summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorRui Teng <rui.teng@amd.com>2019-10-18 16:15:42 +0800
committerRui Teng <rui.teng@amd.com>2019-10-18 16:16:15 +0800
commitddb75ca062ee46d047d79bf8e8da3c81a6373aa4 (patch)
tree6fd0f9b3d9a899d01b3197d5e7754beca5fe622e
parentae73402ee2e4a00f53819715b52d6e3f7ec814d2 (diff)
parent65bde8c9e4af39d123a6773a8f5a9e63ae332758 (diff)
Merge amd-staging-dkms-5.0 into amd-mainline-dkms-5.0
Change-Id: Ia84590ea2eca9058cd1b9b5f73e12783ed8639ee Signed-off-by: Rui Teng <rui.teng@amd.com>
-rw-r--r--drivers/gpu/drm/Kconfig4
-rw-r--r--drivers/gpu/drm/amd/amdgpu/Makefile4
-rw-r--r--drivers/gpu/drm/amd/amdgpu/amdgpu.h12
-rw-r--r--drivers/gpu/drm/amd/amdgpu/amdgpu_acp.c34
-rw-r--r--drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c71
-rw-r--r--drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h6
-rw-r--r--drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c134
-rw-r--r--drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c279
-rw-r--r--drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c207
-rw-r--r--drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c206
-rw-r--r--drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c155
-rw-r--r--drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h6
-rw-r--r--drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c14
-rw-r--r--drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.h2
-rw-r--r--drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c21
-rw-r--r--drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c13
-rw-r--r--drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c2
-rw-r--r--drivers/gpu/drm/amd/amdgpu/amdgpu_device.c90
-rw-r--r--drivers/gpu/drm/amd/amdgpu/amdgpu_display.c5
-rw-r--r--drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c121
-rw-r--r--drivers/gpu/drm/amd/amdgpu/amdgpu_encoders.c40
-rw-r--r--drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c10
-rw-r--r--drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c4
-rw-r--r--drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h1
-rw-r--r--drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h1
-rw-r--r--drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c4
-rw-r--r--drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c23
-rw-r--r--drivers/gpu/drm/amd/amdgpu/amdgpu_job.h2
-rw-r--r--drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c34
-rw-r--r--drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h2
-rw-r--r--drivers/gpu/drm/amd/amdgpu/amdgpu_object.c3
-rw-r--r--drivers/gpu/drm/amd/amdgpu/amdgpu_object.h11
-rw-r--r--drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c51
-rw-r--r--drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h9
-rw-r--r--drivers/gpu/drm/amd/amdgpu/amdgpu_tmz.c49
-rw-r--r--drivers/gpu/drm/amd/amdgpu/amdgpu_tmz.h39
-rw-r--r--drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c13
-rw-r--r--drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c6
-rw-r--r--drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h3
-rw-r--r--drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c43
-rw-r--r--drivers/gpu/drm/amd/amdgpu/arct_reg_init.c1
-rw-r--r--drivers/gpu/drm/amd/amdgpu/dce_v10_0.c34
-rw-r--r--drivers/gpu/drm/amd/amdgpu/dce_v11_0.c34
-rw-r--r--drivers/gpu/drm/amd/amdgpu/dce_v6_0.c40
-rw-r--r--drivers/gpu/drm/amd/amdgpu/dce_v8_0.c34
-rw-r--r--drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c45
-rw-r--r--drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c3
-rw-r--r--drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c3
-rw-r--r--drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c3
-rw-r--r--drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c1012
-rw-r--r--drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c20
-rw-r--r--drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.h2
-rw-r--r--drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c6
-rw-r--r--drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c89
-rw-r--r--drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c1
-rw-r--r--drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c19
-rw-r--r--drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.h2
-rw-r--r--drivers/gpu/drm/amd/amdgpu/navi10_reg_init.c1
-rw-r--r--drivers/gpu/drm/amd/amdgpu/navi12_reg_init.c1
-rw-r--r--drivers/gpu/drm/amd/amdgpu/navi14_reg_init.c1
-rw-r--r--drivers/gpu/drm/amd/amdgpu/nbio_v2_3.c16
-rw-r--r--drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c8
-rw-r--r--drivers/gpu/drm/amd/amdgpu/nv.c15
-rw-r--r--drivers/gpu/drm/amd/amdgpu/nvd.h1
-rw-r--r--drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c3
-rw-r--r--drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c2
-rw-r--r--drivers/gpu/drm/amd/amdgpu/soc15.c38
-rw-r--r--drivers/gpu/drm/amd/amdgpu/soc15.h2
-rw-r--r--drivers/gpu/drm/amd/amdgpu/soc15d.h1
-rw-r--r--drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c1
-rw-r--r--drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c21
-rw-r--r--drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c26
-rw-r--r--drivers/gpu/drm/amd/amdgpu/vega10_ih.c39
-rw-r--r--drivers/gpu/drm/amd/amdgpu/vega10_reg_init.c1
