFP_KERNEL); if (!tu->filename) { @@ -558,7 +552,7 @@ error: return ret; fail_address_parse: - iput(inode); + path_put(&path); pr_info("Failed to parse address or file.\n"); @@ -922,6 +916,7 @@ probe_event_enable(struct trace_uprobe *tu, struct trace_event_file *file, goto err_flags; tu->consumer.filter = filter; + tu->inode = d_real_inode(tu->path.dentry); ret = uprobe_register(tu->inode, tu->offset, &tu->consumer); if (ret) goto err_buffer; @@ -967,6 +962,7 @@ probe_event_disable(struct trace_uprobe *tu, struct trace_event_file *file) WARN_ON(!uprobe_filter_is_empty(&tu->filter)); uprobe_unregister(tu->inode, tu->offset, &tu->consumer); + tu->inode = NULL; tu->tp.flags &= file ? ~TP_FLAG_TRACE : ~TP_FLAG_PROFILE; uprobe_buffer_disable(); @@ -1337,7 +1333,6 @@ struct trace_event_call * create_local_trace_uprobe(char *name, unsigned long offs, bool is_return) { struct trace_uprobe *tu; - struct inode *inode; struct path path; int ret; @@ -1345,11 +1340,8 @@ create_local_trace_uprobe(char *name, unsigned long offs, bool is_return) if (ret) return ERR_PTR(ret); - inode = igrab(d_inode(path.dentry)); - path_put(&path); - - if (!inode || !S_ISREG(inode->i_mode)) { - iput(inode); + if (!d_is_reg(path.dentry)) { + path_put(&path); return ERR_PTR(-EINVAL); } @@ -1364,11 +1356,12 @@ create_local_trace_uprobe(char *name, unsigned long offs, bool is_return) if (IS_ERR(tu)) { pr_info("Failed to allocate trace_uprobe.(%d)\n", (int)PTR_ERR(tu)); + path_put(&path); return ERR_CAST(tu); } tu->offset = offs; - tu->inode = inode; + tu->path = path; tu->filename = kstrdup(name, GFP_KERNEL); init_trace_event_call(tu, &tu->tp.call); diff --git a/kernel/tracepoint.c b/kernel/tracepoint.c index 671b13457387..1e37da2e0c25 100644 --- a/kernel/tracepoint.c +++ b/kernel/tracepoint.c @@ -207,7 +207,7 @@ static int tracepoint_add_func(struct tracepoint *tp, lockdep_is_held(&tracepoints_mutex)); old = func_add(&tp_funcs, func, prio); if (IS_ERR(old)) { - WARN_ON_ONCE(1); + WARN_ON_ONCE(PTR_ERR(old) != -ENOMEM); return PTR_ERR(old); } @@ -239,7 +239,7 @@ static int tracepoint_remove_func(struct tracepoint *tp, lockdep_is_held(&tracepoints_mutex)); old = func_remove(&tp_funcs, func); if (IS_ERR(old)) { - WARN_ON_ONCE(1); + WARN_ON_ONCE(PTR_ERR(old) != -ENOMEM); return PTR_ERR(old); } diff --git a/lib/errseq.c b/lib/errseq.c index df782418b333..81f9e33aa7e7 100644 --- a/lib/errseq.c +++ b/lib/errseq.c @@ -111,27 +111,22 @@ EXPORT_SYMBOL(errseq_set); * errseq_sample() - Grab current errseq_t value. * @eseq: Pointer to errseq_t to be sampled. * - * This function allows callers to sample an errseq_t value, marking it as - * "seen" if required. + * This function allows callers to initialise their errseq_t variable. + * If the error has been "seen", new callers will not see an old error. + * If there is an unseen error in @eseq, the caller of this function will + * see it the next time it checks for an error. * + * Context: Any context. * Return: The current errseq value. */ errseq_t errseq_sample(errseq_t *eseq) { errseq_t old = READ_ONCE(*eseq); - errseq_t new = old; - /* - * For the common case of no errors ever having been set, we can skip - * marking the SEEN bit. Once an error has been set, the value will - * never go back to zero. - */ - if (old != 0) { - new |= ERRSEQ_SEEN; - if (old != new) - cmpxchg(eseq, old, new); - } - return new; + /* If nobody has seen this error yet, then we can be the first. */ + if (!(old & ERRSEQ_SEEN)) + old = 0; + return old; } EXPORT_SYMBOL(errseq_sample); diff --git a/lib/find_bit_benchmark.c b/lib/find_bit_benchmark.c index 5985a25e6cbc..5367ffa5c18f 100644 --- a/lib/find_bit_benchmark.c +++ b/lib/find_bit_benchmark.c @@ -132,7 +132,12 @@ static int __init find_bit_test(void) test_find_next_bit(bitmap, BITMAP_LEN); test_find_next_zero_bit(bitmap, BITMAP_LEN); test_find_last_bit(bitmap, BITMAP_LEN); - test_find_first_bit(bitmap, BITMAP_LEN); + + /* + * test_find_first_bit() may take some time, so + * traverse only part of bitmap to avoid soft lockup. + */ + test_find_first_bit(bitmap, BITMAP_LEN / 10); test_find_next_and_bit(bitmap, bitmap2, BITMAP_LEN); pr_err("\nStart testing find_bit() with sparse bitmap\n"); diff --git a/lib/swiotlb.c b/lib/swiotlb.c index fece57566d45..cc640588f145 100644 --- a/lib/swiotlb.c +++ b/lib/swiotlb.c @@ -714,7 +714,7 @@ swiotlb_alloc_buffer(struct device *dev, size_t size, dma_addr_t *dma_handle, phys_addr = swiotlb_tbl_map_single(dev, __phys_to_dma(dev, io_tlb_start), - 0, size, DMA_FROM_DEVICE, 0); + 0, size, DMA_FROM_DEVICE, attrs); if (phys_addr == SWIOTLB_MAP_ERROR) goto out_warn; @@ -737,7 +737,7 @@ out_unmap: swiotlb_tbl_unmap_single(dev, phys_addr, size, DMA_TO_DEVICE, DMA_ATTR_SKIP_CPU_SYNC); out_warn: - if ((attrs & DMA_ATTR_NO_WARN) && printk_ratelimit()) { + if (!(attrs & DMA_ATTR_NO_WARN) && printk_ratelimit()) { dev_warn(dev, "swiotlb: coherent allocation failed, size=%zu\n", size); diff --git a/mm/backing-dev.c b/mm/backing-dev.c index 023190c69dce..7441bd93b732 100644 --- a/mm/backing-dev.c +++ b/mm/backing-dev.c @@ -115,6 +115,7 @@ static int bdi_debug_register(struct backing_dev_info *bdi, const char *name) bdi, &bdi_debug_stats_fops); if (!bdi->debug_stats) { debugfs_remove(bdi->debug_dir); + bdi->debug_dir = NULL; return -ENOMEM; } @@ -383,7 +384,7 @@ static void wb_shutdown(struct bdi_writeback *wb) * the barrier provided by test_and_clear_bit() above. */ smp_wmb(); - clear_bit(WB_shutting_down, &wb->state); + clear_and_wake_up_bit(WB_shutting_down, &wb->state); } static void wb_exit(struct bdi_writeback *wb) diff --git a/mm/migrate.c b/mm/migrate.c index 568433023831..8c0af0f7cab1 100644 --- a/mm/migrate.c +++ b/mm/migrate.c @@ -528,14 +528,12 @@ int migrate_page_move_mapping(struct address_space *mapping, int i; int index = page_index(page); - for (i = 0; i < HPAGE_PMD_NR; i++) { + for (i = 1; i < HPAGE_PMD_NR; i++) { pslot = radix_tree_lookup_slot(&mapping->i_pages, index + i); radix_tree_replace_slot(&mapping->i_pages, pslot, newpage + i); } - } else { - radix_tree_replace_slot(&mapping->i_pages, pslot, newpage); } /* diff --git a/mm/mmap.c b/mm/mmap.c index 9d5968d1e8e3..78e14facdb6e 100644 --- a/mm/mmap.c +++ b/mm/mmap.c @@ -1324,6 +1324,35 @@ static inline int mlock_future_check(struct mm_struct *mm, return 0; } +static inline u64 file_mmap_size_max(struct file *file, struct inode *inode) +{ + if (S_ISREG(inode->i_mode)) + return inode->i_sb->s_maxbytes; + + if (S_ISBLK(inode->i_mode)) + return MAX_LFS_FILESIZE; + + /* Special "we do even unsigned file positions" case */ + if (file->f_mode & FMODE_UNSIGNED_OFFSET) + return 0; + + /* Yes, random drivers might want more. But I'm tired of buggy drivers */ + return ULONG_MAX; +} + +static inline bool file_mmap_ok(struct file *file, struct inode *inode, + unsigned long pgoff, unsigned long len) +{ + u64 maxsize = file_mmap_size_max(file, inode); + + if (maxsize && len > maxsize) + return false; + maxsize -= len; + if (pgoff > maxsize >> PAGE_SHIFT) + return false; + return true; +} + /* * The caller must hold down_write(¤t->mm->mmap_sem). */ @@ -1409,6 +1438,9 @@ unsigned long do_mmap(struct file *file, unsigned long addr, struct inode *inode = file_inode(file); unsigned long flags_mask; + if (!file_mmap_ok(file, inode, pgoff, len)) + return -EOVERFLOW; + flags_mask = LEGACY_MAP_MASK | file->f_op->mmap_supported_flags; switch (flags & MAP_TYPE) { @@ -3024,6 +3056,32 @@ void exit_mmap(struct mm_struct *mm) /* mm's last user has gone, and its about to be pulled down */ mmu_notifier_release(mm); + if (unlikely(mm_is_oom_victim(mm))) { + /* + * Manually reap the mm to free as much memory as possible. + * Then, as the oom reaper does, set MMF_OOM_SKIP to disregard + * this mm from further consideration. Taking mm->mmap_sem for + * write after setting MMF_OOM_SKIP will guarantee that the oom + * reaper will not run on this mm again after mmap_sem is + * dropped. + * + * Nothing can be holding mm->mmap_sem here and the above call + * to mmu_notifier_release(mm) ensures mmu notifier callbacks in + * __oom_reap_task_mm() will not block. + * + * This needs to be done before calling munlock_vma_pages_all(), + * which clears VM_LOCKED, otherwise the oom reaper cannot + * reliably test it. + */ + mutex_lock(&oom_lock); + __oom_reap_task_mm(mm); + mutex_unlock(&oom_lock); + + set_bit(MMF_OOM_SKIP, &mm->flags); + down_write(&mm->mmap_sem); + up_write(&mm->mmap_sem); + } + if (mm->locked_vm) { vma = mm->mmap; while (vma) { @@ -3045,24 +3103,6 @@ void exit_mmap(struct mm_struct *mm) /* update_hiwater_rss(mm) here? but nobody should be looking */ /* Use -1 here to ensure all VMAs in the mm are unmapped */ unmap_vmas(&tlb, vma, 0, -1); - - if (unlikely(mm_is_oom_victim(mm))) { - /* - * Wait for oom_reap_task() to stop working on this - * mm. Because MMF_OOM_SKIP is already set before - * calling down_read(), oom_reap_task() will not run - * on this "mm" post up_write(). - * - * mm_is_oom_victim() cannot be set from under us - * either because victim->mm is already set to NULL - * under task_lock before calling mmput and oom_mm is - * set not NULL by the OOM killer only if victim->mm - * is found not NULL while holding the task_lock. - */ - set_bit(MMF_OOM_SKIP, &mm->flags); - down_write(&mm->mmap_sem); - up_write(&mm->mmap_sem); - } free_pgtables(&tlb, vma, FIRST_USER_ADDRESS, USER_PGTABLES_CEILING); tlb_finish_mmu(&tlb, 0, -1); diff --git a/mm/oom_kill.c b/mm/oom_kill.c index ff992fa8760a..8ba6cb88cf58 100644 --- a/mm/oom_kill.c +++ b/mm/oom_kill.c @@ -469,7 +469,6 @@ bool process_shares_mm(struct task_struct *p, struct mm_struct *mm) return false; } - #ifdef CONFIG_MMU /* * OOM Reaper kernel thread which tries to reap the memory used by the OOM @@ -480,16 +479,54 @@ static DECLARE_WAIT_QUEUE_HEAD(oom_reaper_wait); static struct task_struct *oom_reaper_list; static DEFINE_SPINLOCK(oom_reaper_lock); -static bool __oom_reap_task_mm(struct task_struct *tsk, struct mm_struct *mm) +void __oom_reap_task_mm(struct mm_struct *mm) { - struct mmu_gather tlb; struct vm_area_struct *vma; + + /* + * Tell all users of get_user/copy_from_user etc... that the content + * is no longer stable. No barriers really needed because unmapping + * should imply barriers already and the reader would hit a page fault + * if it stumbled over a reaped memory. + */ + set_bit(MMF_UNSTABLE, &mm->flags); + + for (vma = mm->mmap ; vma; vma = vma->vm_next) { + if (!can_madv_dontneed_vma(vma)) + continue; + + /* + * Only anonymous pages have a good chance to be dropped + * without additional steps which we cannot afford as we + * are OOM already. + * + * We do not even care about fs backed pages because all + * which are reclaimable have already been reclaimed and + * we do not want to block exit_mmap by keeping mm ref + * count elevated without a good reason. + */ + if (vma_is_anonymous(vma) || !(vma->vm_flags & VM_SHARED)) { + const unsigned long start = vma->vm_start; + const unsigned long end = vma->vm_end; + struct mmu_gather tlb; + + tlb_gather_mmu(&tlb, mm, start, end); + mmu_notifier_invalidate_range_start(mm, start, end); + unmap_page_range(&tlb, vma, start, end, NULL); + mmu_notifier_invalidate_range_end(mm, start, end); + tlb_finish_mmu(&tlb, start, end); + } + } +} + +static bool oom_reap_task_mm(struct task_struct *tsk, struct mm_struct *mm) +{ bool ret = true; /* * We have to make sure to not race with the victim exit path * and cause premature new oom victim selection: - * __oom_reap_task_mm exit_mm + * oom_reap_task_mm exit_mm * mmget_not_zero * mmput * atomic_dec_and_test @@ -534,39 +571,8 @@ static bool __oom_reap_task_mm(struct task_struct *tsk, struct mm_struct *mm) trace_start_task_reaping(tsk->pid); - /* - * Tell all users of get_user/copy_from_user etc... that the content - * is no longer stable. No barriers really needed because unmapping - * should imply barriers already and the reader would hit a page fault - * if it stumbled over a reaped memory. - */ - set_bit(MMF_UNSTABLE, &mm->flags); - - for (vma = mm->mmap ; vma; vma = vma->vm_next) { - if (!can_madv_dontneed_vma(vma)) - continue; + __oom_reap_task_mm(mm); - /* - * Only anonymous pages have a good chance to be dropped - * without additional steps which we cannot afford as we - * are OOM already. - * - * We do not even care about fs backed pages because all - * which are reclaimable have already been reclaimed and - * we do not want to block exit_mmap by keeping mm ref - * count elevated without a good reason. - */ - if (vma_is_anonymous(vma) || !(vma->vm_flags & VM_SHARED)) { - const unsigned long start = vma->vm_start; - const unsigned long end = vma->vm_end; - - tlb_gather_mmu(&tlb, mm, start, end); - mmu_notifier_invalidate_range_start(mm, start, end); - unmap_page_range(&tlb, vma, start, end, NULL); - mmu_notifier_invalidate_range_end(mm, start, end); - tlb_finish_mmu(&tlb, start, end); - } - } pr_info("oom_reaper: reaped process %d (%s), now anon-rss:%lukB, file-rss:%lukB, shmem-rss:%lukB\n", task_pid_nr(tsk), tsk->comm, K(get_mm_counter(mm, MM_ANONPAGES)), @@ -587,14 +593,13 @@ static void oom_reap_task(struct task_struct *tsk) struct mm_struct *mm = tsk->signal->oom_mm; /* Retry the down_read_trylock(mmap_sem) a few times */ - while (attempts++ < MAX_OOM_REAP_RETRIES && !__oom_reap_task_mm(tsk, mm)) + while (attempts++ < MAX_OOM_REAP_RETRIES && !oom_reap_task_mm(tsk, mm)) schedule_timeout_idle(HZ/10); if (attempts <= MAX_OOM_REAP_RETRIES || test_bit(MMF_OOM_SKIP, &mm->flags)) goto done; - pr_info("oom_reaper: unable to reap pid:%d (%s)\n", task_pid_nr(tsk), tsk->comm); debug_show_all_locks(); diff --git a/mm/sparse.c b/mm/sparse.c index 62eef264a7bd..73dc2fcc0eab 100644 --- a/mm/sparse.c +++ b/mm/sparse.c @@ -629,7 +629,7 @@ void offline_mem_sections(unsigned long start_pfn, unsigned long end_pfn) unsigned long pfn; for (pfn = start_pfn; pfn < end_pfn; pfn += PAGES_PER_SECTION) { - unsigned long section_nr = pfn_to_section_nr(start_pfn); + unsigned long section_nr = pfn_to_section_nr(pfn); struct mem_section *ms; /* diff --git a/mm/vmstat.c b/mm/vmstat.c index 536332e988b8..a2b9518980ce 100644 --- a/mm/vmstat.c +++ b/mm/vmstat.c @@ -1161,7 +1161,7 @@ const char * const vmstat_text[] = { "nr_vmscan_immediate_reclaim", "nr_dirtied", "nr_written", - "nr_indirectly_reclaimable", + "", /* nr_indirectly_reclaimable */ /* enum writeback_stat_item counters */ "nr_dirty_threshold", @@ -1740,6 +1740,10 @@ static int vmstat_show(struct seq_file *m, void *arg) unsigned long *l = arg; unsigned long off = l - (unsigned long *)m->private; + /* Skip hidden vmstat items. */ + if (*vmstat_text[off] == '\0') + return 0; + seq_puts(m, vmstat_text[off]); seq_put_decimal_ull(m, " ", *l); seq_putc(m, '\n'); diff --git a/mm/z3fold.c b/mm/z3fold.c index c0bca6153b95..4b366d181f35 100644 --- a/mm/z3fold.c +++ b/mm/z3fold.c @@ -144,7 +144,8 @@ enum z3fold_page_flags { PAGE_HEADLESS = 0, MIDDLE_CHUNK_MAPPED, NEEDS_COMPACTING, - PAGE_STALE + PAGE_STALE, + UNDER_RECLAIM }; /***************** @@ -173,6 +174,7 @@ static struct z3fold_header *init_z3fold_page(struct page *page, clear_bit(MIDDLE_CHUNK_MAPPED, &page->private); clear_bit(NEEDS_COMPACTING, &page->private); clear_bit(PAGE_STALE, &page->private); + clear_bit(UNDER_RECLAIM, &page->private); spin_lock_init(&zhdr->page_lock); kref_init(&zhdr->refcount); @@ -756,6 +758,10 @@ static void z3fold_free(struct z3fold_pool *pool, unsigned long handle) atomic64_dec(&pool->pages_nr); return; } + if (test_bit(UNDER_RECLAIM, &page->private)) { + z3fold_page_unlock(zhdr); + return; + } if (test_and_set_bit(NEEDS_COMPACTING, &page->private)) { z3fold_page_unlock(zhdr); return; @@ -840,6 +846,8 @@ static int z3fold_reclaim_page(struct z3fold_pool *pool, unsigned int retries) kref_get(&zhdr->refcount); list_del_init(&zhdr->buddy); zhdr->cpu = -1; + set_bit(UNDER_RECLAIM, &page->private); + break; } list_del_init(&page->lru); @@ -887,25 +895,35 @@ static int z3fold_reclaim_page(struct z3fold_pool *pool, unsigned int retries) goto next; } next: - spin_lock(&pool->lock); if (test_bit(PAGE_HEADLESS, &page->private)) { if (ret == 0) { - spin_unlock(&pool->lock); free_z3fold_page(page); return 0; } - } else if (kref_put(&zhdr->refcount, release_z3fold_page)) { - atomic64_dec(&pool->pages_nr); + spin_lock(&pool->lock); + list_add(&page->lru, &pool->lru); + spin_unlock(&pool->lock); + } else { + z3fold_page_lock(zhdr); + clear_bit(UNDER_RECLAIM, &page->private); + if (kref_put(&zhdr->refcount, + release_z3fold_page_locked)) { + atomic64_dec(&pool->pages_nr); + return 0; + } + /* + * if we are here, the page is still not completely + * free. Take the global pool lock then to be able + * to add it back to the lru list + */ + spin_lock(&pool->lock); + list_add(&page->lru, &pool->lru); spin_unlock(&pool->lock); - return 0; + z3fold_page_unlock(zhdr); } - /* - * Add to the beginning of LRU. - * Pool lock has to be kept here to ensure the page has - * not already been released - */ - list_add(&page->lru, &pool->lru); + /* We started off locked to we need to lock the pool back */ + spin_lock(&pool->lock); } spin_unlock(&pool->lock); return -EAGAIN; diff --git a/net/9p/trans_common.c b/net/9p/trans_common.c index 38aa6345bdfa..b718db2085b2 100644 --- a/net/9p/trans_common.c +++ b/net/9p/trans_common.c @@ -16,7 +16,7 @@ #include <linux/module.h> /** - * p9_release_req_pages - Release pages after the transaction. + * p9_release_pages - Release pages after the transaction. */ void p9_release_pages(struct page **pages, int nr_pages) { diff --git a/net/9p/trans_fd.c b/net/9p/trans_fd.c index 0cfba919d167..848969fe7979 100644 --- a/net/9p/trans_fd.c +++ b/net/9p/trans_fd.c @@ -1092,8 +1092,8 @@ static struct p9_trans_module p9_fd_trans = { }; /** - * p9_poll_proc - poll worker thread - * @a: thread state and arguments + * p9_poll_workfn - poll worker thread + * @work: work queue * * polls all v9fs transports for new events and queues the appropriate * work to the work queue diff --git a/net/9p/trans_rdma.c b/net/9p/trans_rdma.c index 6d8e3031978f..3d414acb7015 100644 --- a/net/9p/trans_rdma.c +++ b/net/9p/trans_rdma.c @@ -68,8 +68,6 @@ * @pd: Protection Domain pointer * @qp: Queue Pair pointer * @cq: Completion Queue pointer - * @dm_mr: DMA Memory Region pointer - * @lkey: The local access only memory region key * @timeout: Number of uSecs to wait for connection management events * @privport: Whether a privileged port may be used * @port: The port to use @@ -632,7 +630,7 @@ static int p9_rdma_bind_privport(struct p9_trans_rdma *rdma) } /** - * trans_create_rdma - Transport method for creating atransport instance + * rdma_create_trans - Transport method for creating a transport instance * @client: client instance * @addr: IP address string * @args: Mount options string diff --git a/net/9p/trans_virtio.c b/net/9p/trans_virtio.c index 3aa5a93ad107..4d0372263e5d 100644 --- a/net/9p/trans_virtio.c +++ b/net/9p/trans_virtio.c @@ -60,7 +60,6 @@ static atomic_t vp_pinned = ATOMIC_INIT(0); /** * struct virtio_chan - per-instance transport information - * @initialized: whether the channel is initialized * @inuse: whether the channel is in use * @lock: protects multiple elements within this structure * @client: client instance @@ -385,8 +384,8 @@ static int p9_get_mapped_pages(struct virtio_chan *chan, * @uidata: user bffer that should be ued for zero copy read * @uodata: user buffer that shoud be user for zero copy write * @inlen: read buffer size - * @olen: write buffer size - * @hdrlen: reader header size, This is the size of response protocol data + * @outlen: write buffer size + * @in_hdr_len: reader header size, This is the size of response protocol data * */ static int diff --git a/net/9p/trans_xen.c b/net/9p/trans_xen.c index 086a4abdfa7c..0f19960390a6 100644 --- a/net/9p/trans_xen.c +++ b/net/9p/trans_xen.c @@ -485,7 +485,7 @@ static int xen_9pfs_front_probe(struct xenbus_device *dev, static int xen_9pfs_front_resume(struct xenbus_device *dev) { - dev_warn(&dev->dev, "suspsend/resume unsupported\n"); + dev_warn(&dev->dev, "suspend/resume unsupported\n"); return 0; } diff --git a/net/atm/lec.c b/net/atm/lec.c index 01d5d20a6eb1..3138a869b5c0 100644 --- a/net/atm/lec.c +++ b/net/atm/lec.c @@ -41,6 +41,9 @@ static unsigned char bridge_ula_lec[] = { 0x01, 0x80, 0xc2, 0x00, 0x00 }; #include <linux/module.h> #include <linux/init.h> +/* Hardening for Spectre-v1 */ +#include <linux/nospec.h> + #include "lec.h" #include "lec_arpc.h" #include "resources.h" @@ -687,8 +690,10 @@ static int lec_vcc_attach(struct atm_vcc *vcc, void __user *arg) bytes_left = copy_from_user(&ioc_data, arg, sizeof(struct atmlec_ioc)); if (bytes_left != 0) pr_info("copy from user failed for %d bytes\n", bytes_left); - if (ioc_data.dev_num < 0 || ioc_data.dev_num >= MAX_LEC_ITF || - !dev_lec[ioc_data.dev_num]) + if (ioc_data.dev_num < 0 || ioc_data.dev_num >= MAX_LEC_ITF) + return -EINVAL; + ioc_data.dev_num = array_index_nospec(ioc_data.dev_num, MAX_LEC_ITF); + if (!dev_lec[ioc_data.dev_num]) return -EINVAL; vpriv = kmalloc(sizeof(struct lec_vcc_priv), GFP_KERNEL); if (!vpriv) diff --git a/net/bridge/br_if.c b/net/bridge/br_if.c index 82c1a6f430b3..5bb6681fa91e 100644 --- a/net/bridge/br_if.c +++ b/net/bridge/br_if.c @@ -518,8 +518,8 @@ int br_add_if(struct net_bridge *br, struct net_device *dev, return -ELOOP; } - /* Device is already being bridged */ - if (br_port_exists(dev)) + /* Device has master upper dev */ + if (netdev_master_upper_dev_get(dev)) return -EBUSY; /* No bridging devices that dislike that (e.g. wireless) */ diff --git a/net/ceph/osd_client.c b/net/ceph/osd_client.c index ea2a6c9fb7ce..d2667e5dddc3 100644 --- a/net/ceph/osd_client.c +++ b/net/ceph/osd_client.c @@ -157,10 +157,12 @@ static void ceph_osd_data_bio_init(struct ceph_osd_data *osd_data, #endif /* CONFIG_BLOCK */ static void ceph_osd_data_bvecs_init(struct ceph_osd_data *osd_data, - struct ceph_bvec_iter *bvec_pos) + struct ceph_bvec_iter *bvec_pos, + u32 num_bvecs) { osd_data->type = CEPH_OSD_DATA_TYPE_BVECS; osd_data->bvec_pos = *bvec_pos; + osd_data->num_bvecs = num_bvecs; } #define osd_req_op_data(oreq, whch, typ, fld) \ @@ -237,6 +239,22 @@ void osd_req_op_extent_osd_data_bio(struct ceph_osd_request *osd_req, EXPORT_SYMBOL(osd_req_op_extent_osd_data_bio); #endif /* CONFIG_BLOCK */ +void osd_req_op_extent_osd_data_bvecs(struct ceph_osd_request *osd_req, + unsigned int which, + struct bio_vec *bvecs, u32 num_bvecs, + u32 bytes) +{ + struct ceph_osd_data *osd_data; + struct ceph_bvec_iter it = { + .bvecs = bvecs, + .iter = { .bi_size = bytes }, + }; + + osd_data = osd_req_op_data(osd_req, which, extent, osd_data); + ceph_osd_data_bvecs_init(osd_data, &it, num_bvecs); +} +EXPORT_SYMBOL(osd_req_op_extent_osd_data_bvecs); + void osd_req_op_extent_osd_data_bvec_pos(struct ceph_osd_request *osd_req, unsigned int which, struct ceph_bvec_iter *bvec_pos) @@ -244,7 +262,7 @@ void osd_req_op_extent_osd_data_bvec_pos(struct ceph_osd_request *osd_req, struct ceph_osd_data *osd_data; osd_data = osd_req_op_data(osd_req, which, extent, osd_data); - ceph_osd_data_bvecs_init(osd_data, bvec_pos); + ceph_osd_data_bvecs_init(osd_data, bvec_pos, 0); } EXPORT_SYMBOL(osd_req_op_extent_osd_data_bvec_pos); @@ -287,7 +305,8 @@ EXPORT_SYMBOL(osd_req_op_cls_request_data_pages); void osd_req_op_cls_request_data_bvecs(struct ceph_osd_request *osd_req, unsigned int which, - struct bio_vec *bvecs, u32 bytes) + struct bio_vec *bvecs, u32 num_bvecs, + u32 bytes) { struct ceph_osd_data *osd_data; struct ceph_bvec_iter it = { @@ -296,7 +315,7 @@ void osd_req_op_cls_request_data_bvecs(struct ceph_osd_request *osd_req, }; osd_data = osd_req_op_data(osd_req, which, cls, request_data); - ceph_osd_data_bvecs_init(osd_data, &it); + ceph_osd_data_bvecs_init(osd_data, &it, num_bvecs); osd_req->r_ops[which].cls.indata_len += bytes; osd_req->r_ops[which].indata_len += bytes; } diff --git a/net/compat.c b/net/compat.c index 5ae7437d3853..7242cce5631b 100644 --- a/net/compat.c +++ b/net/compat.c @@ -377,7 +377,8 @@ static int compat_sock_setsockopt(struct socket *sock, int level, int optname, optname == SO_ATTACH_REUSEPORT_CBPF) return do_set_attach_filter(sock, level, optname, optval, optlen); - if (optname == SO_RCVTIMEO || optname == SO_SNDTIMEO) + if (!COMPAT_USE_64BIT_TIME && + (optname == SO_RCVTIMEO || optname == SO_SNDTIMEO)) return do_set_sock_timeout(sock, level, optname, optval, optlen); return sock_setsockopt(sock, level, optname, optval, optlen); @@ -448,7 +449,8 @@ static int do_get_sock_timeout(struct socket *sock, int level, int optname, static int compat_sock_getsockopt(struct socket *sock, int level, int optname, char __user *optval, int __user *optlen) { - if (optname == SO_RCVTIMEO || optname == SO_SNDTIMEO) + if (!COMPAT_USE_64BIT_TIME && + (optname == SO_RCVTIMEO || optname == SO_SNDTIMEO)) return do_get_sock_timeout(sock, level, optname, optval, optlen); return sock_getsockopt(sock, level, optname, optval, optlen); } diff --git a/net/core/ethtool.c b/net/core/ethtool.c index 03416e6dd5d7..ba02f0dfe85c 100644 --- a/net/core/ethtool.c +++ b/net/core/ethtool.c @@ -1032,6 +1032,11 @@ static noinline_for_stack int ethtool_get_rxnfc(struct net_device *dev, info_size = sizeof(info); if (copy_from_user(&info, useraddr, info_size)) return -EFAULT; + /* Since malicious users may modify the original data, + * we need to check whether FLOW_RSS is still requested. + */ + if (!(info.flow_type & FLOW_RSS)) + return -EINVAL; } if (info.cmd == ETHTOOL_GRXCLSRLALL) { diff --git a/net/core/filter.c b/net/core/filter.c index d31aff93270d..e77c30ca491d 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -3240,6 +3240,7 @@ BPF_CALL_4(bpf_skb_set_tunnel_key, struct sk_buff *, skb, skb_dst_set(skb, (struct dst_entry *) md); info = &md->u.tun_info; + memset(info, 0, sizeof(*info)); info->mode = IP_TUNNEL_INFO_TX; info->key.tun_flags = TUNNEL_KEY | TUNNEL_CSUM | TUNNEL_NOCACHE; diff --git a/net/dccp/ccids/ccid2.c b/net/dccp/ccids/ccid2.c index 92d016e87816..385f153fe031 100644 --- a/net/dccp/ccids/ccid2.c +++ b/net/dccp/ccids/ccid2.c @@ -126,6 +126,16 @@ static void ccid2_change_l_seq_window(struct sock *sk, u64 val) DCCPF_SEQ_WMAX)); } +static void dccp_tasklet_schedule(struct sock *sk) +{ + struct tasklet_struct *t = &dccp_sk(sk)->dccps_xmitlet; + + if (!test_and_set_bit(TASKLET_STATE_SCHED, &t->state)) { + sock_hold(sk); + __tasklet_schedule(t); + } +} + static void ccid2_hc_tx_rto_expire(struct timer_list *t) { struct ccid2_hc_tx_sock *hc = from_timer(hc, t, tx_rtotimer); @@ -166,7 +176,7 @@ static void ccid2_hc_tx_rto_expire(struct timer_list *t) /* if we were blocked before, we may now send cwnd=1 packet */ if (sender_was_blocked) - tasklet_schedule(&dccp_sk(sk)->dccps_xmitlet); + dccp_tasklet_schedule(sk); /* restart backed-off timer */ sk_reset_timer(sk, &hc->tx_rtotimer, jiffies + hc->tx_rto); out: @@ -706,7 +716,7 @@ static void ccid2_hc_tx_packet_recv(struct sock *sk, struct sk_buff *skb) done: /* check if incoming Acks allow pending packets to be sent */ if (sender_was_blocked && !ccid2_cwnd_network_limited(hc)) - tasklet_schedule(&dccp_sk(sk)->dccps_xmitlet); + dccp_tasklet_schedule(sk); dccp_ackvec_parsed_cleanup(&hc->tx_av_chunks); } diff --git a/net/dccp/timer.c b/net/dccp/timer.c index b50a8732ff43..1501a20a94ca 100644 --- a/net/dccp/timer.c +++ b/net/dccp/timer.c @@ -232,6 +232,7 @@ static void dccp_write_xmitlet(unsigned long data) else dccp_write_xmit(sk); bh_unlock_sock(sk); + sock_put(sk); } static void dccp_write_xmit_timer(struct timer_list *t) @@ -240,7 +241,6 @@ static void dccp_write_xmit_timer(struct timer_list *t) struct sock *sk = &dp->dccps_inet_connection.icsk_inet.sk; dccp_write_xmitlet((unsigned long)sk); - sock_put(sk); } void dccp_init_xmit_timers(struct sock *sk) diff --git a/net/ieee802154/6lowpan/6lowpan_i.h b/net/ieee802154/6lowpan/6lowpan_i.h index b8d95cb71c25..44a7e16bf3b5 100644 --- a/net/ieee802154/6lowpan/6lowpan_i.h +++ b/net/ieee802154/6lowpan/6lowpan_i.h @@ -20,8 +20,8 @@ typedef unsigned __bitwise lowpan_rx_result; struct frag_lowpan_compare_key { u16 tag; u16 d_size; - const struct ieee802154_addr src; - const struct ieee802154_addr dst; + struct ieee802154_addr src; + struct ieee802154_addr dst; }; /* Equivalent of ipv4 struct ipq diff --git a/net/ieee802154/6lowpan/reassembly.c b/net/ieee802154/6lowpan/reassembly.c index 1790b65944b3..2cc224106b69 100644 --- a/net/ieee802154/6lowpan/reassembly.c +++ b/net/ieee802154/6lowpan/reassembly.c @@ -75,14 +75,14 @@ fq_find(struct net *net, const struct lowpan_802154_cb *cb, { struct netns_ieee802154_lowpan *ieee802154_lowpan = net_ieee802154_lowpan(net); - struct frag_lowpan_compare_key key = { - .tag = cb->d_tag, - .d_size = cb->d_size, - .src = *src, - .dst = *dst, - }; + struct frag_lowpan_compare_key key = {}; struct inet_frag_queue *q; + key.tag = cb->d_tag; + key.d_size = cb->d_size; + key.src = *src; + key.dst = *dst; + q = inet_frag_find(&ieee802154_lowpan->frags, &key); if (!q) return NULL; @@ -372,7 +372,7 @@ int lowpan_frag_rcv(struct sk_buff *skb, u8 frag_type) struct lowpan_frag_queue *fq; struct net *net = dev_net(skb->dev); struct lowpan_802154_cb *cb = lowpan_802154_cb(skb); - struct ieee802154_hdr hdr; + struct ieee802154_hdr hdr = {}; int err; if (ieee802154_hdr_peek_addrs(skb, &hdr) < 0) diff --git a/net/ipv4/ping.c b/net/ipv4/ping.c index 05e47d777009..56a010622f70 100644 --- a/net/ipv4/ping.c +++ b/net/ipv4/ping.c @@ -775,8 +775,10 @@ static int ping_v4_sendmsg(struct sock *sk, struct msghdr *msg, size_t len) ipc.addr = faddr = daddr; if (ipc.opt && ipc.opt->opt.srr) { - if (!daddr) - return -EINVAL; + if (!daddr) { + err = -EINVAL; + goto out_free; + } faddr = ipc.opt->opt.faddr; } tos = get_rttos(&ipc, inet); @@ -842,6 +844,7 @@ back_from_confirm: out: ip_rt_put(rt); +out_free: if (free) kfree(ipc.opt); if (!err) { diff --git a/net/ipv4/route.c b/net/ipv4/route.c index ccb25d80f679..29268efad247 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -709,7 +709,7 @@ static void update_or_create_fnhe(struct fib_nh *nh, __be32 daddr, __be32 gw, fnhe->fnhe_gw = gw; fnhe->fnhe_pmtu = pmtu; fnhe->fnhe_mtu_locked = lock; - fnhe->fnhe_expires = expires; + fnhe->fnhe_expires = max(1UL, expires); /* Exception created; mark the cached routes for the nexthop * stale, so anyone caching it rechecks if this exception @@ -1297,6 +1297,36 @@ static unsigned int ipv4_mtu(const struct dst_entry *dst) return mtu - lwtunnel_headroom(dst->lwtstate, mtu); } +static void ip_del_fnhe(struct fib_nh *nh, __be32 daddr) +{ + struct fnhe_hash_bucket *hash; + struct fib_nh_exception *fnhe, __rcu **fnhe_p; + u32 hval = fnhe_hashfun(daddr); + + spin_lock_bh(&fnhe_lock); + + hash = rcu_dereference_protected(nh->nh_exceptions, + lockdep_is_held(&fnhe_lock)); + hash += hval; + + fnhe_p = &hash->chain; + fnhe = rcu_dereference_protected(*fnhe_p, lockdep_is_held(&fnhe_lock)); + while (fnhe) { + if (fnhe->fnhe_daddr == daddr) { + rcu_assign_pointer(*fnhe_p, rcu_dereference_protected( + fnhe->fnhe_next, lockdep_is_held(&fnhe_lock))); + fnhe_flush_routes(fnhe); + kfree_rcu(fnhe, rcu); + break; + } + fnhe_p = &fnhe->fnhe_next; + fnhe = rcu_dereference_protected(fnhe->fnhe_next, + lockdep_is_held(&fnhe_lock)); + } + + spin_unlock_bh(&fnhe_lock); +} + static struct fib_nh_exception *find_exception(struct fib_nh *nh, __be32 daddr) { struct fnhe_hash_bucket *hash = rcu_dereference(nh->nh_exceptions); @@ -1310,8 +1340,14 @@ static struct fib_nh_exception *find_exception(struct fib_nh *nh, __be32 daddr) for (fnhe = rcu_dereference(hash[hval].chain); fnhe; fnhe = rcu_dereference(fnhe->fnhe_next)) { - if (fnhe->fnhe_daddr == daddr) + if (fnhe->fnhe_daddr == daddr) { + if (fnhe->fnhe_expires && + time_after(jiffies, fnhe->fnhe_expires)) { + ip_del_fnhe(nh, daddr); + break; + } return fnhe; + } } return NULL; } @@ -1339,6 +1375,7 @@ static bool rt_bind_exception(struct rtable *rt, struct fib_nh_exception *fnhe, fnhe->fnhe_gw = 0; fnhe->fnhe_pmtu = 0; fnhe->fnhe_expires = 0; + fnhe->fnhe_mtu_locked = false; fnhe_flush_routes(fnhe); orig = NULL; } @@ -1636,36 +1673,6 @@ static void ip_handle_martian_source(struct net_device *dev, #endif } -static void ip_del_fnhe(struct fib_nh *nh, __be32 daddr) -{ - struct fnhe_hash_bucket *hash; - struct fib_nh_exception *fnhe, __rcu **fnhe_p; - u32 hval = fnhe_hashfun(daddr); - - spin_lock_bh(&fnhe_lock); - - hash = rcu_dereference_protected(nh->nh_exceptions, - lockdep_is_held(&fnhe_lock)); - hash += hval; - - fnhe_p = &hash->chain; - fnhe = rcu_dereference_protected(*fnhe_p, lockdep_is_held(&fnhe_lock)); - while (fnhe) { - if (fnhe->fnhe_daddr == daddr) { - rcu_assign_pointer(*fnhe_p, rcu_dereference_protected( - fnhe->fnhe_next, lockdep_is_held(&fnhe_lock))); - fnhe_flush_routes(fnhe); - kfree_rcu(fnhe, rcu); - break; - } - fnhe_p = &fnhe->fnhe_next; - fnhe = rcu_dereference_protected(fnhe->fnhe_next, - lockdep_is_held(&fnhe_lock)); - } - - spin_unlock_bh(&fnhe_lock); -} - /* called in rcu_read_lock() section */ static int __mkroute_input(struct sk_buff *skb, const struct fib_result *res, @@ -1719,20 +1726,10 @@ static int __mkroute_input(struct sk_buff *skb, fnhe = find_exception(&FIB_RES_NH(*res), daddr); if (do_cache) { - if (fnhe) { + if (fnhe) rth = rcu_dereference(fnhe->fnhe_rth_input); - if (rth && rth->dst.expires && - time_after(jiffies, rth->dst.expires)) { - ip_del_fnhe(&FIB_RES_NH(*res), daddr); - fnhe = NULL; - } else { - goto rt_cache; - } - } - - rth = rcu_dereference(FIB_RES_NH(*res).nh_rth_input); - -rt_cache: + else + rth = rcu_dereference(FIB_RES_NH(*res).nh_rth_input); if (rt_cache_valid(rth)) { skb_dst_set_noref(skb, &rth->dst); goto out; @@ -2216,39 +2213,31 @@ static struct rtable *__mkroute_output(const struct fib_result *res, * the loopback interface and the IP_PKTINFO ipi_ifindex will * be set to the loopback interface as well. */ - fi = NULL; + do_cache = false; } fnhe = NULL; do_cache &= fi != NULL; - if (do_cache) { + if (fi) { struct rtable __rcu **prth; struct fib_nh *nh = &FIB_RES_NH(*res); fnhe = find_exception(nh, fl4->daddr); + if (!do_cache) + goto add; if (fnhe) { prth = &fnhe->fnhe_rth_output; - rth = rcu_dereference(*prth); - if (rth && rth->dst.expires && - time_after(jiffies, rth->dst.expires)) { - ip_del_fnhe(nh, fl4->daddr); - fnhe = NULL; - } else { - goto rt_cache; + } else { + if (unlikely(fl4->flowi4_flags & + FLOWI_FLAG_KNOWN_NH && + !(nh->nh_gw && + nh->nh_scope == RT_SCOPE_LINK))) { + do_cache = false; + goto add; } + prth = raw_cpu_ptr(nh->nh_pcpu_rth_output); } - - if (unlikely(fl4->flowi4_flags & - FLOWI_FLAG_KNOWN_NH && - !(nh->nh_gw && - nh->nh_scope == RT_SCOPE_LINK))) { - do_cache = false; - goto add; - } - prth = raw_cpu_ptr(nh->nh_pcpu_rth_output); rth = rcu_dereference(*prth); - -rt_cache: if (rt_cache_valid(rth) && dst_hold_safe(&rth->dst)) return rth; } diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 9ce1c726185e..c9d00ef54dec 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -697,7 +697,7 @@ static bool tcp_should_autocork(struct sock *sk, struct sk_buff *skb, { return skb->len < size_goal && sock_net(sk)->ipv4.sysctl_tcp_autocorking && - skb != tcp_write_queue_head(sk) && + !tcp_rtx_queue_empty(sk) && refcount_read(&sk->sk_wmem_alloc) > skb->truesize; } @@ -1204,7 +1204,8 @@ int tcp_sendmsg_locked(struct sock *sk, struct msghdr *msg, size_t size) uarg->zerocopy = 0; } - if (unlikely(flags & MSG_FASTOPEN || inet_sk(sk)->defer_connect)) { + if (unlikely(flags & MSG_FASTOPEN || inet_sk(sk)->defer_connect) && + !tp->repair) { err = tcp_sendmsg_fastopen(sk, msg, &copied_syn, size); if (err == -EINPROGRESS && copied_syn > 0) goto out; @@ -2673,7 +2674,7 @@ static int do_tcp_setsockopt(struct sock *sk, int level, case TCP_REPAIR_QUEUE: if (!tp->repair) err = -EPERM; - else if (val < TCP_QUEUES_NR) + else if ((unsigned int)val < TCP_QUEUES_NR) tp->repair_queue = val; else err = -EINVAL; diff --git a/net/ipv4/tcp_bbr.c b/net/ipv4/tcp_bbr.c index 158d105e76da..58e2f479ffb4 100644 --- a/net/ipv4/tcp_bbr.c +++ b/net/ipv4/tcp_bbr.c @@ -806,7 +806,9 @@ static void bbr_update_min_rtt(struct sock *sk, const struct rate_sample *rs) } } } - bbr->idle_restart = 0; + /* Restart after idle ends only once we process a new S/ACK for data */ + if (rs->delivered > 0) + bbr->idle_restart = 0; } static void bbr_update_model(struct sock *sk, const struct rate_sample *rs) diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index 24b5c59b1c53..b61a770884fa 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -401,9 +401,9 @@ static int compute_score(struct sock *sk, struct net *net, bool dev_match = (sk->sk_bound_dev_if == dif || sk->sk_bound_dev_if == sdif); - if (exact_dif && !dev_match) + if (!dev_match) return -1; - if (sk->sk_bound_dev_if && dev_match) + if (sk->sk_bound_dev_if) score += 4; } @@ -952,8 +952,10 @@ int udp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len) sock_tx_timestamp(sk, ipc.sockc.tsflags, &ipc.tx_flags); if (ipc.opt && ipc.opt->opt.srr) { - if (!daddr) - return -EINVAL; + if (!daddr) { + err = -EINVAL; + goto out_free; + } faddr = ipc.opt->opt.faddr; connected = 0; } @@ -1074,6 +1076,7 @@ do_append_data: out: ip_rt_put(rt); +out_free: if (free) kfree(ipc.opt); if (!err) diff --git a/net/ipv6/Kconfig b/net/ipv6/Kconfig index 6794ddf0547c..11e4e80cf7e9 100644 --- a/net/ipv6/Kconfig +++ b/net/ipv6/Kconfig @@ -34,16 +34,15 @@ config IPV6_ROUTE_INFO bool "IPv6: Route Information (RFC 4191) support" depends on IPV6_ROUTER_PREF ---help--- - This is experimental support of Route Information. + Support of Route Information. If unsure, say N. config IPV6_OPTIMISTIC_DAD bool "IPv6: Enable RFC 4429 Optimistic DAD" ---help--- - This is experimental support for optimistic Duplicate - Address Detection. It allows for autoconfigured addresses - to be used more quickly. + Support for optimistic Duplicate Address Detection. It allows for + autoconfigured addresses to be used more quickly. If unsure, say N. @@ -280,7 +279,7 @@ config IPV6_MROUTE depends on IPV6 select IP_MROUTE_COMMON ---help--- - Experimental support for IPv6 multicast forwarding. + Support for IPv6 multicast forwarding. If unsure, say N. config IPV6_MROUTE_MULTIPLE_TABLES diff --git a/net/ipv6/ip6_vti.c b/net/ipv6/ip6_vti.c index c214ffec02f0..ca957dd93a29 100644 --- a/net/ipv6/ip6_vti.c +++ b/net/ipv6/ip6_vti.c @@ -669,7 +669,7 @@ static void vti6_link_config(struct ip6_tnl *t, bool keep_mtu) else mtu = ETH_DATA_LEN - LL_MAX_HEADER - sizeof(struct ipv6hdr); - dev->mtu = max_t(int, mtu, IPV6_MIN_MTU); + dev->mtu = max_t(int, mtu, IPV4_MIN_MTU); } /** @@ -881,7 +881,7 @@ static void vti6_dev_setup(struct net_device *dev) dev->priv_destructor = vti6_dev_free; dev->type = ARPHRD_TUNNEL6; - dev->min_mtu = IPV6_MIN_MTU; + dev->min_mtu = IPV4_MIN_MTU; dev->max_mtu = IP_MAX_MTU - sizeof(struct ipv6hdr); dev->flags |= IFF_NOARP; dev->addr_len = sizeof(struct in6_addr); diff --git a/net/ipv6/route.c b/net/ipv6/route.c index cde7d8251377..f4d61736c41a 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -1835,11 +1835,16 @@ static void ip6_multipath_l3_keys(const struct sk_buff *skb, const struct ipv6hdr *inner_iph; const struct icmp6hdr *icmph; struct ipv6hdr _inner_iph; + struct icmp6hdr _icmph; if (likely(outer_iph->nexthdr != IPPROTO_ICMPV6)) goto out; - icmph = icmp6_hdr(skb); + icmph = skb_header_pointer(skb, skb_transport_offset(skb), + sizeof(_icmph), &_icmph); + if (!icmph) + goto out; + if (icmph->icmp6_type != ICMPV6_DEST_UNREACH && icmph->icmp6_type != ICMPV6_PKT_TOOBIG && icmph->icmp6_type != ICMPV6_TIME_EXCEED && diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index 4ec76a87aeb8..ea0730028e5d 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -148,9 +148,9 @@ static int compute_score(struct sock *sk, struct net *net, bool dev_match = (sk->sk_bound_dev_if == dif || sk->sk_bound_dev_if == sdif); - if (exact_dif && !dev_match) + if (!dev_match) return -1; - if (sk->sk_bound_dev_if && dev_match) + if (sk->sk_bound_dev_if) score++; } diff --git a/net/ipv6/xfrm6_tunnel.c b/net/ipv6/xfrm6_tunnel.c index f85f0d7480ac..4a46df8441c9 100644 --- a/net/ipv6/xfrm6_tunnel.c +++ b/net/ipv6/xfrm6_tunnel.c @@ -341,6 +341,9 @@ static void __net_exit xfrm6_tunnel_net_exit(struct net *net) struct xfrm6_tunnel_net *xfrm6_tn = xfrm6_tunnel_pernet(net); unsigned int i; + xfrm_state_flush(net, IPSEC_PROTO_ANY, false); + xfrm_flush_gc(); + for (i = 0; i < XFRM6_TUNNEL_SPI_BYADDR_HSIZE; i++) WARN_ON_ONCE(!hlist_empty(&xfrm6_tn->spi_byaddr[i])); diff --git a/net/key/af_key.c b/net/key/af_key.c index 7e2e7188e7f4..e62e52e8f141 100644 --- a/net/key/af_key.c +++ b/net/key/af_key.c @@ -437,6 +437,24 @@ static int verify_address_len(const void *p) return 0; } +static inline int sadb_key_len(const struct sadb_key *key) +{ + int key_bytes = DIV_ROUND_UP(key->sadb_key_bits, 8); + + return DIV_ROUND_UP(sizeof(struct sadb_key) + key_bytes, + sizeof(uint64_t)); +} + +static int verify_key_len(const void *p) +{ + const struct sadb_key *key = p; + + if (sadb_key_len(key) > key->sadb_key_len) + return -EINVAL; + + return 0; +} + static inline int pfkey_sec_ctx_len(const struct sadb_x_sec_ctx *sec_ctx) { return DIV_ROUND_UP(sizeof(struct sadb_x_sec_ctx) + @@ -533,16 +551,25 @@ static int parse_exthdrs(struct sk_buff *skb, const struct sadb_msg *hdr, void * return -EINVAL; if (ext_hdrs[ext_type-1] != NULL) return -EINVAL; - if (ext_type == SADB_EXT_ADDRESS_SRC || - ext_type == SADB_EXT_ADDRESS_DST || - ext_type == SADB_EXT_ADDRESS_PROXY || - ext_type == SADB_X_EXT_NAT_T_OA) { + switch (ext_type) { + case SADB_EXT_ADDRESS_SRC: + case SADB_EXT_ADDRESS_DST: + case SADB_EXT_ADDRESS_PROXY: + case SADB_X_EXT_NAT_T_OA: if (verify_address_len(p)) return -EINVAL; - } - if (ext_type == SADB_X_EXT_SEC_CTX) { + break; + case SADB_X_EXT_SEC_CTX: if (verify_sec_ctx_len(p)) return -EINVAL; + break; + case SADB_EXT_KEY_AUTH: + case SADB_EXT_KEY_ENCRYPT: + if (verify_key_len(p)) + return -EINVAL; + break; + default: + break; } ext_hdrs[ext_type-1] = (void *) p; } @@ -1104,14 +1131,12 @@ static struct xfrm_state * pfkey_msg2xfrm_state(struct net *net, key = ext_hdrs[SADB_EXT_KEY_AUTH - 1]; if (key != NULL && sa->sadb_sa_auth != SADB_X_AALG_NULL && - ((key->sadb_key_bits+7) / 8 == 0 || - (key->sadb_key_bits+7) / 8 > key->sadb_key_len * sizeof(uint64_t))) + key->sadb_key_bits == 0) return ERR_PTR(-EINVAL); key = ext_hdrs[SADB_EXT_KEY_ENCRYPT-1]; if (key != NULL && sa->sadb_sa_encrypt != SADB_EALG_NULL && - ((key->sadb_key_bits+7) / 8 == 0 || - (key->sadb_key_bits+7) / 8 > key->sadb_key_len * sizeof(uint64_t))) + key->sadb_key_bits == 0) return ERR_PTR(-EINVAL); x = xfrm_state_alloc(net); diff --git a/net/llc/af_llc.c b/net/llc/af_llc.c index cb80ebb38311..1beeea9549fa 100644 --- a/net/llc/af_llc.c +++ b/net/llc/af_llc.c @@ -930,6 +930,9 @@ static int llc_ui_sendmsg(struct socket *sock, struct msghdr *msg, size_t len) if (size > llc->dev->mtu) size = llc->dev->mtu; copied = size - hdrlen; + rc = -EINVAL; + if (copied < 0) + goto release; release_sock(sk); skb = sock_alloc_send_skb(sk, size, noblock, &rc); lock_sock(sk); diff --git a/net/mac80211/agg-tx.c b/net/mac80211/agg-tx.c index 595c662a61e8..ac4295296514 100644 --- a/net/mac80211/agg-tx.c +++ b/net/mac80211/agg-tx.c @@ -8,6 +8,7 @@ * Copyright 2007, Michael Wu <flamingice@sourmilk.net> * Copyright 2007-2010, Intel Corporation * Copyright(c) 2015-2017 Intel Deutschland GmbH + * Copyright (C) 2018 Intel Corporation * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License version 2 as @@ -970,6 +971,9 @@ void ieee80211_process_addba_resp(struct ieee80211_local *local, sta->ampdu_mlme.addba_req_num[tid] = 0; + tid_tx->timeout = + le16_to_cpu(mgmt->u.action.u.addba_resp.timeout); + if (tid_tx->timeout) { mod_timer(&tid_tx->session_timer, TU_TO_EXP_TIME(tid_tx->timeout)); diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c index 69449db7e283..233068756502 100644 --- a/net/mac80211/mlme.c +++ b/net/mac80211/mlme.c @@ -36,6 +36,7 @@ #define IEEE80211_AUTH_TIMEOUT (HZ / 5) #define IEEE80211_AUTH_TIMEOUT_LONG (HZ / 2) #define IEEE80211_AUTH_TIMEOUT_SHORT (HZ / 10) +#define IEEE80211_AUTH_TIMEOUT_SAE (HZ * 2) #define IEEE80211_AUTH_MAX_TRIES 3 #define IEEE80211_AUTH_WAIT_ASSOC (HZ * 5) #define IEEE80211_ASSOC_TIMEOUT (HZ / 5) @@ -1787,7 +1788,7 @@ static bool ieee80211_sta_wmm_params(struct ieee80211_local *local, params[ac].acm = acm; params[ac].uapsd = uapsd; - if (params->cw_min == 0 || + if (params[ac].cw_min == 0 || params[ac].cw_min > params[ac].cw_max) { sdata_info(sdata, "AP has invalid WMM params (CWmin/max=%d/%d for ACI %d), using defaults\n", @@ -3814,16 +3815,19 @@ static int ieee80211_auth(struct