-rw-r--r--drivers/gpu/drm/amd/amdgpu/vega20_reg_init.c1
-rw-r--r--drivers/gpu/drm/amd/amdkcl/kcl_device_cgroup.c8
-rw-r--r--drivers/gpu/drm/amd/amdkcl/kcl_pci.c2
-rw-r--r--drivers/gpu/drm/amd/amdkfd/cik_event_interrupt.c8
-rw-r--r--drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler.h139
-rw-r--r--drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler_gfx10.asm1
-rw-r--r--drivers/gpu/drm/amd/amdkfd/kfd_chardev.c17
-rw-r--r--drivers/gpu/drm/amd/amdkfd/kfd_crat.c1
-rw-r--r--drivers/gpu/drm/amd/amdkfd/kfd_dbgdev.c18
-rw-r--r--drivers/gpu/drm/amd/amdkfd/kfd_dbgmgr.c8
-rw-r--r--drivers/gpu/drm/amd/amdkfd/kfd_device.c64
-rw-r--r--drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c42
-rw-r--r--drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h4
-rw-r--r--drivers/gpu/drm/amd/amdkfd/kfd_events.c12
-rw-r--r--drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c1
-rw-r--r--drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c3
-rw-r--r--drivers/gpu/drm/amd/amdkfd/kfd_iommu.c6
-rw-r--r--drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c1
-rw-r--r--drivers/gpu/drm/amd/amdkfd/kfd_module.c2
-rw-r--r--drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c37
-rw-r--r--drivers/gpu/drm/amd/amdkfd/kfd_packet_manager.c1
-rw-r--r--drivers/gpu/drm/amd/amdkfd/kfd_priv.h8
-rw-r--r--drivers/gpu/drm/amd/amdkfd/kfd_process.c20
-rw-r--r--drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c6
-rw-r--r--drivers/gpu/drm/amd/amdkfd/kfd_topology.c1
-rw-r--r--drivers/gpu/drm/amd/display/Kconfig32
-rw-r--r--drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c65
-rw-r--r--drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.c9
-rw-r--r--drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c12
-rw-r--r--drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.c10
-rw-r--r--drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c20
-rw-r--r--drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_pp_smu.c109
-rw-r--r--drivers/gpu/drm/amd/display/dc/bios/bios_parser.c7
-rw-r--r--drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c8
-rw-r--r--drivers/gpu/drm/amd/display/dc/clk_mgr/clk_mgr.c25
-rw-r--r--drivers/gpu/drm/amd/display/dc/clk_mgr/dcn20/dcn20_clk_mgr.c24
-rw-r--r--drivers/gpu/drm/amd/display/dc/clk_mgr/dcn20/dcn20_clk_mgr.h1
-rw-r--r--drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c135
-rw-r--r--drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.h11
-rw-r--r--drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr_vbios_smu.c19
-rw-r--r--drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr_vbios_smu.h4
-rw-r--r--drivers/gpu/drm/amd/display/dc/core/dc.c100
-rw-r--r--drivers/gpu/drm/amd/display/dc/core/dc_link.c45
-rw-r--r--drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c33
-rw-r--r--drivers/gpu/drm/amd/display/dc/core/dc_stream.c4
-rw-r--r--drivers/gpu/drm/amd/display/dc/dc.h17
-rw-r--r--drivers/gpu/drm/amd/display/dc/dc_hw_types.h5
-rw-r--r--drivers/gpu/drm/amd/display/dc/dc_link.h17
-rw-r--r--drivers/gpu/drm/amd/display/dc/dc_stream.h3
-rw-r--r--drivers/gpu/drm/amd/display/dc/dce/dce_abm.c3
-rw-r--r--drivers/gpu/drm/amd/display/dc/dce/dce_aux.c11
-rw-r--r--drivers/gpu/drm/amd/display/dc/dce100/dce100_resource.c1
-rw-r--r--drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c28
-rw-r--r--drivers/gpu/drm/amd/display/dc/dce110/dce110_resource.c1
-rw-r--r--drivers/gpu/drm/amd/display/dc/dce112/dce112_resource.c1
-rw-r--r--drivers/gpu/drm/amd/display/dc/dce120/dce120_resource.c1
-rw-r--r--drivers/gpu/drm/amd/display/dc/dce80/dce80_resource.c1
-rw-r--r--drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.c4
-rw-r--r--drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.h5
-rw-r--r--drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c16
-rw-r--r--drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.h1
-rw-r--r--drivers/gpu/drm/amd/display/dc/dcn10/dcn10_opp.c3
-rw-r--r--drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c60
-rw-r--r--drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.h5
-rw-r--r--drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c1
-rw-r--r--drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.c62
-rw-r--r--drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.h5
-rw-r--r--drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp.c9
-rw-r--r--drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp.h5
-rw-r--r--drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dwb_scl.c4
-rw-r--r--drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c38
-rw-r--r--drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.c2
-rw-r--r--drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c27
-rw-r--r--drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.h3
-rw-r--r--drivers/gpu/drm/amd/display/dc/dcn20/dcn20_stream_encoder.c4
-rw-r--r--drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c36
-rw-r--r--drivers/gpu/drm/amd/display/dc/dm_pp_smu.h2
-rw-r--r--drivers/gpu/drm/amd/display/dc/dml/dcn21/display_mode_vba_21.c13
-rw-r--r--drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h2