ieee80211_sub_if_data *sdata) tx_flags); if (tx_flags == 0) { - auth_data->timeout = jiffies + IEEE80211_AUTH_TIMEOUT; - auth_data->timeout_started = true; - run_again(sdata, auth_data->timeout); + if (auth_data->algorithm == WLAN_AUTH_SAE) + auth_data->timeout = jiffies + + IEEE80211_AUTH_TIMEOUT_SAE; + else + auth_data->timeout = jiffies + IEEE80211_AUTH_TIMEOUT; } else { auth_data->timeout = round_jiffies_up(jiffies + IEEE80211_AUTH_TIMEOUT_LONG); - auth_data->timeout_started = true; - run_again(sdata, auth_data->timeout); } + auth_data->timeout_started = true; + run_again(sdata, auth_data->timeout); + return 0; } @@ -3894,8 +3898,15 @@ void ieee80211_sta_work(struct ieee80211_sub_if_data *sdata) ifmgd->status_received = false; if (ifmgd->auth_data && ieee80211_is_auth(fc)) { if (status_acked) { - ifmgd->auth_data->timeout = - jiffies + IEEE80211_AUTH_TIMEOUT_SHORT; + if (ifmgd->auth_data->algorithm == + WLAN_AUTH_SAE) + ifmgd->auth_data->timeout = + jiffies + + IEEE80211_AUTH_TIMEOUT_SAE; + else + ifmgd->auth_data->timeout = + jiffies + + IEEE80211_AUTH_TIMEOUT_SHORT; run_again(sdata, ifmgd->auth_data->timeout); } else { ifmgd->auth_data->timeout = jiffies - 1; diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c index 535de3161a78..05a265cd573d 100644 --- a/net/mac80211/tx.c +++ b/net/mac80211/tx.c @@ -4,6 +4,7 @@ * Copyright 2006-2007 Jiri Benc <jbenc@suse.cz> * Copyright 2007 Johannes Berg <johannes@sipsolutions.net> * Copyright 2013-2014 Intel Mobile Communications GmbH + * Copyright (C) 2018 Intel Corporation * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License version 2 as @@ -1135,7 +1136,7 @@ static bool ieee80211_tx_prep_agg(struct ieee80211_tx_data *tx, } /* reset session timer */ - if (reset_agg_timer && tid_tx->timeout) + if (reset_agg_timer) tid_tx->last_tx = jiffies; return queued; diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index 55342c4d5cec..2e2dd88fc79f 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -2606,13 +2606,13 @@ static int netlink_seq_show(struct seq_file *seq, void *v) { if (v == SEQ_START_TOKEN) { seq_puts(seq, - "sk Eth Pid Groups " - "Rmem Wmem Dump Locks Drops Inode\n"); + "sk Eth Pid Groups " + "Rmem Wmem Dump Locks Drops Inode\n"); } else { struct sock *s = v; struct netlink_sock *nlk = nlk_sk(s); - seq_printf(seq, "%pK %-3d %-6u %08x %-8d %-8d %d %-8d %-8d %-8lu\n", + seq_printf(seq, "%pK %-3d %-10u %08x %-8d %-8d %-5d %-8d %-8d %-8lu\n", s, s->sk_protocol, nlk->portid, diff --git a/net/nsh/nsh.c b/net/nsh/nsh.c index d7da99a0b0b8..9696ef96b719 100644 --- a/net/nsh/nsh.c +++ b/net/nsh/nsh.c @@ -57,6 +57,8 @@ int nsh_pop(struct sk_buff *skb) return -ENOMEM; nh = (struct nshhdr *)(skb->data); length = nsh_hdr_len(nh); + if (length < NSH_BASE_HDR_LEN) + return -EINVAL; inner_proto = tun_p_to_eth_p(nh->np); if (!pskb_may_pull(skb, length)) return -ENOMEM; @@ -90,6 +92,8 @@ static struct sk_buff *nsh_gso_segment(struct sk_buff *skb, if (unlikely(!pskb_may_pull(skb, NSH_BASE_HDR_LEN))) goto out; nsh_len = nsh_hdr_len(nsh_hdr(skb)); + if (nsh_len < NSH_BASE_HDR_LEN) + goto out; if (unlikely(!pskb_may_pull(skb, nsh_len))) goto out; diff --git a/net/openvswitch/flow_netlink.c b/net/openvswitch/flow_netlink.c index 7322aa1e382e..492ab0c36f7c 100644 --- a/net/openvswitch/flow_netlink.c +++ b/net/openvswitch/flow_netlink.c @@ -1712,13 +1712,10 @@ static void nlattr_set(struct nlattr *attr, u8 val, /* The nlattr stream should already have been validated */ nla_for_each_nested(nla, attr, rem) { - if (tbl[nla_type(nla)].len == OVS_ATTR_NESTED) { - if (tbl[nla_type(nla)].next) - tbl = tbl[nla_type(nla)].next; - nlattr_set(nla, val, tbl); - } else { + if (tbl[nla_type(nla)].len == OVS_ATTR_NESTED) + nlattr_set(nla, val, tbl[nla_type(nla)].next ? : tbl); + else memset(nla_data(nla), val, nla_len(nla)); - } if (nla_type(nla) == OVS_KEY_ATTR_CT_STATE) *(u32 *)nla_data(nla) &= CT_SUPPORTED_MASK; diff --git a/net/rds/ib_cm.c b/net/rds/ib_cm.c index eea1d8611b20..13b38ad0fa4a 100644 --- a/net/rds/ib_cm.c +++ b/net/rds/ib_cm.c @@ -547,7 +547,7 @@ static int rds_ib_setup_qp(struct rds_connection *conn) rdsdebug("conn %p pd %p cq %p %p\n", conn, ic->i_pd, ic->i_send_cq, ic->i_recv_cq); - return ret; + goto out; sends_out: vfree(ic->i_sends); @@ -572,6 +572,7 @@ send_cq_out: ic->i_send_cq = NULL; rds_ibdev_out: rds_ib_remove_conn(rds_ibdev, conn); +out: rds_ib_dev_put(rds_ibdev); return ret; diff --git a/net/rds/recv.c b/net/rds/recv.c index de50e2126e40..dc67458b52f0 100644 --- a/net/rds/recv.c +++ b/net/rds/recv.c @@ -558,6 +558,7 @@ static int rds_cmsg_recv(struct rds_incoming *inc, struct msghdr *msg, struct rds_cmsg_rx_trace t; int i, j; + memset(&t, 0, sizeof(t)); inc->i_rx_lat_trace[RDS_MSG_RX_CMSG] = local_clock(); t.rx_traces = rs->rs_rx_traces; for (i = 0; i < rs->rs_rx_traces; i++) { diff --git a/net/rfkill/rfkill-gpio.c b/net/rfkill/rfkill-gpio.c index 41bd496531d4..00192a996be0 100644 --- a/net/rfkill/rfkill-gpio.c +++ b/net/rfkill/rfkill-gpio.c @@ -137,13 +137,18 @@ static int rfkill_gpio_probe(struct platform_device *pdev) ret = rfkill_register(rfkill->rfkill_dev); if (ret < 0) - return ret; + goto err_destroy; platform_set_drvdata(pdev, rfkill); dev_info(&pdev->dev, "%s device registered.\n", rfkill->name); return 0; + +err_destroy: + rfkill_destroy(rfkill->rfkill_dev); + + return ret; } static int rfkill_gpio_remove(struct platform_device *pdev) diff --git a/net/rxrpc/af_rxrpc.c b/net/rxrpc/af_rxrpc.c index 9a2c8e7c000e..2b463047dd7b 100644 --- a/net/rxrpc/af_rxrpc.c +++ b/net/rxrpc/af_rxrpc.c @@ -313,7 +313,7 @@ struct rxrpc_call *rxrpc_kernel_begin_call(struct socket *sock, memset(&cp, 0, sizeof(cp)); cp.local = rx->local; cp.key = key; - cp.security_level = 0; + cp.security_level = rx->min_sec_level; cp.exclusive = false; cp.upgrade = upgrade; cp.service_id = srx->srx_service; diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h index 90d7079e0aa9..19975d2ca9a2 100644 --- a/net/rxrpc/ar-internal.h +++ b/net/rxrpc/ar-internal.h @@ -476,6 +476,7 @@ enum rxrpc_call_flag { RXRPC_CALL_SEND_PING, /* A ping will need to be sent */ RXRPC_CALL_PINGING, /* Ping in process */ RXRPC_CALL_RETRANS_TIMEOUT, /* Retransmission due to timeout occurred */ + RXRPC_CALL_BEGAN_RX_TIMER, /* We began the expect_rx_by timer */ }; /* diff --git a/net/rxrpc/conn_event.c b/net/rxrpc/conn_event.c index c717152070df..1350f1be8037 100644 --- a/net/rxrpc/conn_event.c +++ b/net/rxrpc/conn_event.c @@ -40,7 +40,7 @@ static void rxrpc_conn_retransmit_call(struct rxrpc_connection *conn, } __attribute__((packed)) pkt; struct rxrpc_ackinfo ack_info; size_t len; - int ioc; + int ret, ioc; u32 serial, mtu, call_id, padding; _enter("%d", conn->debug_id); @@ -135,10 +135,13 @@ static void rxrpc_conn_retransmit_call(struct rxrpc_connection *conn, break; } - kernel_sendmsg(conn->params.local->socket, &msg, iov, ioc, len); + ret = kernel_sendmsg(conn->params.local->socket, &msg, iov, ioc, len); conn->params.peer->last_tx_at = ktime_get_real(); + if (ret < 0) + trace_rxrpc_tx_fail(conn->debug_id, serial, ret, + rxrpc_tx_fail_call_final_resend); + _leave(""); - return; } /* @@ -236,6 +239,8 @@ static int rxrpc_abort_connection(struct rxrpc_connection *conn, ret = kernel_sendmsg(conn->params.local->socket, &msg, iov, 2, len); if (ret < 0) { + trace_rxrpc_tx_fail(conn->debug_id, serial, ret, + rxrpc_tx_fail_conn_abort); _debug("sendmsg failed: %d", ret); return -EAGAIN; } diff --git a/net/rxrpc/input.c b/net/rxrpc/input.c index 0410d2277ca2..b5fd6381313d 100644 --- a/net/rxrpc/input.c +++ b/net/rxrpc/input.c @@ -971,7 +971,7 @@ static void rxrpc_input_call_packet(struct rxrpc_call *call, if (timo) { unsigned long now = jiffies, expect_rx_by; - expect_rx_by = jiffies + timo; + expect_rx_by = now + timo; WRITE_ONCE(call->expect_rx_by, expect_rx_by); rxrpc_reduce_call_timer(call, expect_rx_by, now, rxrpc_timer_set_for_normal); diff --git a/net/rxrpc/local_event.c b/net/rxrpc/local_event.c index 93b5d910b4a1..8325f1b86840 100644 --- a/net/rxrpc/local_event.c +++ b/net/rxrpc/local_event.c @@ -71,7 +71,8 @@ static void rxrpc_send_version_request(struct rxrpc_local *local, ret = kernel_sendmsg(local->socket, &msg, iov, 2, len); if (ret < 0) - _debug("sendmsg failed: %d", ret); + trace_rxrpc_tx_fail(local->debug_id, 0, ret, + rxrpc_tx_fail_version_reply); _leave(""); } diff --git a/net/rxrpc/local_object.c b/net/rxrpc/local_object.c index 8b54e9531d52..b493e6b62740 100644 --- a/net/rxrpc/local_object.c +++ b/net/rxrpc/local_object.c @@ -134,22 +134,49 @@ static int rxrpc_open_socket(struct rxrpc_local *local, struct net *net) } } - /* we want to receive ICMP errors */ - opt = 1; - ret = kernel_setsockopt(local->socket, SOL_IP, IP_RECVERR, - (char *) &opt, sizeof(opt)); - if (ret < 0) { - _debug("setsockopt failed"); - goto error; - } + switch (local->srx.transport.family) { + case AF_INET: + /* we want to receive ICMP errors */ + opt = 1; + ret = kernel_setsockopt(local->socket, SOL_IP, IP_RECVERR, + (char *) &opt, sizeof(opt)); + if (ret < 0) { + _debug("setsockopt failed"); + goto error; + } - /* we want to set the don't fragment bit */ - opt = IP_PMTUDISC_DO; - ret = kernel_setsockopt(local->socket, SOL_IP, IP_MTU_DISCOVER, - (char *) &opt, sizeof(opt)); - if (ret < 0) { - _debug("setsockopt failed"); - goto error; + /* we want to set the don't fragment bit */ + opt = IP_PMTUDISC_DO; + ret = kernel_setsockopt(local->socket, SOL_IP, IP_MTU_DISCOVER, + (char *) &opt, sizeof(opt)); + if (ret < 0) { + _debug("setsockopt failed"); + goto error; + } + break; + + case AF_INET6: + /* we want to receive ICMP errors */ + opt = 1; + ret = kernel_setsockopt(local->socket, SOL_IPV6, IPV6_RECVERR, + (char *) &opt, sizeof(opt)); + if (ret < 0) { + _debug("setsockopt failed"); + goto error; + } + + /* we want to set the don't fragment bit */ + opt = IPV6_PMTUDISC_DO; + ret = kernel_setsockopt(local->socket, SOL_IPV6, IPV6_MTU_DISCOVER, + (char *) &opt, sizeof(opt)); + if (ret < 0) { + _debug("setsockopt failed"); + goto error; + } + break; + + default: + BUG(); } /* set the socket up */ diff --git a/net/rxrpc/output.c b/net/rxrpc/output.c index 7f1fc04775b3..f03de1c59ba3 100644 --- a/net/rxrpc/output.c +++ b/net/rxrpc/output.c @@ -210,6 +210,9 @@ int rxrpc_send_ack_packet(struct rxrpc_call *call, bool ping, if (ping) call->ping_time = now; conn->params.peer->last_tx_at = ktime_get_real(); + if (ret < 0) + trace_rxrpc_tx_fail(call->debug_id, serial, ret, + rxrpc_tx_fail_call_ack); if (call->state < RXRPC_CALL_COMPLETE) { if (ret < 0) { @@ -294,6 +297,10 @@ int rxrpc_send_abort_packet(struct rxrpc_call *call) ret = kernel_sendmsg(conn->params.local->socket, &msg, iov, 1, sizeof(pkt)); conn->params.peer->last_tx_at = ktime_get_real(); + if (ret < 0) + trace_rxrpc_tx_fail(call->debug_id, serial, ret, + rxrpc_tx_fail_call_abort); + rxrpc_put_connection(conn); return ret; @@ -387,6 +394,9 @@ int rxrpc_send_data_packet(struct rxrpc_call *call, struct sk_buff *skb, conn->params.peer->last_tx_at = ktime_get_real(); up_read(&conn->params.local->defrag_sem); + if (ret < 0) + trace_rxrpc_tx_fail(call->debug_id, serial, ret, + rxrpc_tx_fail_call_data_nofrag); if (ret == -EMSGSIZE) goto send_fragmentable; @@ -414,6 +424,17 @@ done: rxrpc_timer_set_for_lost_ack); } } + + if (sp->hdr.seq == 1 && + !test_and_set_bit(RXRPC_CALL_BEGAN_RX_TIMER, + &call->flags)) { + unsigned long nowj = jiffies, expect_rx_by; + + expect_rx_by = nowj + call->next_rx_timo; + WRITE_ONCE(call->expect_rx_by, expect_rx_by); + rxrpc_reduce_call_timer(call, expect_rx_by, nowj, + rxrpc_timer_set_for_normal); + } } rxrpc_set_keepalive(call); @@ -465,6 +486,10 @@ send_fragmentable: #endif } + if (ret < 0) + trace_rxrpc_tx_fail(call->debug_id, serial, ret, + rxrpc_tx_fail_call_data_frag); + up_write(&conn->params.local->defrag_sem); goto done; } @@ -482,6 +507,7 @@ void rxrpc_reject_packets(struct rxrpc_local *local) struct kvec iov[2]; size_t size; __be32 code; + int ret; _enter("%d", local->debug_id); @@ -516,7 +542,10 @@ void rxrpc_reject_packets(struct rxrpc_local *local) whdr.flags ^= RXRPC_CLIENT_INITIATED; whdr.flags &= RXRPC_CLIENT_INITIATED; - kernel_sendmsg(local->socket, &msg, iov, 2, size); + ret = kernel_sendmsg(local->socket, &msg, iov, 2, size); + if (ret < 0) + trace_rxrpc_tx_fail(local->debug_id, 0, ret, + rxrpc_tx_fail_reject); } rxrpc_free_skb(skb, rxrpc_skb_rx_freed); @@ -567,7 +596,8 @@ void rxrpc_send_keepalive(struct rxrpc_peer *peer) ret = kernel_sendmsg(peer->local->socket, &msg, iov, 2, len); if (ret < 0) - _debug("sendmsg failed: %d", ret); + trace_rxrpc_tx_fail(peer->debug_id, 0, ret, + rxrpc_tx_fail_version_keepalive); peer->last_tx_at = ktime_get_real(); _leave(""); diff --git a/net/rxrpc/peer_event.c b/net/rxrpc/peer_event.c index 78c2f95d1f22..0ed8b651cec2 100644 --- a/net/rxrpc/peer_event.c +++ b/net/rxrpc/peer_event.c @@ -28,39 +28,39 @@ static void rxrpc_store_error(struct rxrpc_peer *, struct sock_exterr_skb *); * Find the peer associated with an ICMP packet. */ static struct rxrpc_peer *rxrpc_lookup_peer_icmp_rcu(struct rxrpc_local *local, - const struct sk_buff *skb) + const struct sk_buff *skb, + struct sockaddr_rxrpc *srx) { struct sock_exterr_skb *serr = SKB_EXT_ERR(skb); - struct sockaddr_rxrpc srx; _enter(""); - memset(&srx, 0, sizeof(srx)); - srx.transport_type = local->srx.transport_type; - srx.transport_len = local->srx.transport_len; - srx.transport.family = local->srx.transport.family; + memset(srx, 0, sizeof(*srx)); + srx->transport_type = local->srx.transport_type; + srx->transport_len = local->srx.transport_len; + srx->transport.family = local->srx.transport.family; /* Can we see an ICMP4 packet on an ICMP6 listening socket? and vice * versa? */ - switch (srx.transport.family) { + switch (srx->transport.family) { case AF_INET: - srx.transport.sin.sin_port = serr->port; + srx->transport.sin.sin_port = serr->port; switch (serr->ee.ee_origin) { case SO_EE_ORIGIN_ICMP: _net("Rx ICMP"); - memcpy(&srx.transport.sin.sin_addr, + memcpy(&srx->transport.sin.sin_addr, skb_network_header(skb) + serr->addr_offset, sizeof(struct in_addr)); break; case SO_EE_ORIGIN_ICMP6: _net("Rx ICMP6 on v4 sock"); - memcpy(&srx.transport.sin.sin_addr, + memcpy(&srx->transport.sin.sin_addr, skb_network_header(skb) + serr->addr_offset + 12, sizeof(struct in_addr)); break; default: - memcpy(&srx.transport.sin.sin_addr, &ip_hdr(skb)->saddr, + memcpy(&srx->transport.sin.sin_addr, &ip_hdr(skb)->saddr, sizeof(struct in_addr)); break; } @@ -68,25 +68,25 @@ static struct rxrpc_peer *rxrpc_lookup_peer_icmp_rcu(struct rxrpc_local *local, #ifdef CONFIG_AF_RXRPC_IPV6 case AF_INET6: - srx.transport.sin6.sin6_port = serr->port; + srx->transport.sin6.sin6_port = serr->port; switch (serr->ee.ee_origin) { case SO_EE_ORIGIN_ICMP6: _net("Rx ICMP6"); - memcpy(&srx.transport.sin6.sin6_addr, + memcpy(&srx->transport.sin6.sin6_addr, skb_network_header(skb) + serr->addr_offset, sizeof(struct