-rw-r--r--drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.c2
-rw-r--r--drivers/gpu/drm/amd/display/dc/dsc/rc_calc.c3
-rw-r--r--drivers/gpu/drm/amd/display/dc/inc/dc_link_dp.h3
-rw-r--r--drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h11
-rw-r--r--drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr_internal.h25
-rw-r--r--drivers/gpu/drm/amd/display/dc/inc/hw/link_encoder.h4
-rw-r--r--drivers/gpu/drm/amd/display/dc/inc/hw/opp.h1
-rw-r--r--drivers/gpu/drm/amd/display/dc/inc/hw/stream_encoder.h5
-rw-r--r--drivers/gpu/drm/amd/display/dc/inc/hw/timing_generator.h2
-rw-r--r--drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h13
-rw-r--r--drivers/gpu/drm/amd/display/include/ddc_service_types.h2
-rw-r--r--drivers/gpu/drm/amd/display/modules/freesync/freesync.c21
-rw-r--r--drivers/gpu/drm/amd/display/modules/hdcp/hdcp.h4
-rw-r--r--drivers/gpu/drm/amd/display/modules/hdcp/hdcp1_execution.c4
-rw-r--r--drivers/gpu/drm/amd/display/modules/hdcp/hdcp1_transition.c12
-rw-r--r--drivers/gpu/drm/amd/display/modules/hdcp/hdcp_log.c8
-rw-r--r--drivers/gpu/drm/amd/display/modules/inc/mod_info_packet.h3
-rw-r--r--drivers/gpu/drm/amd/display/modules/info_packet/info_packet.c98
-rw-r--r--drivers/gpu/drm/amd/dkms/m4/devcgroup-check-permission.m416
-rw-r--r--drivers/gpu/drm/amd/dkms/m4/kernel.m42
-rw-r--r--drivers/gpu/drm/amd/include/asic_reg/gc/gc_9_0_offset.h18
-rw-r--r--drivers/gpu/drm/amd/include/asic_reg/gc/gc_9_0_sh_mask.h18
-rw-r--r--drivers/gpu/drm/amd/include/asic_reg/oss/osssys_4_0_sh_mask.h4
-rw-r--r--drivers/gpu/drm/amd/include/asic_reg/smuio/smuio_11_0_0_offset.h92
-rw-r--r--drivers/gpu/drm/amd/include/asic_reg/smuio/smuio_11_0_0_sh_mask.h176
-rw-r--r--drivers/gpu/drm/amd/include/kgd_kfd_interface.h8
-rw-r--r--drivers/gpu/drm/amd/include/kgd_pp_interface.h6
-rw-r--r--drivers/gpu/drm/amd/powerplay/amd_powerplay.c18
-rw-r--r--drivers/gpu/drm/amd/powerplay/amdgpu_smu.c148
-rw-r--r--drivers/gpu/drm/amd/powerplay/arcturus_ppt.c32
-rw-r--r--drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c6
-rw-r--r--drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c6
-rw-r--r--drivers/gpu/drm/amd/powerplay/hwmgr/vega20_baco.c23
-rw-r--r--drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.c22
-rw-r--r--drivers/gpu/drm/amd/powerplay/inc/amdgpu_smu.h37
-rw-r--r--drivers/gpu/drm/amd/powerplay/inc/hwmgr.h1
-rw-r--r--drivers/gpu/drm/amd/powerplay/inc/smu_types.h1
-rw-r--r--drivers/gpu/drm/amd/powerplay/inc/vega20_ppsmc.h3
-rw-r--r--drivers/gpu/drm/amd/powerplay/navi10_ppt.c51
-rw-r--r--drivers/gpu/drm/amd/powerplay/renoir_ppt.c366
-rw-r--r--drivers/gpu/drm/amd/powerplay/smu_v11_0.c113
-rw-r--r--drivers/gpu/drm/amd/powerplay/smu_v12_0.c84
-rw-r--r--drivers/gpu/drm/amd/powerplay/smumgr/polaris10_smumgr.c2
-rw-r--r--drivers/gpu/drm/amd/powerplay/smumgr/smu8_smumgr.c2
-rw-r--r--drivers/gpu/drm/amd/powerplay/smumgr/vegam_smumgr.c2
-rw-r--r--drivers/gpu/drm/amd/powerplay/vega20_ppt.c50
-rw-r--r--drivers/gpu/drm/drm_mode_object.c4
-rw-r--r--include/drm/drm_dp_helper.h26
-rw-r--r--include/kcl/kcl_device_cgroup.h6
-rw-r--r--include/linux/device_cgroup.h19
-rw-r--r--include/uapi/drm/amdgpu_drm.h12
-rw-r--r--security/device_cgroup.c16
205 files changed, 4236 insertions, 2114 deletions
diff --git a/drivers/gpu/drm/Kconfig b/drivers/gpu/drm/Kconfig
index 98b033e675db..262c638f176d 100644
--- a/drivers/gpu/drm/Kconfig
+++ b/drivers/gpu/drm/Kconfig
@@ -213,9 +213,9 @@ config DRM_AMDGPU
tristate "AMD GPU"
depends on DRM && PCI && MMU
select FW_LOADER
- select DRM_KMS_HELPER
+ select DRM_KMS_HELPER
select DRM_SCHED
- select DRM_TTM
+ select DRM_TTM
select POWER_SUPPLY
select HWMON
select BACKLIGHT_CLASS_DEVICE
diff --git a/drivers/gpu/drm/amd/amdgpu/Makefile b/drivers/gpu/drm/amd/amdgpu/Makefile
index ea9122b5796e..1a1b9463f83b 100644
--- a/drivers/gpu/drm/amd/amdgpu/Makefile
+++ b/drivers/gpu/drm/amd/amdgpu/Makefile
@@ -54,9 +54,9 @@ amdgpu-y += amdgpu_device.o amdgpu_kms.o \
amdgpu_gtt_mgr.o amdgpu_vram_mgr.o amdgpu_virt.o amdgpu_atomfirmware.o \
amdgpu_vf_error.o amdgpu_sched.o amdgpu_debugfs.o amdgpu_ids.o \
amdgpu_gmc.o amdgpu_mmhub.o amdgpu_xgmi.o amdgpu_csa.o amdgpu_sem.o \
- amdgpu_ras.o amdgpu_vm_cpu.o amdgpu_vm_sdma.o amdgpu_pmu.o \
+ amdgpu_ras.o amdgpu_vm_cpu.o amdgpu_vm_sdma.o \
amdgpu_discovery.o amdgpu_ras_eeprom.o amdgpu_nbio.o amdgpu_umc.o \
- smu_v11_0_i2c.o
+ smu_v11_0_i2c.o amdgpu_tmz.o
amdgpu-$(CONFIG_PERF_EVENTS) += amdgpu_pmu.o
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index 6775647f0ba5..8efaa957743e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -89,6 +89,7 @@
#include "amdgpu_mes.h"
#include "amdgpu_umc.h"
#include "amdgpu_mmhub.h"
+#include "amdgpu_tmz.h"
#define MAX_GPU_INSTANCE 16
@@ -107,6 +108,8 @@ struct amdgpu_mgpu_info
uint32_t num_apu;
};
+#define AMDGPU_MAX_TIMEOUT_PARAM_LENTH 256
+
/*
* Modules parameters.
*/
@@ -123,6 +126,7 @@ extern int amdgpu_disp_priority;
extern int amdgpu_hw_i2c;
extern int amdgpu_pcie_gen2;
extern int amdgpu_msi;
+extern char amdgpu_lockup_timeout[AMDGPU_MAX_TIMEOUT_PARAM_LENTH];
extern int amdgpu_dpm;
extern int amdgpu_fw_load_type;
extern int amdgpu_aspm;