in6_addr)); break; case SO_EE_ORIGIN_ICMP: _net("Rx ICMP on v6 sock"); - srx.transport.sin6.sin6_addr.s6_addr32[0] = 0; - srx.transport.sin6.sin6_addr.s6_addr32[1] = 0; - srx.transport.sin6.sin6_addr.s6_addr32[2] = htonl(0xffff); - memcpy(srx.transport.sin6.sin6_addr.s6_addr + 12, + srx->transport.sin6.sin6_addr.s6_addr32[0] = 0; + srx->transport.sin6.sin6_addr.s6_addr32[1] = 0; + srx->transport.sin6.sin6_addr.s6_addr32[2] = htonl(0xffff); + memcpy(srx->transport.sin6.sin6_addr.s6_addr + 12, skb_network_header(skb) + serr->addr_offset, sizeof(struct in_addr)); break; default: - memcpy(&srx.transport.sin6.sin6_addr, + memcpy(&srx->transport.sin6.sin6_addr, &ipv6_hdr(skb)->saddr, sizeof(struct in6_addr)); break; @@ -98,7 +98,7 @@ static struct rxrpc_peer *rxrpc_lookup_peer_icmp_rcu(struct rxrpc_local *local, BUG(); } - return rxrpc_lookup_peer_rcu(local, &srx); + return rxrpc_lookup_peer_rcu(local, srx); } /* @@ -146,6 +146,7 @@ static void rxrpc_adjust_mtu(struct rxrpc_peer *peer, struct sock_exterr_skb *se void rxrpc_error_report(struct sock *sk) { struct sock_exterr_skb *serr; + struct sockaddr_rxrpc srx; struct rxrpc_local *local = sk->sk_user_data; struct rxrpc_peer *peer; struct sk_buff *skb; @@ -166,7 +167,7 @@ void rxrpc_error_report(struct sock *sk) } rcu_read_lock(); - peer = rxrpc_lookup_peer_icmp_rcu(local, skb); + peer = rxrpc_lookup_peer_icmp_rcu(local, skb, &srx); if (peer && !rxrpc_get_peer_maybe(peer)) peer = NULL; if (!peer) { @@ -176,6 +177,8 @@ void rxrpc_error_report(struct sock *sk) return; } + trace_rxrpc_rx_icmp(peer, &serr->ee, &srx); + if ((serr->ee.ee_origin == SO_EE_ORIGIN_ICMP && serr->ee.ee_type == ICMP_DEST_UNREACH && serr->ee.ee_code == ICMP_FRAG_NEEDED)) { @@ -209,9 +212,6 @@ static void rxrpc_store_error(struct rxrpc_peer *peer, ee = &serr->ee; - _net("Rx Error o=%d t=%d c=%d e=%d", - ee->ee_origin, ee->ee_type, ee->ee_code, ee->ee_errno); - err = ee->ee_errno; switch (ee->ee_origin) { diff --git a/net/rxrpc/rxkad.c b/net/rxrpc/rxkad.c index 588fea0dd362..6c0ae27fff84 100644 --- a/net/rxrpc/rxkad.c +++ b/net/rxrpc/rxkad.c @@ -664,7 +664,8 @@ static int rxkad_issue_challenge(struct rxrpc_connection *conn) ret = kernel_sendmsg(conn->params.local->socket, &msg, iov, 2, len); if (ret < 0) { - _debug("sendmsg failed: %d", ret); + trace_rxrpc_tx_fail(conn->debug_id, serial, ret, + rxrpc_tx_fail_conn_challenge); return -EAGAIN; } @@ -719,7 +720,8 @@ static int rxkad_send_response(struct rxrpc_connection *conn, ret = kernel_sendmsg(conn->params.local->socket, &msg, iov, 3, len); if (ret < 0) { - _debug("sendmsg failed: %d", ret); + trace_rxrpc_tx_fail(conn->debug_id, serial, ret, + rxrpc_tx_fail_conn_response); return -EAGAIN; } diff --git a/net/rxrpc/sendmsg.c b/net/rxrpc/sendmsg.c index 206e802ccbdc..be01f9c5d963 100644 --- a/net/rxrpc/sendmsg.c +++ b/net/rxrpc/sendmsg.c @@ -223,6 +223,15 @@ static void rxrpc_queue_packet(struct rxrpc_sock *rx, struct rxrpc_call *call, ret = rxrpc_send_data_packet(call, skb, false); if (ret < 0) { + switch (ret) { + case -ENETUNREACH: + case -EHOSTUNREACH: + case -ECONNREFUSED: + rxrpc_set_call_completion(call, + RXRPC_CALL_LOCAL_ERROR, + 0, ret); + goto out; + } _debug("need instant resend %d", ret); rxrpc_instant_resend(call, ix); } else { @@ -241,6 +250,7 @@ static void rxrpc_queue_packet(struct rxrpc_sock *rx, struct rxrpc_call *call, rxrpc_timer_set_for_send); } +out: rxrpc_free_skb(skb, rxrpc_skb_tx_freed); _leave(""); } diff --git a/net/sched/act_skbedit.c b/net/sched/act_skbedit.c index ddf69fc01bdf..6138d1d71900 100644 --- a/net/sched/act_skbedit.c +++ b/net/sched/act_skbedit.c @@ -121,7 +121,8 @@ static int tcf_skbedit_init(struct net *net, struct nlattr *nla, return 0; if (!flags) { - tcf_idr_release(*a, bind); + if (exists) + tcf_idr_release(*a, bind); return -EINVAL; } diff --git a/net/sched/act_skbmod.c b/net/sched/act_skbmod.c index bbcbdce732cc..ad050d7d4b46 100644 --- a/net/sched/act_skbmod.c +++ b/net/sched/act_skbmod.c @@ -131,8 +131,11 @@ static int tcf_skbmod_init(struct net *net, struct nlattr *nla, if (exists && bind) return 0; - if (!lflags) + if (!lflags) { + if (exists) + tcf_idr_release(*a, bind); return -EINVAL; + } if (!exists) { ret = tcf_idr_create(tn, parm->index, est, a, diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c index b66754f52a9f..963e4bf0aab8 100644 --- a/net/sched/cls_api.c +++ b/net/sched/cls_api.c @@ -152,8 +152,8 @@ static struct tcf_proto *tcf_proto_create(const char *kind, u32 protocol, NL_SET_ERR_MSG(extack, "TC classifier not found"); err = -ENOENT; } - goto errout; #endif + goto errout; } tp->classify = tp->ops->classify; tp->protocol = protocol; diff --git a/net/sched/sch_fq.c b/net/sched/sch_fq.c index a366e4c9413a..4808713c73b9 100644 --- a/net/sched/sch_fq.c +++ b/net/sched/sch_fq.c @@ -128,6 +128,28 @@ static bool fq_flow_is_detached(const struct fq_flow *f) return f->next == &detached; } +static bool fq_flow_is_throttled(const struct fq_flow *f) +{ + return f->next == &throttled; +} + +static void fq_flow_add_tail(struct fq_flow_head *head, struct fq_flow *flow) +{ + if (head->first) + head->last->next = flow; + else + head->first = flow; + head->last = flow; + flow->next = NULL; +} + +static void fq_flow_unset_throttled(struct fq_sched_data *q, struct fq_flow *f) +{ + rb_erase(&f->rate_node, &q->delayed); + q->throttled_flows--; + fq_flow_add_tail(&q->old_flows, f); +} + static void fq_flow_set_throttled(struct fq_sched_data *q, struct fq_flow *f) { struct rb_node **p = &q->delayed.rb_node, *parent = NULL; @@ -155,15 +177,6 @@ static void fq_flow_set_throttled(struct fq_sched_data *q, struct fq_flow *f) static struct kmem_cache *fq_flow_cachep __read_mostly; -static void fq_flow_add_tail(struct fq_flow_head *head, struct fq_flow *flow) -{ - if (head->first) - head->last->next = flow; - else - head->first = flow; - head->last = flow; - flow->next = NULL; -} /* limit number of collected flows per round */ #define FQ_GC_MAX 8 @@ -267,6 +280,8 @@ static struct fq_flow *fq_classify(struct sk_buff *skb, struct fq_sched_data *q) f->socket_hash != sk->sk_hash)) { f->credit = q->initial_quantum; f->socket_hash = sk->sk_hash; + if (fq_flow_is_throttled(f)) + fq_flow_unset_throttled(q, f); f->time_next_packet = 0ULL; } return f; @@ -438,9 +453,7 @@ static void fq_check_throttled(struct fq_sched_data *q, u64 now) q->time_next_delayed_flow = f->time_next_packet; break; } - rb_erase(p, &q->delayed); - q->throttled_flows--; - fq_flow_add_tail(&q->old_flows, f); + fq_flow_unset_throttled(q, f); } } diff --git a/net/sctp/associola.c b/net/sctp/associola.c index 837806dd5799..a47179da24e6 100644 --- a/net/sctp/associola.c +++ b/net/sctp/associola.c @@ -1024,8 +1024,9 @@ static void sctp_assoc_bh_rcv(struct work_struct *work) struct sctp_endpoint *ep; struct sctp_chunk *chunk; struct sctp_inq *inqueue; - int state; + int first_time = 1; /* is this the first time through the loop */ int error = 0; + int state; /* The association should be held so we should be safe. */ ep = asoc->ep; @@ -1036,6 +1037,30 @@ static void sctp_assoc_bh_rcv(struct work_struct *work) state = asoc->state; subtype = SCTP_ST_CHUNK(chunk->chunk_hdr->type); + /* If the first chunk in the packet is AUTH, do special + * processing specified in Section 6.3 of SCTP-AUTH spec + */ + if (first_time && subtype.chunk == SCTP_CID_AUTH) { + struct sctp_chunkhdr *next_hdr; + + next_hdr = sctp_inq_peek(inqueue); + if (!next_hdr) + goto normal; + + /* If the next chunk is COOKIE-ECHO, skip the AUTH + * chunk while saving a pointer to it so we can do + * Authentication later (during cookie-echo + * processing). + */ + if (next_hdr->type == SCTP_CID_COOKIE_ECHO) { + chunk->auth_chunk = skb_clone(chunk->skb, + GFP_ATOMIC); + chunk->auth = 1; + continue; + } + } + +normal: /* SCTP-AUTH, Section 6.3: * The receiver has a list of chunk types which it expects * to be received only after an AUTH-chunk. This list has @@ -1074,6 +1099,9 @@ static void sctp_assoc_bh_rcv(struct work_struct *work) /* If there is an error on chunk, discard this packet. */ if (error && chunk) chunk->pdiscard = 1; + + if (first_time) + first_time = 0; } sctp_association_put(asoc); } diff --git a/net/sctp/inqueue.c b/net/sctp/inqueue.c index 23ebc5318edc..eb93ffe2408b 100644 --- a/net/sctp/inqueue.c +++ b/net/sctp/inqueue.c @@ -217,7 +217,7 @@ new_skb: skb_pull(chunk->skb, sizeof(*ch)); chunk->subh.v = NULL; /* Subheader is no longer valid. */ - if (chunk->chunk_end + sizeof(*ch) < skb_tail_pointer(chunk->skb)) { + if (chunk->chunk_end + sizeof(*ch) <= skb_tail_pointer(chunk->skb)) { /* This is not a singleton */ chunk->singleton = 0; } else if (chunk->chunk_end > skb_tail_pointer(chunk->skb)) { diff --git a/net/sctp/ipv6.c b/net/sctp/ipv6.c index 2e3f7b75a8ec..42247110d842 100644 --- a/net/sctp/ipv6.c +++ b/net/sctp/ipv6.c @@ -895,6 +895,9 @@ static int sctp_inet6_cmp_addr(const union sctp_addr *addr1, if (sctp_is_any(sk, addr1) || sctp_is_any(sk, addr2)) return 1; + if (addr1->sa.sa_family == AF_INET && addr2->sa.sa_family == AF_INET) + return addr1->v4.sin_addr.s_addr == addr2->v4.sin_addr.s_addr; + return __sctp_v6_cmp_addr(addr1, addr2); } diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c index 5a4fb1dc8400..e62addb60434 100644 --- a/net/sctp/sm_make_chunk.c +++ b/net/sctp/sm_make_chunk.c @@ -1152,7 +1152,7 @@ struct sctp_chunk *sctp_make_violation_max_retrans( const struct sctp_association *asoc, const struct sctp_chunk *chunk) { - static const char error[] = "Association exceeded its max_retans count"; + static const char error[] = "Association exceeded its max_retrans count"; size_t payload_len = sizeof(error) + sizeof(struct sctp_errhdr); struct sctp_chunk *retval; diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c index dd0594a10961..c9ae3404b1bb 100644 --- a/net/sctp/sm_statefuns.c +++ b/net/sctp/sm_statefuns.c @@ -153,10 +153,7 @@ static enum sctp_disposition sctp_sf_violation_chunk( struct sctp_cmd_seq *commands); static enum sctp_ierror sctp_sf_authenticate( - struct net *net, - const struct sctp_endpoint *ep, const struct sctp_association *asoc, - const union sctp_subtype type, struct sctp_chunk *chunk); static enum sctp_disposition __sctp_sf_do_9_1_abort( @@ -626,6 +623,38 @@ enum sctp_disposition sctp_sf_do_5_1C_ack(struct net *net, return SCTP_DISPOSITION_CONSUME; } +static bool sctp_auth_chunk_verify(struct net *net, struct sctp_chunk *chunk, + const struct sctp_association *asoc) +{ + struct sctp_chunk auth; + + if (!chunk->auth_chunk) + return true; + + /* SCTP-AUTH: auth_chunk pointer is only set when the cookie-echo + * is supposed to be authenticated and we have to do delayed + * authentication. We've just recreated the association using + * the information in the cookie and now it's much easier to + * do the authentication. + */ + + /* Make sure that we and the peer are AUTH capable */ + if (!net->sctp.auth_enable || !asoc->peer.auth_capable) + return false; + + /* set-up our fake chunk so that we can process it */ + auth.skb = chunk->auth_chunk; + auth.asoc = chunk->asoc; + auth.sctp_hdr = chunk->sctp_hdr; + auth.chunk_hdr = (struct sctp_chunkhdr *) + skb_push(chunk->auth_chunk, + sizeof(struct sctp_chunkhdr)); + skb_pull(chunk->auth_chunk, sizeof(struct sctp_chunkhdr)); + auth.transport = chunk->transport; + + return sctp_sf_authenticate(asoc, &auth) == SCTP_IERROR_NO_ERROR; +} + /* * Respond to a normal COOKIE ECHO chunk. * We are the side that is being asked for an association. @@ -763,37 +792,9 @@ enum sctp_disposition sctp_sf_do_5_1D_ce(struct net *net, if (error) goto nomem_init; - /* SCTP-AUTH: auth_chunk pointer is only set when the cookie-echo - * is supposed to be authenticated and we have to do delayed - * authentication. We've just recreated the association using - * the information in the cookie and now it's much easier to - * do the authentication. - */ - if (chunk->auth_chunk) { - struct sctp_chunk auth; - enum sctp_ierror ret; - - /* Make sure that we and the peer are AUTH capable */ - if (!net->sctp.auth_enable || !new_asoc->peer.auth_capable) { - sctp_association_free(new_asoc); - return sctp_sf_pdiscard(net, ep, asoc, type, arg, commands); - } - - /* set-up our fake chunk so that we can process it */ - auth.skb = chunk->auth_chunk; - auth.asoc = chunk->asoc; - auth.sctp_hdr = chunk->sctp_hdr; - auth.chunk_hdr = (struct sctp_chunkhdr *) - skb_push(chunk->auth_chunk, - sizeof(struct sctp_chunkhdr)); - skb_pull(chunk->auth_chunk, sizeof(struct sctp_chunkhdr)); - auth.transport = chunk->transport; - - ret = sctp_sf_authenticate(net, ep, new_asoc, type, &auth); - if (ret != SCTP_IERROR_NO_ERROR) { - sctp_association_free(new_asoc); - return sctp_sf_pdiscard(net, ep, asoc, type, arg, commands); - } + if (!sctp_auth_chunk_verify(net, chunk, new_asoc)) { + sctp_association_free(new_asoc); + return sctp_sf_pdiscard(net, ep, asoc, type, arg, commands); } repl = sctp_make_cookie_ack(new_asoc, chunk); @@ -1794,13 +1795,18 @@ static enum sctp_disposition sctp_sf_do_dupcook_a( GFP_ATOMIC)) goto nomem; + if (sctp_auth_asoc_init_active_key(new_asoc, GFP_ATOMIC)) + goto nomem; + + if (!sctp_auth_chunk_verify(net, chunk, new_asoc)) + return SCTP_DISPOSITION_DISCARD; + /* Make sure no new addresses are being added during the * restart. Though this is a pretty complicated attack * since you'd have to get inside the cookie. */ - if (!sctp_sf_check_restart_addrs(new_asoc, asoc, chunk, commands)) { + if (!sctp_sf_check_restart_addrs(new_asoc, asoc, chunk, commands)) return SCTP_DISPOSITION_CONSUME; - } /* If the endpoint is in the SHUTDOWN-ACK-SENT state and recognizes * the peer has restarted (Action A), it MUST NOT setup a new @@ -1906,6 +1912,12 @@ static enum sctp_disposition sctp_sf_do_dupcook_b( GFP_ATOMIC)) goto nomem; + if (sctp_auth_asoc_init_active_key(new_asoc, GFP_ATOMIC)) + goto nomem; + + if (!sctp_auth_chunk_verify(net, chunk, new_asoc)) + return SCTP_DISPOSITION_DISCARD; + /* Update the content of current association. */ sctp_add_cmd_sf(commands, SCTP_CMD_UPDATE_ASSOC, SCTP_ASOC(new_asoc)); sctp_add_cmd_sf(commands, SCTP_CMD_NEW_STATE, @@ -2003,6 +2015,9 @@ static enum sctp_disposition sctp_sf_do_dupcook_d( * a COOKIE ACK. */ + if (!sctp_auth_chunk_verify(net, chunk, asoc)) + return SCTP_DISPOSITION_DISCARD; + /* Don't accidentally move back into established state. */ if (asoc->state < SCTP_STATE_ESTABLISHED) { sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_STOP, @@ -2050,7 +2065,7 @@ static enum sctp_disposition sctp_sf_do_dupcook_d( } } - repl = sctp_make_cookie_ack(new_asoc, chunk); + repl = sctp_make_cookie_ack(asoc, chunk); if (!repl) goto nomem; @@ -4165,10 +4180,7 @@ gen_shutdown: * The return value is the disposition of the chunk. */ static enum sctp_ierror sctp_sf_authenticate( - struct net *net, - const struct sctp_endpoint *ep, const struct sctp_association *asoc, - const union sctp_subtype type, struct sctp_chunk *chunk) { struct sctp_shared_key *sh_key = NULL; @@ -4269,7 +4281,7 @@ enum sctp_disposition sctp_sf_eat_auth(struct net *net, commands); auth_hdr = (struct sctp_authhdr *)chunk->skb->data; - error = sctp_sf_authenticate(net, ep, asoc, type, chunk); + error = sctp_sf_authenticate(asoc, chunk); switch (error) { case SCTP_IERROR_AUTH_BAD_HMAC: /* Generate the ERROR chunk and discard the rest diff --git a/net/sctp/stream.c b/net/sctp/stream.c index f799043abec9..f1f1d1b232ba 100644 --- a/net/sctp/stream.c +++ b/net/sctp/stream.c @@ -240,6 +240,8 @@ void sctp_stream_update(struct sctp_stream *stream, struct sctp_stream *new) new->out = NULL; new->in = NULL; + new->outcnt = 0; + new->incnt = 0; } static int sctp_send_reconf(struct sctp_association *asoc, diff --git a/net/sctp/ulpevent.c b/net/sctp/ulpevent.c index 84207ad33e8e..8cb7d9858270 100644 --- a/net/sctp/ulpevent.c +++ b/net/sctp/ulpevent.c @@ -715,7 +715,6 @@ struct sctp_ulpevent *sctp_ulpevent_make_rcvmsg(struct sctp_association *asoc, return event; fail_mark: - sctp_chunk_put(chunk); kfree_skb(skb); fail: return NULL; diff --git a/net/smc/af_smc.c b/net/smc/af_smc.c index f5d4b69dbabc..544bab42f925 100644 --- a/net/smc/af_smc.c +++ b/net/smc/af_smc.c @@ -292,6 +292,17 @@ static void smc_copy_sock_settings_to_smc(struct smc_sock *smc) smc_copy_sock_settings(&smc->sk, smc->clcsock->sk, SK_FLAGS_CLC_TO_SMC); } +/* register a new rmb */ +static int smc_reg_rmb(struct smc_link *link, struct smc_buf_desc *rmb_desc) +{ + /* register memory region for new rmb */ + if (smc_wr_reg_send(link, rmb_desc->mr_rx[SMC_SINGLE_LINK])) { + rmb_desc->regerr = 1; + return -EFAULT; + } + return 0; +} + static int smc_clnt_conf_first_link(struct smc_sock *smc) { struct smc_link_group *lgr = smc->conn.lgr; @@ -321,9 +332,7 @@ static int smc_clnt_conf_first_link(struct smc_sock *smc) smc_wr_remember_qp_attr(link); - rc = smc_wr_reg_send(link, - smc->conn.rmb_desc->mr_rx[SMC_SINGLE_LINK]); - if (rc) + if (smc_reg_rmb(link, smc->conn.rmb_desc)) return SMC_CLC_DECL_INTERR; /* send CONFIRM LINK response over RoCE fabric */ @@ -473,13 +482,8 @@ static int smc_connect_rdma(struct smc_sock *smc) goto decline_rdma_unlock; } } else { - struct smc_buf_desc *buf_desc = smc->conn.rmb_desc; - - if (!buf_desc->reused) { - /* register memory region for new rmb */ - rc = smc_wr_reg_send(link, - buf_desc->mr_rx[SMC_SINGLE_LINK]); - if (rc) { + if (!smc->conn.rmb_desc->reused) { + if (smc_reg_rmb(link, smc->conn.rmb_desc)) { reason_code = SMC_CLC_DECL_INTERR; goto decline_rdma_unlock; } @@ -719,9 +723,7 @@ static int smc_serv_conf_first_link(struct smc_sock *smc) link = &lgr->lnk[SMC_SINGLE_LINK]; - rc = smc_wr_reg_send(link, - smc->conn.rmb_desc->mr_rx[SMC_SINGLE_LINK]); - if (rc) + if (smc_reg_rmb(link, smc->conn.rmb_desc)) return SMC_CLC_DECL_INTERR; /* send CONFIRM LINK request to client over the RoCE fabric */ @@ -854,13 +856,8 @@ static void smc_listen_work(struct work_struct *work) smc_rx_init(new_smc); if (local_contact != SMC_FIRST_CONTACT) { - struct smc_buf_desc *buf_desc = new_smc->conn.rmb_desc; - - if (!buf_desc->reused) { - /* register memory region for new rmb */ - rc = smc_wr_reg_send(link, - buf_desc->mr_rx[SMC_SINGLE_LINK]); - if (rc) { + if (!new_smc->conn.rmb_desc->reused) { + if (smc_reg_rmb(link, new_smc->conn.rmb_desc)) { reason_code = SMC_CLC_DECL_INTERR; goto decline_rdma_unlock; } @@ -978,10 +975,6 @@ static void smc_tcp_listen_work(struct work_struct *work) } out: - if (lsmc->clcsock) { - sock_release(lsmc->clcsock); - lsmc->clcsock = NULL; - } release_sock(lsk); sock_put(&lsmc->sk); /* sock_hold in smc_listen */ } @@ -1170,13 +1163,15 @@ static __poll_t smc_poll(struct file *file, struct socket *sock, /* delegate to CLC child sock */ release_sock(sk); mask = smc->clcsock->ops->poll(file, smc->clcsock, wait); - /* if non-blocking connect finished ... */ lock_sock(sk); - if ((sk->sk_state == SMC_INIT) && (mask & EPOLLOUT)) { - sk->sk_err = smc->clcsock->sk->sk_err; - if (sk->sk_err) { - mask |= EPOLLERR; - } else { + sk->sk_err = smc->clcsock->sk->sk_err; + if (sk->sk_err) { + mask |= EPOLLERR; + } else { + /* if non-blocking connect finished ... */ + if (sk->sk_state == SMC_INIT && + mask & EPOLLOUT && + smc->clcsock->sk->sk_state != TCP_CLOSE) { rc = smc_connect_rdma(smc); if (rc < 0) mask |= EPOLLERR; @@ -1320,8 +1315,11 @@ static ssize_t smc_sendpage(struct socket *sock, struct page *page, smc = smc_sk(sk); lock_sock(sk); - if (sk->sk_state != SMC_ACTIVE) + if (sk->sk_state != SMC_ACTIVE) { + release_sock(sk); goto out; + } + release_sock(sk); if (smc->use_fallback) rc = kernel_sendpage(smc->clcsock, page, offset, size, flags); @@ -1329,7 +1327,6 @@ static ssize_t smc_sendpage(struct socket *sock, struct page *page, rc = sock_no_sendpage(sock, page, offset, size, flags); out: - release_sock(sk); return rc; } diff --git a/net/smc/smc_core.c b/net/smc/smc_core.c index f44f6803f7ff..d4bd01bb44e1 100644 --- a/net/smc/smc_core.c +++ b/net/smc/smc_core.c @@ -32,6 +32,9 @@ static u32 smc_lgr_num; /* unique link group number */ +static void smc_buf_free(struct smc_buf_desc *buf_desc, struct smc_link *lnk, + bool is_rmb); + static void smc_lgr_schedule_free_work(struct smc_link_group *lgr) { /* client link group creation always follows the server link group @@ -234,9 +237,22 @@ static void smc_buf_unuse(struct smc_connection *conn) conn->sndbuf_size = 0; } if (conn->rmb_desc) { - conn->rmb_desc->reused = true; - conn->rmb_desc->used = 0; - conn->rmbe_size = 0; + if (!conn->rmb_desc->regerr) { + conn->rmb_desc->reused = 1; + conn->rmb_desc->used = 0; + conn->rmbe_size = 0; + } else { + /* buf registration failed, reuse not possible */ + struct smc_link_group *lgr = conn->lgr; + struct smc_link *lnk; + + write_lock_bh(&lgr->rmbs_lock); + list_del(&conn->rmb_desc->list); + write_unlock_bh(&lgr->rmbs_lock); + + lnk = &lgr->lnk[SMC_SINGLE_LINK]; + smc_buf_free(conn->rmb_desc, lnk, true); + } } } diff --git a/net/smc/smc_core.h b/net/smc/smc_core.h index 07e2a393e6d9..5dfcb15d529f 100644 --- a/net/smc/smc_core.h +++ b/net/smc/smc_core.h @@ -123,7 +123,8 @@ struct smc_buf_desc { */ u32 order; /* allocation order */ u32 used; /* currently used / unused */ - bool reused; /* new created / reused */ + u8 reused : 1; /* new created / reused */ + u8 regerr : 1; /* err during registration */ }; struct smc_rtoken { /* address/key of remote RMB */ diff --git a/net/sunrpc/xprtrdma/fmr_ops.c b/net/sunrpc/xprtrdma/fmr_ops.c index 5cc68a824f45..f2f63959fddd 100644 --- a/net/sunrpc/xprtrdma/fmr_ops.c +++ b/net/sunrpc/xprtrdma/fmr_ops.c @@ -72,6 +72,7 @@ fmr_op_init_mr(struct rpcrdma_ia *ia, struct rpcrdma_mr *mr) if (IS_ERR(mr->fmr.fm_mr)) goto out_fmr_err; + INIT_LIST_HEAD(&mr->mr_list); return 0; out_fmr_err: @@ -102,10 +103,6 @@ fmr_op_release_mr(struct rpcrdma_mr *mr) LIST_HEAD(unmap_list); int rc; - /* Ensure MW is not on any rl_registered list */ - if (!list_empty(&mr->mr_list)) - list_del(&mr->mr_list); - kfree(mr->fmr.fm_physaddrs); kfree(mr->mr_sg); diff --git a/net/sunrpc/xprtrdma/frwr_ops.c b/net/sunrpc/xprtrdma/frwr_ops.c index c5743a0960be..c59c5c788db0 100644 --- a/net/sunrpc/xprtrdma/frwr_ops.c +++ b/net/sunrpc/xprtrdma/frwr_ops.c @@ -110,6 +110,7 @@ frwr_op_init_mr(struct rpcrdma_ia *ia, struct rpcrdma_mr *mr) if (!mr->mr_sg) goto out_list_err; + INIT_LIST_HEAD(&mr->mr_list); sg_init_table(mr->mr_sg, depth); init_completion(&frwr->fr_linv_done); return 0; @@ -133,10 +134,6 @@ frwr_op_release_mr(struct rpcrdma_mr *mr) { int rc; - /* Ensure MR is not on any rl_registered list */ - if (!list_empty(&mr->mr_list)) - list_del(&mr->mr_list); - rc = ib_dereg_mr(mr->frwr.fr_mr); if (rc) pr_err("rpcrdma: final ib_dereg_mr for %p returned %i\n", @@ -195,7 +192,7 @@ frwr_op_recover_mr(struct rpcrdma_mr *mr) return; out_release: - pr_err("rpcrdma: FRWR reset failed %d, %p release\n", rc, mr); + pr_err("rpcrdma: FRWR reset failed %d, %p released\n", rc, mr); r_xprt->rx_stats.mrs_orphaned++; spin_lock(&r_xprt->rx_buf.rb_mrlock); @@ -476,7 +473,7 @@ frwr_op_reminv(struct rpcrdma_rep *rep, struct list_head *mrs) list_for_each_entry(mr, mrs, mr_list) if (mr->mr_handle == rep->rr_inv_rkey) { - list_del(&mr->mr_list); + list_del_init(&mr->mr_list); trace_xprtrdma_remoteinv(mr); mr->frwr.fr_state = FRWR_IS_INVALID; rpcrdma_mr_unmap_and_put(mr); diff --git a/net/sunrpc/xprtrdma/verbs.c b/net/sunrpc/xprtrdma/verbs.c index fe5eaca2d197..c345d365af88 100644 --- a/net/sunrpc/xprtrdma/verbs.c +++ b/net/sunrpc/xprtrdma/verbs.c @@ -1254,6 +1254,11 @@ rpcrdma_mrs_destroy(struct rpcrdma_buffer *buf) list_del(&mr->mr_all); spin_unlock(&buf->rb_mrlock); + + /* Ensure MW is not on any rl_registered list */ + if (!list_empty(&mr->mr_list)) + list_del(&mr->mr_list); + ia->ri_ops->ro_release_mr(mr); count++; spin_lock(&buf->rb_mrlock); diff --git a/net/sunrpc/xprtrdma/xprt_rdma.h b/net/sunrpc/xprtrdma/xprt_rdma.h index 3d3b423fa9c1..cb41b12a3bf8 100644 --- a/net/sunrpc/xprtrdma/xprt_rdma.h +++ b/net/sunrpc/xprtrdma/xprt_rdma.h @@ -380,7 +380,7 @@ rpcrdma_mr_pop(struct list_head *list) struct rpcrdma_mr *mr; mr = list_first_entry(list, struct rpcrdma_mr, mr_list); - list_del(&mr->mr_list); + list_del_init(&mr->mr_list); return mr; } diff --git a/net/tipc/node.c b/net/tipc/node.c index 6f98b56dd48e..f29549de9245 100644 --- a/net/tipc/node.c +++ b/net/tipc/node.c @@ -1950,6 +1950,7 @@ out: int tipc_nl_node_get_link(struct sk_buff *skb, struct genl_info *info) { struct net *net = genl_info_net(info); + struct nlattr *attrs[TIPC_NLA_LINK_MAX + 1]; struct tipc_nl_msg msg; char *name; int err; @@ -1957,9 +1958,19 @@ int tipc_nl_node_get_link(struct sk_buff *skb, struct genl_info *info) msg.portid = info->snd_portid; msg.seq = info->snd_seq; - if (!info->attrs[TIPC_NLA_LINK_NAME]) + if (!info->attrs[TIPC_NLA_LINK]) return -EINVAL; - name = nla_data(info->attrs[TIPC_NLA_LINK_NAME]); + + err = nla_parse_nested(attrs, TIPC_NLA_LINK_MAX, + info->attrs[TIPC_NLA_LINK], + tipc_nl_link_policy, info->extack); + if (err) + return err; + + if (!attrs[TIPC_NLA_LINK_NAME]) + return -EINVAL; + + name = nla_data(attrs[TIPC_NLA_LINK_NAME]); msg.skb = nlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL); if (!msg.skb) @@ -2244,7 +2255,7 @@ int tipc_nl_node_dump_monitor(struct sk_buff *skb, struct netlink_callback *cb) rtnl_lock(); for (bearer_id = prev_bearer; bearer_id < MAX_BEARERS; bearer_id++) { - err = __tipc_nl_add_monitor(net, &msg, prev_bearer); + err = __tipc_nl_add_monitor(net, &msg, bearer_id); if (err) break; } diff --git a/net/tipc/socket.c b/net/tipc/socket.c index 252a52ae0893..6be21575503a 100644 --- a/net/tipc/socket.c +++ b/net/tipc/socket.c @@ -1516,10 +1516,10 @@ static void tipc_sk_set_orig_addr(struct msghdr *m, struct sk_buff *skb) srcaddr->sock.family = AF_TIPC; srcaddr->sock.addrtype = TIPC_ADDR_ID; + srcaddr->sock.scope = 0; srcaddr->sock.addr.id.ref = msg_origport(hdr); srcaddr->sock.addr.id.node = msg_orignode(hdr); srcaddr->sock.addr.name.domain = 0; - srcaddr->sock.scope = 0; m->msg_namelen = sizeof(struct sockaddr_tipc); if (!msg_in_group(hdr)) @@ -1528,6 +1528,7 @@ static void tipc_sk_set_orig_addr(struct msghdr *m, struct sk_buff *skb) /* Group message users may also want to know sending member's id */ srcaddr->member.family = AF_TIPC; srcaddr->member.addrtype = TIPC_ADDR_NAME; + srcaddr->member.scope = 0; srcaddr->member.addr.name.name.type = msg_nametype(hdr); srcaddr->member.addr.name.name.instance = TIPC_SKB_CB(skb)->orig_member; srcaddr->member.addr.name.domain = 0; diff --git a/net/tls/tls_main.c b/net/tls/tls_main.c index 0d379970960e..20cd93be6236 100644 --- a/net/tls/tls_main.c +++ b/net/tls/tls_main.c @@ -114,6 +114,7 @@ int tls_push_sg(struct sock *sk, size = sg->length - offset; offset += sg->offset; + ctx->in_tcp_sendpages = true; while (1) { if (sg_is_last(sg)) sendpage_flags = flags; @@ -134,6 +135,7 @@ retry: offset -= sg->offset; ctx->partially_sent_offset = offset; ctx->partially_sent_record = (void *)sg; + ctx->in_tcp_sendpages = false; return ret; } @@ -148,6 +150,8 @@ retry: } clear_bit(TLS_PENDING_CLOSED_RECORD, &ctx->flags); + ctx->in_tcp_sendpages = false; + ctx->sk_write_space(sk); return 0; } @@ -217,6 +221,10 @@ static void tls_write_space(struct sock *sk) { struct tls_context *ctx = tls_get_ctx(sk); + /* We are already sending pages, ignore notification */ + if (ctx->in_tcp_sendpages) + return; + if (!sk->sk_write_pending && tls_is_pending_closed_record(ctx)) { gfp_t sk_allocation = sk->sk_allocation; int rc; @@ -241,16 +249,13 @@ static void tls_sk_proto_close(struct sock *sk, long timeout) struct tls_context *ctx = tls_get_ctx(sk); long timeo = sock_sndtimeo(sk, 0); void (*sk_proto_close)(struct sock *sk, long timeout); + bool free_ctx = false; lock_sock(sk); sk_proto_close = ctx->sk_proto_close; - if (ctx->conf == TLS_HW_RECORD) - goto skip_tx_cleanup; - - if (ctx->conf == TLS_BASE) { - kfree(ctx); - ctx = NULL; + if (ctx->conf == TLS_BASE || ctx->conf == TLS_HW_RECORD) { + free_ctx = true; goto skip_tx_cleanup; } @@ -287,7 +292,7 @@ skip_tx_cleanup: /* free ctx for TLS_HW_RECORD, used by tcp_set_state * for sk->sk_prot->unhash [tls_hw_unhash] */ - if (ctx && ctx->conf == TLS_HW_RECORD) + if (free_ctx) kfree(ctx); } diff --git a/net/wireless/core.c b/net/wireless/core.c index a6f3cac8c640..c0fd8a85e7f7 100644 --- a/net/wireless/core.c +++ b/net/wireless/core.c @@ -95,6 +95,9 @@ static int cfg80211_dev_check_name(struct cfg80211_registered_device *rdev, ASSERT_RTNL(); + if (strlen(newname) > NL80211_WIPHY_NAME_MAXLEN) + return -EINVAL; + /* prohibit calling the thing phy%d when %d is not its number */ sscanf(newname, PHY_NAME "%d%n", &wiphy_idx, &taken); if (taken == strlen(newname) && wiphy_idx != rdev->wiphy_idx) { diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index ff28f8feeb09..a052693c2e85 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -9214,6 +9214,7 @@ static int nl80211_connect(struct sk_buff *skb, struct genl_info *info) if (nla_get_flag(info->attrs[NL80211_ATTR_EXTERNAL_AUTH_SUPPORT])) { if (!info->attrs[NL80211_ATTR_SOCKET_OWNER]) { + kzfree(connkeys); GENL_SET_ERR_MSG(info, "external auth requires connection ownership"); return -EINVAL; diff --git a/net/wireless/reg.c b/net/wireless/reg.c index 16c7e4ef5820..ac3e12c32aa3 100644 --- a/net/wireless/reg.c +++ b/net/wireless/reg.c @@ -1026,6 +1026,7 @@ static int regdb_query_country(const struct fwdb_header *db, if (!tmp_rd) { kfree(regdom); + kfree(wmm_ptrs); return -ENOMEM; } regdom = tmp_rd; diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c index f9d2f2233f09..6c177ae7a6d9 100644 --- a/net/xfrm/xfrm_state.c +++ b/net/xfrm/xfrm_state.c @@ -2175,6 +2175,12 @@ struct xfrm_state_afinfo *xfrm_state_get_afinfo(unsigned int family) return afinfo; } +void xfrm_flush_gc(void) +{ + flush_work(&xfrm_state_gc_work); +} +EXPORT_SYMBOL(xfrm_flush_gc); + /* Temporarily