@@ -173,6 +177,8 @@ extern int sched_policy;
static const int sched_policy = KFD_SCHED_POLICY_HWS;
#endif
+extern int amdgpu_tmz;
+
#ifdef CONFIG_DRM_AMDGPU_SI
extern int amdgpu_si_support;
#endif
@@ -439,7 +445,6 @@ struct amdgpu_fpriv {
};
int amdgpu_file_to_fpriv(struct file *filp, struct amdgpu_fpriv **fpriv);
-int amdgpu_device_get_job_timeout_settings(struct amdgpu_device *adev);
int amdgpu_ib_get(struct amdgpu_device *adev, struct amdgpu_vm *vm,
unsigned size, struct amdgpu_ib *ib);
@@ -972,6 +977,9 @@ struct amdgpu_device {
bool enable_mes;
struct amdgpu_mes mes;
+ /* tmz */
+ struct amdgpu_tmz tmz;
+
struct amdgpu_ip_block ip_blocks[AMDGPU_MAX_IP_NUM];
int num_ip_blocks;
struct mutex mn_lock;
@@ -983,7 +991,7 @@ struct amdgpu_device {
atomic64_t gart_pin_size;
/* soc15 register offset based on ip, instance and segment */
- uint32_t *reg_offset[MAX_HWIP][HWIP_MAX_INSTANCE];
+ uint32_t *reg_offset[MAX_HWIP][HWIP_MAX_INSTANCE];
const struct amdgpu_df_funcs *df_funcs;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_acp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_acp.c
index 500d890d19cd..35482f3c1841 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_acp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_acp.c
@@ -192,7 +192,7 @@ static int acp_hw_init(void *handle)
u32 val = 0;
u32 count = 0;
struct device *dev;
- struct i2s_platform_data *i2s_pdata;
+ struct i2s_platform_data *i2s_pdata = NULL;
struct amdgpu_device *adev = (struct amdgpu_device *)handle;
@@ -234,20 +234,21 @@ static int acp_hw_init(void *handle)
adev->acp.acp_cell = kcalloc(ACP_DEVS, sizeof(struct mfd_cell),
GFP_KERNEL);
- if (adev->acp.acp_cell == NULL)
- return -ENOMEM;
+ if (adev->acp.acp_cell == NULL) {
+ r = -ENOMEM;
+ goto failure;
+ }
adev->acp.acp_res = kcalloc(5, sizeof(struct resource), GFP_KERNEL);
if (adev->acp.acp_res == NULL) {
- kfree(adev->acp.acp_cell);
- return -ENOMEM;
+ r = -ENOMEM;
+ goto failure;
}
i2s_pdata = kcalloc(3, sizeof(struct i2s_platform_data), GFP_KERNEL);
if (i2s_pdata == NULL) {
- kfree(adev->acp.acp_res);
- kfree(adev->acp.acp_cell);
- return -ENOMEM;
+ r = -ENOMEM;
+ goto failure;
}
switch (adev->asic_type) {
@@ -344,14 +345,14 @@ static int acp_hw_init(void *handle)
r = mfd_add_hotplug_devices(adev->acp.parent, adev->acp.acp_cell,
ACP_DEVS);
if (r)
- return r;
+ goto failure;
for (i = 0; i < ACP_DEVS ; i++) {
dev = get_mfd_cell_dev(adev->acp.acp_cell[i].name, i);
r = pm_genpd_add_device(&adev->acp.acp_genpd->gpd, dev);
if (r) {
dev_err(dev, "Failed to add dev to genpd\n");
- return r;
+ goto failure;
}
}
@@ -370,7 +371,8 @@ static int acp_hw_init(void *handle)
break;
if (--count == 0) {
dev_err(&adev->pdev->dev, "Failed to reset ACP\n");
- return -ETIMEDOUT;
+ r = -ETIMEDOUT;
+ goto failure;
}
udelay(100);
}
@@ -387,7 +389,8 @@ static int acp_hw_init(void *handle)
break;
if (--count == 0) {
dev_err(&adev->pdev->dev, "Failed to reset ACP\n");
- return -ETIMEDOUT;
+ r = -ETIMEDOUT;
+ goto failure;
}
udelay(100);
}
@@ -396,6 +399,13 @@ static int acp_hw_init(void *handle)
val &= ~ACP_SOFT_RESET__SoftResetAud_MASK;
cgs_write_register(adev->acp.cgs_device, mmACP_SOFT_RESET, val);
return 0;
+
+failure:
+ kfree(i2s_pdata);
+ kfree(adev->acp.acp_res);
+ kfree(adev->acp.acp_cell);
+ kfree(adev->acp.acp_genpd);
+ return r;
}
/**
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
index f111cc1ee6da..c3d654a25837 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
@@ -68,53 +68,10 @@ void amdgpu_amdkfd_fini(void)
void amdgpu_amdkfd_device_probe(struct amdgpu_device *adev)
{
- const struct kfd2kgd_calls *kfd2kgd;
bool vf = amdgpu_sriov_vf(adev);
- switch (adev->asic_type) {
-#ifdef CONFIG_DRM_AMDGPU_CIK
- case CHIP_KAVERI:
- case CHIP_HAWAII:
- kfd2kgd = amdgpu_amdkfd_gfx_7_get_functions();
- break;
-#endif
- case CHIP_CARRIZO:
- case CHIP_TONGA:
- case CHIP_FIJI:
- case CHIP_POLARIS10:
- case CHIP_POLARIS11:
- case CHIP_POLARIS12:
- case CHIP_VEGAM:
- kfd2kgd = amdgpu_amdkfd_gfx_8_0_get_functions();
- break;
- case CHIP_VEGA10:
- case CHIP_VEGA12:
- case CHIP_VEGA20:
- case CHIP_RAVEN:
- case CHIP_RENOIR:
-#if LINUX_VERSION_CODE < KERNEL_VERSION(4, 16, 0) && defined(BUILD_AS_DKMS)
- if (adev->asic_type == CHIP_RAVEN) {
- dev_dbg(adev->dev, "DKMS installed kfd does not support Raven for kernel < 4.16\n");
- return;
- }
-#endif
- kfd2kgd = amdgpu_amdkfd_gfx_9_0_get_functions();
- break;
- case CHIP_ARCTURUS:
- kfd2kgd = amdgpu_amdkfd_arcturus_get_functions();
- break;
- case CHIP_NAVI10:
- case CHIP_NAVI14:
- case CHIP_NAVI12:
- kfd2kgd = amdgpu_amdkfd_gfx_10_0_get_functions();
- break;
- default:
- dev_info(adev->dev, "kfd not supported on this ASIC\n");
- return;
- }
-
adev->kfd.dev = kgd2kfd_probe((struct kgd_dev *)adev,
- adev->pdev, kfd2kgd, adev->asic_type, vf);
+ adev->pdev, adev->asic_type, vf);
if (adev->kfd.dev)
amdgpu_amdkfd_total_mem_size += adev->gmc.real_vram_size;
@@ -740,33 +697,7 @@ 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;
-}
-
-struct kfd2kgd_calls *amdgpu_amdkfd_arcturus_get_functions(void)
-{
- return NULL;
-}
-
-struct kfd2kgd_calls *amdgpu_amdkfd_gfx_10_0_get_functions(void)