located here until net/xfrm/xfrm_tunnel.c is created */ void xfrm_state_delete_tunnel(struct xfrm_state *x) { diff --git a/samples/sockmap/Makefile b/samples/sockmap/Makefile index 9bf2881bd11b..fa53f4d77834 100644 --- a/samples/sockmap/Makefile +++ b/samples/sockmap/Makefile @@ -65,11 +65,14 @@ $(src)/*.c: verify_target_bpf # asm/sysreg.h - inline assembly used by it is incompatible with llvm. # But, there is no easy way to fix it, so just exclude it since it is # useless for BPF samples. +# +# -target bpf option required with SK_MSG programs, this is to ensure +# reading 'void *' data types for data and data_end are __u64 reads. $(obj)/%.o: $(src)/%.c $(CLANG) $(NOSTDINC_FLAGS) $(LINUXINCLUDE) $(EXTRA_CFLAGS) -I$(obj) \ -D__KERNEL__ -D__ASM_SYSREG_H -Wno-unused-value -Wno-pointer-sign \ -Wno-compare-distinct-pointer-types \ -Wno-gnu-variable-sized-type-not-at-end \ -Wno-address-of-packed-member -Wno-tautological-compare \ - -Wno-unknown-warning-option \ - -O2 -emit-llvm -c $< -o -| $(LLC) -march=bpf -filetype=obj -o $@ + -Wno-unknown-warning-option -O2 -target bpf \ + -emit-llvm -c $< -o -| $(LLC) -march=bpf -filetype=obj -o $@ diff --git a/scripts/Makefile.gcc-plugins b/scripts/Makefile.gcc-plugins index b2a95af7df18..7f5c86246138 100644 --- a/scripts/Makefile.gcc-plugins +++ b/scripts/Makefile.gcc-plugins @@ -14,7 +14,7 @@ ifdef CONFIG_GCC_PLUGINS endif ifdef CONFIG_GCC_PLUGIN_SANCOV - ifeq ($(CFLAGS_KCOV),) + ifeq ($(strip $(CFLAGS_KCOV)),) # It is needed because of the gcc-plugin.sh and gcc version checks. gcc-plugin-$(CONFIG_GCC_PLUGIN_SANCOV) += sancov_plugin.so diff --git a/scripts/Makefile.lib b/scripts/Makefile.lib index 07d07409f16f..5af34a2b0cd9 100644 --- a/scripts/Makefile.lib +++ b/scripts/Makefile.lib @@ -196,7 +196,7 @@ $(obj)/%.tab.c: $(src)/%.y FORCE $(call if_changed,bison) quiet_cmd_bison_h = YACC $@ - cmd_bison_h = bison -o/dev/null --defines=$@ -t -l $< + cmd_bison_h = $(YACC) -o/dev/null --defines=$@ -t -l $< $(obj)/%.tab.h: $(src)/%.y FORCE $(call if_changed,bison_h) diff --git a/scripts/dtc/checks.c b/scripts/dtc/checks.c index c07ba4da9e36..815eaf140ab5 100644 --- a/scripts/dtc/checks.c +++ b/scripts/dtc/checks.c @@ -787,10 +787,9 @@ static void check_pci_bridge(struct check *c, struct dt_info *dti, struct node * FAIL(c, dti, node, "incorrect #size-cells for PCI bridge"); prop = get_property(node, "bus-range"); - if (!prop) { - FAIL(c, dti, node, "missing bus-range for PCI bridge"); + if (!prop) return; - } + if (prop->val.len != (sizeof(cell_t) * 2)) { FAIL_PROP(c, dti, node, prop, "value must be 2 cells"); return; diff --git a/scripts/extract_xc3028.pl b/scripts/extract_xc3028.pl index 61d9b256c658..a1c51b7e4baf 100755 --- a/scripts/extract_xc3028.pl +++ b/scripts/extract_xc3028.pl @@ -1,6 +1,6 @@ #!/usr/bin/env perl -# Copyright (c) Mauro Carvalho Chehab <mchehab@infradead.org> +# Copyright (c) Mauro Carvalho Chehab <mchehab@kernel.org> # Released under GPLv2 # # In order to use, you need to: diff --git a/scripts/faddr2line b/scripts/faddr2line index 9e5735a4d3a5..1876a741087c 100755 --- a/scripts/faddr2line +++ b/scripts/faddr2line @@ -170,7 +170,10 @@ __faddr2line() { echo "$file_lines" | while read -r line do echo $line - eval $(echo $line | awk -F "[ :]" '{printf("n1=%d;n2=%d;f=%s",$NF-5, $NF+5, $(NF-1))}') + n=$(echo $line | sed 's/.*:\([0-9]\+\).*/\1/g') + n1=$[$n-5] + n2=$[$n+5] + f=$(echo $line | sed 's/.*at \(.\+\):.*/\1/g') awk 'NR>=strtonum("'$n1'") && NR<=strtonum("'$n2'") {printf("%d\t%s\n", NR, $0)}' $f done diff --git a/scripts/genksyms/Makefile b/scripts/genksyms/Makefile index ef0287e42957..03b7ce97de14 100644 --- a/scripts/genksyms/Makefile +++ b/scripts/genksyms/Makefile @@ -14,14 +14,14 @@ genksyms-objs := genksyms.o parse.tab.o lex.lex.o # so that 'bison: not found' will be displayed if it is missing. ifeq ($(findstring 1,$(KBUILD_ENABLE_EXTRA_GCC_CHECKS)),) -quiet_cmd_bison_no_warn = $(quet_cmd_bison) +quiet_cmd_bison_no_warn = $(quiet_cmd_bison) cmd_bison_no_warn = $(YACC) --version >/dev/null; \ $(cmd_bison) 2>/dev/null $(obj)/parse.tab.c: $(src)/parse.y FORCE $(call if_changed,bison_no_warn) -quiet_cmd_bison_h_no_warn = $(quet_cmd_bison_h) +quiet_cmd_bison_h_no_warn = $(quiet_cmd_bison_h) cmd_bison_h_no_warn = $(YACC) --version >/dev/null; \ $(cmd_bison_h) 2>/dev/null diff --git a/scripts/mod/sumversion.c b/scripts/mod/sumversion.c index 944418da9fe3..0f6dcb4011a8 100644 --- a/scripts/mod/sumversion.c +++ b/scripts/mod/sumversion.c @@ -330,14 +330,7 @@ static int parse_source_files(const char *objfile, struct md4_ctx *md) goto out; } - /* There will be a line like so: - deps_drivers/net/dummy.o := \ - drivers/net/dummy.c \ - $(wildcard include/config/net/fastroute.h) \ - include/linux/module.h \ - - Sum all files in the same dir or subdirs. - */ + /* Sum all files in the same dir or subdirs. */ while ((line = get_next_line(&pos, file, flen)) != NULL) { char* p = line; diff --git a/scripts/split-man.pl b/scripts/split-man.pl index bfe16cbe42df..c3db607ee9ec 100755 --- a/scripts/split-man.pl +++ b/scripts/split-man.pl @@ -1,7 +1,7 @@ #!/usr/bin/perl # SPDX-License-Identifier: GPL-2.0 # -# Author: Mauro Carvalho Chehab <mchehab@s-opensource.com> +# Author: Mauro Carvalho Chehab <mchehab+samsung@kernel.org> # # Produce manpages from kernel-doc. # See Documentation/doc-guide/kernel-doc.rst for instructions diff --git a/sound/core/pcm_compat.c b/sound/core/pcm_compat.c index 06d7c40af570..6491afbb5fd5 100644 --- a/sound/core/pcm_compat.c +++ b/sound/core/pcm_compat.c @@ -423,6 +423,8 @@ static int snd_pcm_ioctl_xfern_compat(struct snd_pcm_substream *substream, return -ENOTTY; if (substream->stream != dir) return -EINVAL; + if (substream->runtime->status->state == SNDRV_PCM_STATE_OPEN) + return -EBADFD; if ((ch = substream->runtime->channels) > 128) return -EINVAL; diff --git a/sound/core/seq/seq_virmidi.c b/sound/core/seq/seq_virmidi.c index f48a4cd24ffc..289ae6bb81d9 100644 --- a/sound/core/seq/seq_virmidi.c +++ b/sound/core/seq/seq_virmidi.c @@ -174,12 +174,12 @@ static void snd_virmidi_output_trigger(struct snd_rawmidi_substream *substream, } return; } + spin_lock_irqsave(&substream->runtime->lock, flags); if (vmidi->event.type != SNDRV_SEQ_EVENT_NONE) { if (snd_seq_kernel_client_dispatch(vmidi->client, &vmidi->event, in_atomic(), 0) < 0) - return; + goto out; vmidi->event.type = SNDRV_SEQ_EVENT_NONE; } - spin_lock_irqsave(&substream->runtime->lock, flags); while (1) { count = __snd_rawmidi_transmit_peek(substream, buf, sizeof(buf)); if (count <= 0) diff --git a/sound/drivers/aloop.c b/sound/drivers/aloop.c index 58e349fc893f..eab7f594ebe7 100644 --- a/sound/drivers/aloop.c +++ b/sound/drivers/aloop.c @@ -831,9 +831,11 @@ static int loopback_rate_shift_get(struct snd_kcontrol *kcontrol, { struct loopback *loopback = snd_kcontrol_chip(kcontrol); + mutex_lock(&loopback->cable_lock); ucontrol->value.integer.value[0] = loopback->setup[kcontrol->id.subdevice] [kcontrol->id.device].rate_shift; + mutex_unlock(&loopback->cable_lock); return 0; } @@ -865,9 +867,11 @@ static int loopback_notify_get(struct snd_kcontrol *kcontrol, { struct loopback *loopback = snd_kcontrol_chip(kcontrol); + mutex_lock(&loopback->cable_lock); ucontrol->value.integer.value[0] = loopback->setup[kcontrol->id.subdevice] [kcontrol->id.device].notify; + mutex_unlock(&loopback->cable_lock); return 0; } @@ -879,12 +883,14 @@ static int loopback_notify_put(struct snd_kcontrol *kcontrol, int change = 0; val = ucontrol->value.integer.value[0] ? 1 : 0; + mutex_lock(&loopback->cable_lock); if (val != loopback->setup[kcontrol->id.subdevice] [kcontrol->id.device].notify) { loopback->setup[kcontrol->id.subdevice] [kcontrol->id.device].notify = val; change = 1; } + mutex_unlock(&loopback->cable_lock); return change; } @@ -892,15 +898,18 @@ static int loopback_active_get(struct snd_kcontrol *kcontrol, struct snd_ctl_elem_value *ucontrol) { struct loopback *loopback = snd_kcontrol_chip(kcontrol); - struct loopback_cable *cable = loopback->cables - [kcontrol->id.subdevice][kcontrol->id.device ^ 1]; + struct loopback_cable *cable; + unsigned int val = 0; + mutex_lock(&loopback->cable_lock); + cable = loopback->cables[kcontrol->id.subdevice][kcontrol->id.device ^ 1]; if (cable != NULL) { unsigned int running = cable->running ^ cable->pause; val = (running & (1 << SNDRV_PCM_STREAM_PLAYBACK)) ? 1 : 0; } + mutex_unlock(&loopback->cable_lock); ucontrol->value.integer.value[0] = val; return 0; } @@ -943,9 +952,11 @@ static int loopback_rate_get(struct snd_kcontrol *kcontrol, { struct loopback *loopback = snd_kcontrol_chip(kcontrol); + mutex_lock(&loopback->cable_lock); ucontrol->value.integer.value[0] = loopback->setup[kcontrol->id.subdevice] [kcontrol->id.device].rate; + mutex_unlock(&loopback->cable_lock); return 0; } @@ -965,9 +976,11 @@ static int loopback_channels_get(struct snd_kcontrol *kcontrol, { struct loopback *loopback = snd_kcontrol_chip(kcontrol); + mutex_lock(&loopback->cable_lock); ucontrol->value.integer.value[0] = loopback->setup[kcontrol->id.subdevice] [kcontrol->id.device].channels; + mutex_unlock(&loopback->cable_lock); return 0; } diff --git a/sound/firewire/amdtp-stream.c b/sound/firewire/amdtp-stream.c index 4a1dc145327b..cb9acfe60f6a 100644 --- a/sound/firewire/amdtp-stream.c +++ b/sound/firewire/amdtp-stream.c @@ -773,8 +773,6 @@ static void amdtp_stream_first_callback(struct fw_iso_context *context, u32 cycle; unsigned int packets; - s->max_payload_length = amdtp_stream_get_max_payload(s); - /* * For in-stream, first packet has come. * For out-stream, prepared to transmit first packet @@ -879,6 +877,9 @@ int amdtp_stream_start(struct amdtp_stream *s, int channel, int speed) amdtp_stream_update(s); + if (s->direction == AMDTP_IN_STREAM) + s->max_payload_length = amdtp_stream_get_max_payload(s); + if (s->flags & CIP_NO_HEADER) s->tag = TAG_NO_CIP_HEADER; else diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c index 8c238e51bb5a..2dd34dd77447 100644 --- a/sound/pci/hda/patch_realtek.c +++ b/sound/pci/hda/patch_realtek.c @@ -3832,7 +3832,7 @@ static void alc280_fixup_hp_gpio4(struct hda_codec *codec, } } -#if IS_REACHABLE(INPUT) +#if IS_REACHABLE(CONFIG_INPUT) static void gpio2_mic_hotkey_event(struct hda_codec *codec, struct hda_jack_callback *event) { diff --git a/tools/arch/arm/include/uapi/asm/kvm.h b/tools/arch/arm/include/uapi/asm/kvm.h index 2ba95d6fe852..caae4843cb70 100644 --- a/tools/arch/arm/include/uapi/asm/kvm.h +++ b/tools/arch/arm/include/uapi/asm/kvm.h @@ -195,6 +195,12 @@ struct kvm_arch_memory_slot { #define KVM_REG_ARM_VFP_FPINST 0x1009 #define KVM_REG_ARM_VFP_FPINST2 0x100A +/* KVM-as-firmware specific pseudo-registers */ +#define KVM_REG_ARM_FW (0x0014 << KVM_REG_ARM_COPROC_SHIFT) +#define KVM_REG_ARM_FW_REG(r) (KVM_REG_ARM | KVM_REG_SIZE_U64 | \ + KVM_REG_ARM_FW | ((r) & 0xffff)) +#define KVM_REG_ARM_PSCI_VERSION KVM_REG_ARM_FW_REG(0) + /* Device Control API: ARM VGIC */ #define KVM_DEV_ARM_VGIC_GRP_ADDR 0 #define KVM_DEV_ARM_VGIC_GRP_DIST_REGS 1 diff --git a/tools/arch/arm64/include/uapi/asm/kvm.h b/tools/arch/arm64/include/uapi/asm/kvm.h index 9abbf3044654..04b3256f8e6d 100644 --- a/tools/arch/arm64/include/uapi/asm/kvm.h +++ b/tools/arch/arm64/include/uapi/asm/kvm.h @@ -206,6 +206,12 @@ struct kvm_arch_memory_slot { #define KVM_REG_ARM_TIMER_CNT ARM64_SYS_REG(3, 3, 14, 3, 2) #define KVM_REG_ARM_TIMER_CVAL ARM64_SYS_REG(3, 3, 14, 0, 2) +/* KVM-as-firmware specific pseudo-registers */ +#define KVM_REG_ARM_FW (0x0014 << KVM_REG_ARM_COPROC_SHIFT) +#define KVM_REG_ARM_FW_REG(r) (KVM_REG_ARM64 | KVM_REG_SIZE_U64 | \ + KVM_REG_ARM_FW | ((r) & 0xffff)) +#define KVM_REG_ARM_PSCI_VERSION KVM_REG_ARM_FW_REG(0) + /* Device Control API: ARM VGIC */ #define KVM_DEV_ARM_VGIC_GRP_ADDR 0 #define KVM_DEV_ARM_VGIC_GRP_DIST_REGS 1 diff --git a/tools/arch/x86/include/asm/cpufeatures.h b/tools/arch/x86/include/asm/cpufeatures.h index d554c11e01ff..578793e97431 100644 --- a/tools/arch/x86/include/asm/cpufeatures.h +++ b/tools/arch/x86/include/asm/cpufeatures.h @@ -320,6 +320,7 @@ #define X86_FEATURE_AVX512_VPOPCNTDQ (16*32+14) /* POPCNT for vectors of DW/QW */ #define X86_FEATURE_LA57 (16*32+16) /* 5-level page tables */ #define X86_FEATURE_RDPID (16*32+22) /* RDPID instruction */ +#define X86_FEATURE_CLDEMOTE (16*32+25) /* CLDEMOTE instruction */ /* AMD-defined CPU features, CPUID level 0x80000007 (EBX), word 17 */ #define X86_FEATURE_OVERFLOW_RECOV (17*32+ 0) /* MCA overflow recovery support */ diff --git a/tools/bpf/Makefile b/tools/bpf/Makefile index 1ea545965ee3..53b60ad452f5 100644 --- a/tools/bpf/Makefile +++ b/tools/bpf/Makefile @@ -76,6 +76,8 @@ $(OUTPUT)bpf_asm: $(OUTPUT)bpf_asm.o $(OUTPUT)bpf_exp.yacc.o $(OUTPUT)bpf_exp.le $(QUIET_LINK)$(CC) $(CFLAGS) -o $@ $^ $(OUTPUT)bpf_exp.lex.c: $(OUTPUT)bpf_exp.yacc.c +$(OUTPUT)bpf_exp.yacc.o: $(OUTPUT)bpf_exp.yacc.c +$(OUTPUT)bpf_exp.lex.o: $(OUTPUT)bpf_exp.lex.c clean: bpftool_clean $(call QUIET_CLEAN, bpf-progs) diff --git a/tools/bpf/bpf_dbg.c b/tools/bpf/bpf_dbg.c index 4f254bcc4423..61b9aa5d6415 100644 --- a/tools/bpf/bpf_dbg.c +++ b/tools/bpf/bpf_dbg.c @@ -1063,7 +1063,7 @@ static int cmd_load_pcap(char *file) static int cmd_load(char *arg) { - char *subcmd, *cont, *tmp = strdup(arg); + char *subcmd, *cont = NULL, *tmp = strdup(arg); int ret = CMD_OK; subcmd = strtok_r(tmp, " ", &cont); @@ -1073,7 +1073,10 @@ static int cmd_load(char *arg) bpf_reset(); bpf_reset_breakpoints(); - ret = cmd_load_bpf(cont); + if (!cont) + ret = CMD_ERR; + else + ret = cmd_load_bpf(cont); } else if (matches(subcmd, "pcap") == 0) { ret = cmd_load_pcap(cont); } else { diff --git a/tools/include/uapi/linux/kvm.h b/tools/include/uapi/linux/kvm.h index 1065006c9bf5..b02c41e53d56 100644 --- a/tools/include/uapi/linux/kvm.h +++ b/tools/include/uapi/linux/kvm.h @@ -676,6 +676,13 @@ struct kvm_ioeventfd { __u8 pad[36]; }; +#define KVM_X86_DISABLE_EXITS_MWAIT (1 << 0) +#define KVM_X86_DISABLE_EXITS_HTL (1 << 1) +#define KVM_X86_DISABLE_EXITS_PAUSE (1 << 2) +#define KVM_X86_DISABLE_VALID_EXITS (KVM_X86_DISABLE_EXITS_MWAIT | \ + KVM_X86_DISABLE_EXITS_HTL | \ + KVM_X86_DISABLE_EXITS_PAUSE) + /* for KVM_ENABLE_CAP */ struct kvm_enable_cap { /* in */ diff --git a/tools/perf/bench/numa.c b/tools/perf/bench/numa.c index 944070e98a2c..63eb49082774 100644 --- a/tools/perf/bench/numa.c +++ b/tools/perf/bench/numa.c @@ -175,7 +175,7 @@ static const struct option options[] = { OPT_UINTEGER('s', "nr_secs" , &p0.nr_secs, "max number of seconds to run (default: 5 secs)"), OPT_UINTEGER('u', "usleep" , &p0.sleep_usecs, "usecs to sleep per loop iteration"), - OPT_BOOLEAN('R', "data_reads" , &p0.data_reads, "access the data via writes (can be mixed with -W)"), + OPT_BOOLEAN('R', "data_reads" , &p0.data_reads, "access the data via reads (can be mixed with -W)"), OPT_BOOLEAN('W', "data_writes" , &p0.data_writes, "access the data via writes (can be mixed with -R)"), OPT_BOOLEAN('B', "data_backwards", &p0.data_backwards, "access the data backwards as well"), OPT_BOOLEAN('Z', "data_zero_memset", &p0.data_zero_memset,"access the data via glibc bzero only"), diff --git a/tools/perf/pmu-events/arch/x86/mapfile.csv b/tools/perf/pmu-events/arch/x86/mapfile.csv index 93656f2fd53a..7e3cce3bcf3b 100644 --- a/tools/perf/pmu-events/arch/x86/mapfile.csv +++ b/tools/perf/pmu-events/arch/x86/mapfile.csv @@ -29,7 +29,6 @@ GenuineIntel-6-4D,v13,silvermont,core GenuineIntel-6-4C,v13,silvermont,core GenuineIntel-6-2A,v15,sandybridge,core GenuineIntel-6-2C,v2,westmereep-dp,core -GenuineIntel-6-2C,v2,westmereep-dp,core GenuineIntel-6-25,v2,westmereep-sp,core GenuineIntel-6-2F,v2,westmereex,core GenuineIntel-6-55,v1,skylakex,core diff --git a/tools/perf/util/parse-events.y b/tools/perf/util/parse-events.y index d14464c42714..7afeb80cc39e 100644 --- a/tools/perf/util/parse-events.y +++ b/tools/perf/util/parse-events.y @@ -224,15 +224,15 @@ event_def: event_pmu | event_bpf_file event_pmu: -PE_NAME '/' event_config '/' +PE_NAME opt_event_config { struct list_head *list, *orig_terms, *terms; - if (parse_events_copy_term_list($3, &orig_terms)) + if (parse_events_copy_term_list($2, &orig_terms)) YYABORT; ALLOC_LIST(list); - if (parse_events_add_pmu(_parse_state, list, $1, $3, false)) { + if (parse_events_add_pmu(_parse_state, list, $1, $2, false)) { struct perf_pmu *pmu = NULL; int ok = 0; char *pattern; @@ -262,7 +262,7 @@ PE_NAME '/' event_config '/' if (!ok) YYABORT; } - parse_events_terms__delete($3); + parse_events_terms__delete($2); parse_events_terms__delete(orig_terms); $$ = list; } diff --git a/tools/power/acpi/Makefile.config b/tools/power/acpi/Makefile.config index 2cccbba64418..f304be71c278 100644 --- a/tools/power/acpi/Makefile.config +++ b/tools/power/acpi/Makefile.config @@ -56,6 +56,7 @@ INSTALL_SCRIPT = ${INSTALL_PROGRAM} # to compile vs uClibc, that can be done here as well. CROSS = #/usr/i386-linux-uclibc/usr/bin/i386-uclibc- CROSS_COMPILE ?= $(CROSS) +LD = $(CC) HOSTCC = gcc # check if compiler option is supported diff --git a/tools/testing/selftests/bpf/test_progs.c b/tools/testing/selftests/bpf/test_progs.c index faadbe233966..4123d0ab90ba 100644 --- a/tools/testing/selftests/bpf/test_progs.c +++ b/tools/testing/selftests/bpf/test_progs.c @@ -1108,7 +1108,7 @@ static void test_stacktrace_build_id(void) assert(system("dd if=/dev/urandom of=/dev/zero count=4 2> /dev/null") == 0); - assert(system("./urandom_read if=/dev/urandom of=/dev/zero count=4 2> /dev/null") == 0); + assert(system("./urandom_read") == 0); /* disable stack trace collection */ key = 0; val = 1; @@ -1158,7 +1158,7 @@ static void test_stacktrace_build_id(void) } while (bpf_map_get_next_key(stackmap_fd, &previous_key, &key) == 0); CHECK(build_id_matches < 1, "build id match", - "Didn't find expected build ID from the map"); + "Didn't find expected build ID from the map\n"); disable_pmu: ioctl(pmu_fd, PERF_EVENT_IOC_DISABLE); diff --git a/tools/testing/selftests/lib.mk b/tools/testing/selftests/lib.mk index 195e9d4739a9..c1b1a4dc6a96 100644 --- a/tools/testing/selftests/lib.mk +++ b/tools/testing/selftests/lib.mk @@ -20,10 +20,10 @@ all: $(TEST_GEN_PROGS) $(TEST_GEN_PROGS_EXTENDED) $(TEST_GEN_FILES) .ONESHELL: define RUN_TESTS - @export KSFT_TAP_LEVEL=`echo 1`; - @test_num=`echo 0`; - @echo "TAP version 13"; - @for TEST in $(1); do \ + @export KSFT_TAP_LEVEL=`echo 1`; \ + test_num=`echo 0`; \ + echo "TAP version 13"; \ + for TEST in $(1); do \ BASENAME_TEST=`basename $$TEST`; \ test_num=`echo $$test_num+1 | bc`; \ echo "selftests: $$BASENAME_TEST"; \ diff --git a/tools/testing/selftests/net/Makefile b/tools/testing/selftests/net/Makefile index 8f1e13d2e547..3ff81a478dbe 100644 --- a/tools/testing/selftests/net/Makefile +++ b/tools/testing/selftests/net/Makefile @@ -5,7 +5,8 @@ CFLAGS = -Wall -Wl,--no-as-needed -O2 -g CFLAGS += -I../../../../usr/include/ TEST_PROGS := run_netsocktests run_afpackettests test_bpf.sh netdevice.sh rtnetlink.sh -TEST_PROGS += fib_tests.sh fib-onlink-tests.sh in_netns.sh pmtu.sh +TEST_PROGS += fib_tests.sh fib-onlink-tests.sh pmtu.sh +TEST_PROGS_EXTENDED := in_netns.sh TEST_GEN_FILES = socket TEST_GEN_FILES += psock_fanout psock_tpacket msg_zerocopy TEST_GEN_PROGS = reuseport_bpf reuseport_bpf_cpu reuseport_bpf_numa diff --git a/tools/testing/selftests/tc-testing/tc-tests/actions/bpf.json b/tools/testing/selftests/tc-testing/tc-tests/actions/bpf.json index 5b012f4981d4..6f289a49e5ec 100644 --- a/tools/testing/selftests/tc-testing/tc-tests/actions/bpf.json +++ b/tools/testing/selftests/tc-testing/tc-tests/actions/bpf.json @@ -66,7 +66,7 @@ "cmdUnderTest": "$TC action add action bpf object-file _b.o index 667", "expExitCode": "0", "verifyCmd": "$TC action get action bpf index 667", - "matchPattern": "action order [0-9]*: bpf _b.o:\\[action\\] id [0-9]* tag 3b185187f1855c4c default-action pipe.*index 667 ref", + "matchPattern": "action order [0-9]*: bpf _b.o:\\[action\\] id [0-9]* tag 3b185187f1855c4c( jited)? default-action pipe.*index 667 ref", "matchCount": "1", "teardown": [ "$TC action flush action bpf", @@ -92,10 +92,15 @@ "cmdUnderTest": "$TC action add action bpf object-file _c.o index 667", "expExitCode": "255", "verifyCmd": "$TC action get action bpf index 667", - "matchPattern": "action order [0-9]*: bpf _b.o:\\[action\\] id [0-9].*index 667 ref", + "matchPattern": "action order [0-9]*: bpf _c.o:\\[action\\] id [0-9].*index 667 ref", "matchCount": "0", "teardown": [ - "$TC action flush action bpf", + [ + "$TC action flush action bpf", + 0, + 1, + 255 + ], "rm -f _c.o" ] }, diff --git a/virt/kvm/arm/vgic/vgic-init.c b/virt/kvm/arm/vgic/vgic-init.c index 68378fe17a0e..e07156c30323 100644 --- a/virt/kvm/arm/vgic/vgic-init.c +++ b/virt/kvm/arm/vgic/vgic-init.c @@ -423,7 +423,7 @@ static irqreturn_t vgic_maintenance_handler(int irq, void *data) * We cannot rely on the vgic maintenance interrupt to be * delivered synchronously. This means we can only use it to * exit the VM, and we perform the handling of EOIed - * interrupts on the exit path (see vgic_process_maintenance). + * interrupts on the exit path (see vgic_fold_lr_state). */ return IRQ_HANDLED; } diff --git a/virt/kvm/arm/vgic/vgic-mmio.c b/virt/kvm/arm/vgic/vgic-mmio.c index dbe99d635c80..ff9655cfeb2f 100644 --- a/virt/kvm/arm/vgic/vgic-mmio.c +++ b/virt/kvm/arm/vgic/vgic-mmio.c @@ -289,10 +289,16 @@ static void vgic_mmio_change_active(struct kvm_vcpu *vcpu, struct vgic_irq *irq, irq->vcpu->cpu != -1) /* VCPU thread is running */ cond_resched_lock(&irq->irq_lock); - if (irq->hw) + if (irq->hw) { vgic_hw_irq_change_active(vcpu, irq, active, !requester_vcpu); - else + } else { + u32 model = vcpu->kvm->arch.vgic.vgic_model; + irq->active = active; + if (model == KVM_DEV_TYPE_ARM_VGIC_V2 && + active && vgic_irq_is_sgi(irq->intid)) + irq->active_source = requester_vcpu->vcpu_id; + } if (irq->active) vgic_queue_irq_unlock(vcpu->kvm, irq, flags); diff --git a/virt/kvm/arm/vgic/vgic-v2.c b/virt/kvm/arm/vgic/vgic-v2.c index 45aa433f018f..a5f2e44f1c33 100644 --- a/virt/kvm/arm/vgic/vgic-v2.c +++ b/virt/kvm/arm/vgic/vgic-v2.c @@ -37,13 +37,6 @@ void vgic_v2_init_lrs(void) vgic_v2_write_lr(i, 0); } -void vgic_v2_set_npie(struct kvm_vcpu *vcpu) -{ - struct vgic_v2_cpu_if *cpuif = &vcpu->arch.vgic_cpu.vgic_v2; - - cpuif->vgic_hcr |= GICH_HCR_NPIE; -} - void vgic_v2_set_underflow(struct kvm_vcpu *vcpu) { struct vgic_v2_cpu_if *cpuif = &vcpu->arch.vgic_cpu.vgic_v2; @@ -71,13 +64,18 @@ void vgic_v2_fold_lr_state(struct kvm_vcpu *vcpu) int lr; unsigned long flags; - cpuif->vgic_hcr &= ~(GICH_HCR_UIE | GICH_HCR_NPIE); + cpuif->vgic_hcr &= ~GICH_HCR_UIE; for (lr = 0; lr < vgic_cpu->used_lrs; lr++) { u32 val = cpuif->vgic_lr[lr]; - u32 intid = val & GICH_LR_VIRTUALID; + u32 cpuid, intid = val & GICH_LR_VIRTUALID; struct vgic_irq *irq; + /* Extract the source vCPU id from the LR */ + cpuid = val & GICH_LR_PHYSID_CPUID; + cpuid >>= GICH_LR_PHYSID_CPUID_SHIFT; + cpuid &= 7; + /* Notify fds when the guest EOI'ed a level-triggered SPI */ if (lr_signals_eoi_mi(val) && vgic_valid_spi(vcpu->kvm, intid)) kvm_notify_acked_irq(vcpu->kvm, 0, @@ -90,17 +88,16 @@ void vgic_v2_fold_lr_state(struct kvm_vcpu *vcpu) /* Always preserve the active bit */ irq->active = !!(val & GICH_LR_ACTIVE_BIT); + if (irq->active && vgic_irq_is_sgi(intid)) + irq->active_source = cpuid; + /* Edge is the only case where we preserve the pending bit */ if (irq->config == VGIC_CONFIG_EDGE && (val & GICH_LR_PENDING_BIT)) { irq->pending_latch = true; - if (vgic_irq_is_sgi(intid)) { - u32 cpuid = val & GICH_LR_PHYSID_CPUID; - - cpuid >>= GICH_LR_PHYSID_CPUID_SHIFT; + if (vgic_irq_is_sgi(intid)) irq->source |= (1 << cpuid); - } } /* @@ -152,8 +149,15 @@ void vgic_v2_populate_lr(struct kvm_vcpu *vcpu, struct vgic_irq *irq, int lr) u32 val = irq->intid; bool allow_pending = true; - if (irq->active) + if (irq->active) { val |= GICH_LR_ACTIVE_BIT; + if (vgic_irq_is_sgi(irq->intid)) + val |= irq->active_source << GICH_LR_PHYSID_CPUID_SHIFT; + if (vgic_irq_is_multi_sgi(irq)) { + allow_pending = false; + val |= GICH_LR_EOI; + } + } if (irq->hw) { val |= GICH_LR_HW; @@ -190,8 +194,10 @@ void vgic_v2_populate_lr(struct kvm_vcpu *vcpu, struct vgic_irq *irq, int lr) BUG_ON(!src); val |= (src - 1) << GICH_LR_PHYSID_CPUID_SHIFT; irq->source &= ~(1 << (src - 1)); - if (irq->source) + if (irq->source) { irq->pending_latch = true; + val |= GICH_LR_EOI; + } } } diff --git a/virt/kvm/arm/vgic/vgic-v3.c b/virt/kvm/arm/vgic/vgic-v3.c index 8195f52ae6f0..c7423f3768e5 100644 --- a/virt/kvm/arm/vgic/vgic-v3.c +++ b/virt/kvm/arm/vgic/vgic-v3.c @@ -27,13 +27,6 @@ static bool group1_trap; static bool common_trap; static bool gicv4_enable; -void vgic_v3_set_npie(struct kvm_vcpu *vcpu) -{ - struct vgic_v3_cpu_if *cpuif = &vcpu->arch.vgic_cpu.vgic_v3; - - cpuif->vgic_hcr |= ICH_HCR_NPIE; -} - void vgic_v3_set_underflow(struct kvm_vcpu *vcpu) { struct vgic_v3_cpu_if *cpuif = &vcpu->arch.vgic_cpu.vgic_v3; @@ -55,17 +48,23 @@ void vgic_v3_fold_lr_state(struct kvm_vcpu *vcpu) int lr; unsigned long flags; - cpuif->vgic_hcr &= ~(ICH_HCR_UIE | ICH_HCR_NPIE); + cpuif->vgic_hcr &= ~ICH_HCR_UIE; for (lr = 0; lr < vgic_cpu->used_lrs; lr++) { u64 val = cpuif->vgic_lr[lr]; - u32 intid; + u32 intid, cpuid; struct vgic_irq *irq; + bool is_v2_sgi = false; - if (model == KVM_DEV_TYPE_ARM_VGIC_V3) + cpuid = val & GICH_LR_PHYSID_CPUID; + cpuid >>= GICH_LR_PHYSID_CPUID_SHIFT; + + if (model == KVM_DEV_TYPE_ARM_VGIC_V3) { intid = val & ICH_LR_VIRTUAL_ID_MASK; - else + } else { intid = val & GICH_LR_VIRTUALID; + is_v2_sgi = vgic_irq_is_sgi(intid); + } /* Notify fds when the guest EOI'ed a level-triggered IRQ */ if (lr_signals_eoi_mi(val) && vgic_valid_spi(vcpu->kvm, intid)) @@ -81,18 +80,16 @@ void vgic_v3_fold_lr_state(struct kvm_vcpu *vcpu) /* Always preserve the active bit */ irq->active = !!(val & ICH_LR_ACTIVE_BIT); + if (irq->active && is_v2_sgi) + irq->active_source = cpuid; + /* Edge is the only case where we preserve the pending bit */ if (irq->config == VGIC_CONFIG_EDGE && (val & ICH_LR_PENDING_BIT)) { irq->pending_latch = true; - if (vgic_irq_is_sgi(intid) && - model == KVM_DEV_TYPE_ARM_VGIC_V2) { - u32 cpuid = val & GICH_LR_PHYSID_CPUID; - - cpuid >>= GICH_LR_PHYSID_CPUID_SHIFT; + if (is_v2_sgi) irq->source |= (1 << cpuid); - } } /* @@ -133,10 +130,20 @@ void vgic_v3_populate_lr(struct kvm_vcpu *vcpu, struct vgic_irq *irq, int lr) { u32 model = vcpu->kvm->arch.vgic.vgic_model; u64 val = irq->intid; - bool allow_pending = true; + bool allow_pending = true, is_v2_sgi; - if (irq->active) + is_v2_sgi = (vgic_irq_is_sgi(irq->intid) && + model == KVM_DEV_TYPE_ARM_VGIC_V2); + + if (irq->active) { val |= ICH_LR_ACTIVE_BIT; + if (is_v2_sgi) + val |= irq->active_source << GICH_LR_PHYSID_CPUID_SHIFT; + if (vgic_irq_is_multi_sgi(irq)) { + allow_pending = false; + val |= ICH_LR_EOI; + } + } if (irq->hw) { val |= ICH_LR_HW; @@ -174,8 +181,10 @@ void vgic_v3_populate_lr(struct kvm_vcpu *vcpu, struct vgic_irq *irq, int lr) BUG_ON(!src); val |= (src - 1) << GICH_LR_PHYSID_CPUID_SHIFT; irq->source &= ~(1 << (src - 1)); - if (irq->source) + if (irq->source) { irq->pending_latch = true; + val |= ICH_LR_EOI; + } } } diff --git a/virt/kvm/arm/vgic/vgic.c b/virt/kvm/arm/vgic/vgic.c index 702936cbe173..97bfba8d9a59 100644 --- a/virt/kvm/arm/vgic/vgic.c +++ b/virt/kvm/arm/vgic/vgic.c @@ -725,14 +725,6 @@ static inline void vgic_set_underflow(struct kvm_vcpu *vcpu) vgic_v3_set_underflow(vcpu); } -static inline void vgic_set_npie(struct kvm_vcpu *vcpu) -{ - if (kvm_vgic_global_state.type == VGIC_V2) - vgic_v2_set_npie(vcpu); - else - vgic_v3_set_npie(vcpu); -} - /* Requires the ap_list_lock to be held. */ static int compute_ap_list_depth(struct kvm_vcpu *vcpu, bool *multi_sgi) @@ -746,17 +738,15 @@ static int compute_ap_list_depth(struct kvm_vcpu *vcpu, DEBUG_SPINLOCK_BUG_ON(!spin_is_locked(&vgic_cpu->ap_list_lock)); list_for_each_entry(irq, &vgic_cpu->ap_list_head, ap_list) { + int w; + spin_lock(&irq->irq_lock); /* GICv2 SGIs can count for more than one... */ - if (vgic_irq_is_sgi(irq->intid) && irq->source) { - int w = hweight8(irq->source); - - count += w; - *multi_sgi |= (w > 1); - } else { - count++; - } + w = vgic_irq_get_lr_count(irq); spin_unlock(&irq->irq_lock); + + count += w; + *multi_sgi |= (w > 1); } return count; } @@ -767,7 +757,6 @@ static void vgic_flush_lr_state(struct kvm_vcpu *vcpu) struct vgic_cpu *vgic_cpu = &vcpu->arch.vgic_cpu; struct vgic_irq *irq; int count; - bool npie = false; bool multi_sgi; u8 prio = 0xff; @@ -797,10 +786,8 @@ static void vgic_flush_lr_state(struct kvm_vcpu *vcpu) if (likely(vgic_target_oracle(irq) == vcpu)) { vgic_populate_lr(vcpu, irq, count++); - if (irq->source) { - npie = true; + if (irq->source) prio = irq->priority; - } } spin_unlock(&irq->irq_lock); @@ -813,9 +800,6 @@ static void vgic_flush_lr_state(struct kvm_vcpu *vcpu) } } - if (npie) - vgic_set_npie(vcpu); - vcpu->arch.vgic_cpu.used_lrs = count; /* Nuke remaining LRs */ diff --git a/virt/kvm/arm/vgic/vgic.h b/virt/kvm/arm/vgic/vgic.h index 830e815748a0..32c25d42c93f 100644 --- a/virt/kvm/arm/vgic/vgic.h +++ b/virt/kvm/arm/vgic/vgic.h @@ -110,6 +110,20 @@ static inline bool vgic_irq_is_mapped_level(struct vgic_irq *irq) return irq->config == VGIC_CONFIG_LEVEL && irq->hw; } +static inline int vgic_irq_get_lr_count(struct vgic_irq *irq) +{ + /* Account for the active state as an interrupt */ + if (vgic_irq_is_sgi(irq->intid) && irq->source) + return hweight8(irq->source) + irq->active; + + return irq_is_pending(irq) || irq->active; +} + +static inline bool vgic_irq_is_multi_sgi(struct vgic_irq *irq) +{ + return vgic_irq_get_lr_count(irq) > 1; +} + /* * This struct provides an intermediate representation of the fields contained * in the GICH_VMCR and ICH_VMCR registers, such that code exporting the GIC |