-{
- return NULL;
-}
-
struct kfd_dev *kgd2kfd_probe(struct kgd_dev *kgd, struct pci_dev *pdev,
- const struct kfd2kgd_calls *f2g,
unsigned int asic_type, bool vf)
{
return NULL;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
index 2a6d2a4b4039..3cbe66d5f61a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
@@ -138,11 +138,6 @@ int amdgpu_amdkfd_submit_ib(struct kgd_dev *kgd, enum kgd_engine_type engine,
void amdgpu_amdkfd_set_compute_idle(struct kgd_dev *kgd, bool idle);
bool amdgpu_amdkfd_have_atomics_support(struct kgd_dev *kgd);
-struct kfd2kgd_calls *amdgpu_amdkfd_gfx_7_get_functions(void);
-struct kfd2kgd_calls *amdgpu_amdkfd_gfx_8_0_get_functions(void);
-struct kfd2kgd_calls *amdgpu_amdkfd_gfx_9_0_get_functions(void);
-struct kfd2kgd_calls *amdgpu_amdkfd_arcturus_get_functions(void);
-struct kfd2kgd_calls *amdgpu_amdkfd_gfx_10_0_get_functions(void);
int amdgpu_amdkfd_copy_mem_to_mem(struct kgd_dev *kgd, struct kgd_mem *src_mem,
uint64_t src_offset, struct kgd_mem *dst_mem,
uint64_t dest_offset, uint64_t size, struct dma_fence **f,
@@ -262,7 +257,6 @@ void amdgpu_amdkfd_debug_mem_fence(struct kgd_dev *kgd);
int kgd2kfd_init(void);
void kgd2kfd_exit(void);
struct kfd_dev *kgd2kfd_probe(struct kgd_dev *kgd, struct pci_dev *pdev,
- const struct kfd2kgd_calls *f2g,
unsigned int asic_type, bool vf);
bool kgd2kfd_device_init(struct kfd_dev *kfd,
struct drm_device *ddev,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c
index ce0ceb71ef35..db39c6653cce 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c
@@ -70,11 +70,11 @@ static inline struct v9_sdma_mqd *get_sdma_mqd(void *mqd)
return (struct v9_sdma_mqd *)mqd;
}
-static uint32_t get_sdma_base_addr(struct amdgpu_device *adev,
+static uint32_t get_sdma_rlc_reg_offset(struct amdgpu_device *adev,
unsigned int engine_id,
unsigned int queue_id)
{
- uint32_t base[8] = {
+ uint32_t sdma_engine_reg_base[8] = {
SOC15_REG_OFFSET(SDMA0, 0,
mmSDMA0_RLC0_RB_CNTL) - mmSDMA0_RLC0_RB_CNTL,
SOC15_REG_OFFSET(SDMA1, 0,
@@ -92,65 +92,37 @@ static uint32_t get_sdma_base_addr(struct amdgpu_device *adev,
SOC15_REG_OFFSET(SDMA7, 0,
mmSDMA7_RLC0_RB_CNTL) - mmSDMA7_RLC0_RB_CNTL
};
- uint32_t retval;
- retval = base[engine_id] + queue_id * (mmSDMA0_RLC1_RB_CNTL -
- mmSDMA0_RLC0_RB_CNTL);
+ uint32_t retval = sdma_engine_reg_base[engine_id]
+ + queue_id * (mmSDMA0_RLC1_RB_CNTL - mmSDMA0_RLC0_RB_CNTL);
- pr_debug("sdma base address: 0x%x\n", retval);
+ pr_debug("RLC register offset for SDMA%d RLC%d: 0x%x\n", engine_id,
+ queue_id, retval);
return retval;
}
-static u32 sdma_v4_0_get_reg_offset(struct amdgpu_device *adev,
- u32 instance, u32 offset)
-{
- switch (instance) {
- case 0:
- return (adev->reg_offset[SDMA0_HWIP][0][0] + offset);
- case 1:
- return (adev->reg_offset[SDMA1_HWIP][0][1] + offset);
- case 2:
- return (adev->reg_offset[SDMA2_HWIP][0][1] + offset);
- case 3:
- return (adev->reg_offset[SDMA3_HWIP][0][1] + offset);
- case 4:
- return (adev->reg_offset[SDMA4_HWIP][0][1] + offset);
- case 5:
- return (adev->reg_offset[SDMA5_HWIP][0][1] + offset);
- case 6:
- return (adev->reg_offset[SDMA6_HWIP][0][1] + offset);
- case 7:
- return (adev->reg_offset[SDMA7_HWIP][0][1] + offset);
- default:
- break;
- }
- return 0;
-}
-
static int kgd_hqd_sdma_load(struct kgd_dev *kgd, void *mqd,
uint32_t __user *wptr, struct mm_struct *mm)
{
struct amdgpu_device *adev = get_amdgpu_device(kgd);
struct v9_sdma_mqd *m;
- uint32_t sdma_base_addr, sdmax_gfx_context_cntl;
+ uint32_t sdma_rlc_reg_offset;
unsigned long end_jiffies;
uint32_t data;
uint64_t data64;
uint64_t __user *wptr64 = (uint64_t __user *)wptr;
m = get_sdma_mqd(mqd);
- sdma_base_addr = get_sdma_base_addr(adev, m->sdma_engine_id,
+ sdma_rlc_reg_offset = get_sdma_rlc_reg_offset(adev, m->sdma_engine_id,
m->sdma_queue_id);
- sdmax_gfx_context_cntl = sdma_v4_0_get_reg_offset(adev,
- m->sdma_engine_id, mmSDMA0_GFX_CONTEXT_CNTL);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_CNTL,
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_CNTL,
m->sdmax_rlcx_rb_cntl & (~SDMA0_RLC0_RB_CNTL__RB_ENABLE_MASK));
end_jiffies = msecs_to_jiffies(2000) + jiffies;
while (true) {
- data = RREG32(sdma_base_addr + mmSDMA0_RLC0_CONTEXT_STATUS);
+ data = RREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_CONTEXT_STATUS);
if (data & SDMA0_RLC0_CONTEXT_STATUS__IDLE_MASK)
break;
if (time_after(jiffies, end_jiffies)) {
@@ -159,46 +131,43 @@ static int kgd_hqd_sdma_load(struct kgd_dev *kgd, void *mqd,
}
usleep_range(500, 1000);
}
- data = RREG32(sdmax_gfx_context_cntl);
- data = REG_SET_FIELD(data, SDMA0_GFX_CONTEXT_CNTL,
- RESUME_CTX, 0);
- WREG32(sdmax_gfx_context_cntl, data);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_DOORBELL_OFFSET,
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_DOORBELL_OFFSET,
m->sdmax_rlcx_doorbell_offset);
data = REG_SET_FIELD(m->sdmax_rlcx_doorbell, SDMA0_RLC0_DOORBELL,
ENABLE, 1);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_DOORBELL, data);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_RPTR, m->sdmax_rlcx_rb_rptr);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_RPTR_HI,
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_DOORBELL, data);
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_RPTR,
+ m->sdmax_rlcx_rb_rptr);
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_RPTR_HI,
m->sdmax_rlcx_rb_rptr_hi);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_MINOR_PTR_UPDATE, 1);
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_MINOR_PTR_UPDATE, 1);
if (read_user_wptr(mm, wptr64, data64)) {
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_WPTR,
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_WPTR,
lower_32_bits(data64));
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_WPTR_HI,
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_WPTR_HI,
upper_32_bits(data64));
} else {
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_WPTR,
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_WPTR,
m->sdmax_rlcx_rb_rptr);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_WPTR_HI,
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_WPTR_HI,
m->sdmax_rlcx_rb_rptr_hi);
}
- WREG32(sdma_base_addr + mmSDMA0_RLC0_MINOR_PTR_UPDATE, 0);
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_MINOR_PTR_UPDATE, 0);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_BASE, m->sdmax_rlcx_rb_base);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_BASE_HI,
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_BASE, m->sdmax_rlcx_rb_base);
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_BASE_HI,
m->sdmax_rlcx_rb_base_hi);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_RPTR_ADDR_LO,
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_RPTR_ADDR_LO,
m->sdmax_rlcx_rb_rptr_addr_lo);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_RPTR_ADDR_HI,
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_RPTR_ADDR_HI,
m->sdmax_rlcx_rb_rptr_addr_hi);
data = REG_SET_FIELD(m->sdmax_rlcx_rb_cntl, SDMA0_RLC0_RB_CNTL,
RB_ENABLE, 1);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_CNTL, data);
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_CNTL, data);
return 0;
}
@@ -208,7 +177,8 @@ static int kgd_hqd_sdma_dump(struct kgd_dev *kgd,
uint32_t (**dump)[2], uint32_t *n_regs)
{
struct amdgpu_device *adev = get_amdgpu_device(kgd);
- uint32_t sdma_base_addr = get_sdma_base_addr(adev, engine_id, queue_id);
+ uint32_t sdma_rlc_reg_offset = get_sdma_rlc_reg_offset(adev,
+ engine_id, queue_id);
uint32_t i = 0, reg;
#undef HQD_N_REGS
#define HQD_N_REGS (19+6+7+10)
@@ -218,15 +188,15 @@ static int kgd_hqd_sdma_dump(struct kgd_dev *kgd,
return -ENOMEM;
for (reg = mmSDMA0_RLC0_RB_CNTL; reg <= mmSDMA0_RLC0_DOORBELL; reg++)
- DUMP_REG(sdma_base_addr + reg);
+ DUMP_REG(sdma_rlc_reg_offset + reg);
for (reg = mmSDMA0_RLC0_STATUS; reg <= mmSDMA0_RLC0_CSA_ADDR_HI; reg++)
- DUMP_REG(sdma_base_addr + reg);
+ DUMP_REG(sdma_rlc_reg_offset + reg);
for (reg = mmSDMA0_RLC0_IB_SUB_REMAIN;
reg <= mmSDMA0_RLC0_MINOR_PTR_UPDATE; reg++)
- DUMP_REG(sdma_base_addr + reg);
+ DUMP_REG(sdma_rlc_reg_offset + reg);
for (reg = mmSDMA0_RLC0_MIDCMD_DATA0;
reg <= mmSDMA0_RLC0_MIDCMD_CNTL; reg++)
- DUMP_REG(sdma_base_addr + reg);
+ DUMP_REG(sdma_rlc_reg_offset + reg);
WARN_ON_ONCE(i != HQD_N_REGS);
*n_regs = i;
@@ -238,14 +208,14 @@ static bool kgd_hqd_sdma_is_occupied(struct kgd_dev *kgd, void *mqd)
{
struct amdgpu_device *adev = get_amdgpu_device(kgd);
struct v9_sdma_mqd *m;
- uint32_t sdma_base_addr;
+ uint32_t sdma_rlc_reg_offset;
uint32_t sdma_rlc_rb_cntl;
m = get_sdma_mqd(mqd);
- sdma_base_addr = get_sdma_base_addr(adev, m->sdma_engine_id,
+ sdma_rlc_reg_offset = get_sdma_rlc_reg_offset(adev, m->sdma_engine_id,
m->sdma_queue_id);
- sdma_rlc_rb_cntl = RREG32(sdma_base_addr + mmSDMA0_RLC0_RB_CNTL);
+ sdma_rlc_rb_cntl = RREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_CNTL);
if (sdma_rlc_rb_cntl & SDMA0_RLC0_RB_CNTL__RB_ENABLE_MASK)
return true;
@@ -258,20 +228,20 @@ static int kgd_hqd_sdma_destroy(struct kgd_dev *kgd, void *mqd,
{
struct amdgpu_device *adev = get_amdgpu_device(kgd);
struct v9_sdma_mqd *m;
- uint32_t sdma_base_addr;
+ uint32_t sdma_rlc_reg_offset;
uint32_t temp;
unsigned long end_jiffies = (utimeout * HZ / 1000) + jiffies;
m = get_sdma_mqd(mqd);
- sdma_base_addr = get_sdma_base_addr(adev, m->sdma_engine_id,
+ sdma_rlc_reg_offset = get_sdma_rlc_reg_offset(adev, m->sdma_engine_id,
m->sdma_queue_id);
- temp = RREG32(sdma_base_addr + mmSDMA0_RLC0_RB_CNTL);
+ temp = RREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_CNTL);
temp = temp & ~SDMA0_RLC0_RB_CNTL__RB_ENABLE_MASK;
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_CNTL, temp);
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_CNTL, temp);
while (true) {
- temp = RREG32(sdma_base_addr + mmSDMA0_RLC0_CONTEXT_STATUS);
+ temp = RREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_CONTEXT_STATUS);
if (temp & SDMA0_RLC0_CONTEXT_STATUS__IDLE_MASK)
break;
if (time_after(jiffies, end_jiffies)) {
@@ -281,19 +251,19 @@ static int kgd_hqd_sdma_destroy(struct kgd_dev *kgd, void *mqd,
usleep_range(500, 1000);
}
- WREG32(sdma_base_addr + mmSDMA0_RLC0_DOORBELL, 0);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_CNTL,
- RREG32(sdma_base_addr + mmSDMA0_RLC0_RB_CNTL) |
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_DOORBELL, 0);
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_CNTL,
+ RREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_CNTL) |
SDMA0_RLC0_RB_CNTL__RB_ENABLE_MASK);
- m->sdmax_rlcx_rb_rptr = RREG32(sdma_base_addr + mmSDMA0_RLC0_RB_RPTR);
+ m->sdmax_rlcx_rb_rptr = RREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_RPTR);
m->sdmax_rlcx_rb_rptr_hi =
- RREG32(sdma_base_addr + mmSDMA0_RLC0_RB_RPTR_HI);
+ RREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_RPTR_HI);
return 0;
}
-static const struct kfd2kgd_calls kfd2kgd = {
+const struct kfd2kgd_calls arcturus_kfd2kgd = {
.program_sh_mem_settings = kgd_gfx_v9_program_sh_mem_settings,
.set_pasid_vmid_mapping = kgd_gfx_v9_set_pasid_vmid_mapping,
.init_interrupts = kgd_gfx_v9_init_interrupts,
@@ -309,10 +279,8 @@ static const struct kfd2kgd_calls kfd2kgd = {
.address_watch_execute = kgd_gfx_v9_address_watch_execute,
.wave_control_execute = kgd_gfx_v9_wave_control_execute,
.address_watch_get_offset = kgd_gfx_v9_address_watch_get_offset,
- .get_atc_vmid_pasid_mapping_pasid =
- kgd_gfx_v9_get_atc_vmid_pasid_mapping_pasid,
- .get_atc_vmid_pasid_mapping_valid =
- kgd_gfx_v9_get_atc_vmid_pasid_mapping_valid,
+ .get_atc_vmid_pasid_mapping_info =
+ kgd_gfx_v9_get_atc_vmid_pasid_mapping_info,
.get_tile_config = kgd_gfx_v9_get_tile_config,
.set_vm_context_page_table_base = kgd_gfx_v9_set_vm_context_page_table_base,
.invalidate_tlbs = kgd_gfx_v9_invalidate_tlbs,
@@ -326,9 +294,3 @@ static const struct kfd2kgd_calls kfd2kgd = {
.get_iq_wait_times = kgd_gfx_v9_get_iq_wait_times,
.build_grace_period_packet_info = kgd_gfx_v9_build_grace_period_packet_info,
};
-
-struct kfd2kgd_calls *amdgpu_amdkfd_arcturus_get_functions(void)
-{
- return (struct kfd2kgd_calls *)&kfd2kgd;
-}
-
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c
index ae0862b05dd3..b5091e31c83f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c
@@ -22,16 +22,10 @@
#undef pr_fmt
#define pr_fmt(fmt) "kfd2kgd: " fmt
-#include <linux/module.h>
-#include <linux/fdtable.h>
-#include <linux/uaccess.h>
-#include <linux/firmware.h>
#include <linux/mmu_context.h>
#include <drm/drmP.h>
#include "amdgpu.h"
#include "amdgpu_amdkfd.h"
-#include "amdgpu_ucode.h"
-#include "soc15_hw_ip.h"
#include "gc/gc_10_1_0_offset.h"
#include "gc/gc_10_1_0_sh_mask.h"
#include "navi10_enum.h"
@@ -43,6 +37,7 @@
#include "v10_structs.h"
#include "nv.h"
#include "nvd.h"
+#include "gfxhub_v2_0.h"
enum hqd_dequeue_request_type {
NO_ACTION = 0,
@@ -51,63 +46,6 @@ enum hqd_dequeue_request_type {
SAVE_WAVES
};
-/*
- * Register access functions
- */
-
-static void kgd_program_sh_mem_settings(struct kgd_dev *kgd, uint32_t vmid,
- uint32_t sh_mem_config,
- uint32_t sh_mem_ape1_base, uint32_t sh_mem_ape1_limit,
- uint32_t sh_mem_bases);
-static int kgd_set_pasid_vmid_mapping(struct kgd_dev *kgd, unsigned int pasid,
- unsigned int vmid);
-static int kgd_init_interrupts(struct kgd_dev *kgd, uint32_t pipe_id);
-static int kgd_hqd_load(struct kgd_dev *kgd, void *mqd, uint32_t pipe_id,
- uint32_t queue_id, uint32_t __user *wptr,
- uint32_t wptr_shift, uint32_t wptr_mask,
- struct mm_struct *mm);
-static int kgd_hqd_dump(struct kgd_dev *kgd,
- uint32_t pipe_id, uint32_t queue_id,
- uint32_t (**dump)[2], uint32_t *n_regs);
-static int kgd_hqd_sdma_load(struct kgd_dev *kgd, void *mqd,
- uint32_t __user *wptr, struct mm_struct *mm);
-static int kgd_hqd_sdma_dump(struct kgd_dev *kgd,
- uint32_t engine_id, uint32_t queue_id,
- uint32_t (**dump)[2], uint32_t *n_regs);
-static bool kgd_hqd_is_occupied(struct kgd_dev *kgd, uint64_t queue_address,
- uint32_t pipe_id, uint32_t queue_id);
-static bool kgd_hqd_sdma_is_occupied(struct kgd_dev *kgd, void *mqd);
-static int kgd_hqd_destroy(struct kgd_dev *kgd, void *mqd,
- enum kfd_preempt_type reset_type,
- unsigned int utimeout, uint32_t pipe_id,
- uint32_t queue_id);
-static int kgd_hqd_sdma_destroy(struct kgd_dev *kgd, void *mqd,
- unsigned int utimeout);
-#if 0
-static uint32_t get_watch_base_addr(struct amdgpu_device *adev);
-#endif
-static int kgd_address_watch_disable(struct kgd_dev *kgd);
-static int kgd_address_watch_execute(struct kgd_dev *kgd,
- unsigned int watch_point_id,
- uint32_t cntl_val,
- uint32_t addr_hi,
- uint32_t addr_lo);
-static int kgd_wave_control_execute(struct kgd_dev *kgd,
- uint32_t gfx_index_val,
- uint32_t sq_cmd);
-static uint32_t kgd_address_watch_get_offset(struct kgd_dev *kgd,
- unsigned int watch_point_id,
- unsigned int reg_offset);
-
-static bool get_atc_vmid_pasid_mapping_valid(struct kgd_dev *kgd,
- uint8_t vmid);
-static uint16_t get_atc_vmid_pasid_mapping_pasid(struct kgd_dev *kgd,
- uint8_t vmid);
-static void set_vm_context_page_table_base(struct kgd_dev *kgd, uint32_t vmid,
- uint64_t page_table_base);
-static int invalidate_tlbs(struct kgd_dev *kgd, uint16_t pasid);
-static int invalidate_tlbs_vmid(struct kgd_dev *kgd, uint16_t vmid);
-
/* Because of REG_GET_FIELD() being used, we put this function in the
* asic specific file.
*/
@@ -140,38 +78,6 @@ static int amdgpu_amdkfd_get_tile_config(struct kgd_dev *kgd,
return 0;
}
-static const struct kfd2kgd_calls kfd2kgd = {
- .program_sh_mem_settings = kgd_program_sh_mem_settings,
- .set_pasid_vmid_mapping = kgd_set_pasid_vmid_mapping,
- .init_interrupts = kgd_init_interrupts,
- .hqd_load = kgd_hqd_load,
- .hqd_sdma_load = kgd_hqd_sdma_load,
- .hqd_dump = kgd_hqd_dump,
- .hqd_sdma_dump = kgd_hqd_sdma_dump,
- .hqd_is_occupied = kgd_hqd_is_occupied,
- .hqd_sdma_is_occupied = kgd_hqd_sdma_is_occupied,
- .hqd_destroy = kgd_hqd_destroy,
- .hqd_sdma_destroy = kgd_hqd_sdma_destroy,
- .address_watch_disable = kgd_address_watch_disable,
- .address_watch_execute = kgd_address_watch_execute,
- .wave_control_execute = kgd_wave_control_execute,
- .address_watch_get_offset = kgd_address_watch_get_offset,
- .get_atc_vmid_pasid_mapping_pasid =
- get_atc_vmid_pasid_mapping_pasid,
- .get_atc_vmid_pasid_mapping_valid =
- get_atc_vmid_pasid_mapping_valid,
- .get_tile_config = amdgpu_amdkfd_get_tile_config,
- .set_vm_context_page_table_base = set_vm_context_page_table_base,
- .invalidate_tlbs = invalidate_tlbs,
- .invalidate_tlbs_vmid = invalidate_tlbs_vmid,
- .get_hive_id = amdgpu_amdkfd_get_hive_id,
-};
-
-struct kfd2kgd_calls *amdgpu_amdkfd_gfx_10_0_get_functions()
-{
- return (struct kfd2kgd_calls *)&kfd2kgd;
-}
-
static inline struct amdgpu_device *get_amdgpu_device(struct kgd_dev *kgd)
{
return (struct amdgpu_device *)kgd;
@@ -252,11 +158,6 @@ static int kgd_set_pasid_vmid_mapping(struct kgd_dev *kgd, unsigned int pasid,
ATC_VMID0_PASID_MAPPING__VALID_MASK;
pr_debug("pasid 0x%x vmid %d, reg value %x\n", pasid, vmid, pasid_mapping);
- /*
- * need to do this twice, once for gfx and once for mmhub
- * for ATC add 16 to VMID for mmhub, for IH different registers.
- * ATC_VMID0..15 registers are separate from ATC_VMID16..31.
- */
pr_debug("ATHUB, reg %x\n",SOC15_REG_OFFSET(ATHUB, 0, mmATC_VMID0_PASID_MAPPING) + vmid);
WREG32(SOC15_REG_OFFSET(ATHUB, 0, mmATC_VMID0_PASID_MAPPING) + vmid,
@@ -308,11 +209,11 @@ static int kgd_init_interrupts(struct kgd_dev *kgd, uint32_t pipe_id)
return 0;
}
-static uint32_t get_sdma_base_addr(struct amdgpu_device *adev,
+static uint32_t get_sdma_rlc_reg_offset(struct amdgpu_device *adev,
unsigned int engine_id,
unsigned int queue_id)
{
- uint32_t base[2] = {
+ uint32_t sdma_engine_reg_base[2] = {
SOC15_REG_OFFSET(SDMA0, 0,
mmSDMA0_RLC0_RB_CNTL) - mmSDMA0_RLC0_RB_CNTL,
/* On gfx10, mmSDMA1_xxx registers are defined NOT based
@@ -324,12 +225,12 @@ static uint32_t get_sdma_base_addr(struct amdgpu_device *adev,
SOC15_REG_OFFSET(SDMA1, 0,
mmSDMA1_RLC0_RB_CNTL) - mmSDMA0_RLC0_RB_CNTL
};
- uint32_t retval;
- retval = base[engine_id] + queue_id * (mmSDMA0_RLC1_RB_CNTL -
- mmSDMA0_RLC0_RB_CNTL);
+ uint32_t retval = sdma_engine_reg_base[engine_id]
+ + queue_id * (mmSDMA0_RLC1_RB_CNTL - mmSDMA0_RLC0_RB_CNTL);
- pr_debug("sdma base address: 0x%x\n", retval);
+ pr_debug("RLC register offset for SDMA%d RLC%d: 0x%x\n", engine_id,
+ queue_id, retval);
return retval;
}
@@ -490,26 +391,22 @@ static int kgd_hqd_sdma_load(struct kgd_dev *kgd, void *mqd,
{
struct amdgpu_device *adev = get_amdgpu_device(kgd);
struct v10_sdma_mqd *m;
- uint32_t sdma_base_addr, sdmax_gfx_context_cntl;
+ uint32_t sdma_rlc_reg_offset;
unsigned long end_jiffies;
uint32_t data;
uint64_t data64;
uint64_t __user *wptr64 = (uint64_t __user *)wptr;
m = get_sdma_mqd(mqd);
- sdma_base_addr = get_sdma_base_addr(adev, m->sdma_engine_id,
+ sdma_rlc_reg_offset = get_sdma_rlc_reg_offset(adev, m->sdma_engine_id,
m->sdma_queue_id);
- pr_debug("sdma load base addr %x for engine %d, queue %d\n", sdma_base_addr, m->sdma_engine_id, m->sdma_queue_id);
- sdmax_gfx_context_cntl = m->sdma_engine_id ?
- SOC15_REG_OFFSET(SDMA1, 0, mmSDMA1_GFX_CONTEXT_CNTL) :
- SOC15_REG_OFFSET(SDMA0, 0, mmSDMA0_GFX_CONTEXT_CNTL);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_CNTL,
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_CNTL,
m->sdmax_rlcx_rb_cntl & (~SDMA0_RLC0_RB_CNTL__RB_ENABLE_MASK));
end_jiffies = msecs_to_jiffies(2000) + jiffies;
while (true) {
- data = RREG32(sdma_base_addr + mmSDMA0_RLC0_CONTEXT_STATUS);
+ data = RREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_CONTEXT_STATUS);
if (data & SDMA0_RLC0_CONTEXT_STATUS__IDLE_MASK)
break;
if (time_after(jiffies, end_jiffies)) {
@@ -518,46 +415,43 @@ static int kgd_hqd_sdma_load(struct kgd_dev *kgd, void *mqd,
}
usleep_range(500, 1000);
}
- data = RREG32(sdmax_gfx_context_cntl);
- data = REG_SET_FIELD(data, SDMA0_GFX_CONTEXT_CNTL,
- RESUME_CTX, 0);
- WREG32(sdmax_gfx_context_cntl, data);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_DOORBELL_OFFSET,
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_DOORBELL_OFFSET,
m->sdmax_rlcx_doorbell_offset);
data = REG_SET_FIELD(m->sdmax_rlcx_doorbell, SDMA0_RLC0_DOORBELL,
ENABLE, 1);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_DOORBELL, data);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_RPTR, m->sdmax_rlcx_rb_rptr);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_RPTR_HI,
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_DOORBELL, data);
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_RPTR,
+ m->sdmax_rlcx_rb_rptr);
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_RPTR_HI,
m->sdmax_rlcx_rb_rptr_hi);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_MINOR_PTR_UPDATE, 1);
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_MINOR_PTR_UPDATE, 1);
if (read_user_wptr(mm, wptr64, data64)) {
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_WPTR,
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_WPTR,
lower_32_bits(data64));
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_WPTR_HI,
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_WPTR_HI,
upper_32_bits(data64));
} else {
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_WPTR,
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_WPTR,
m->sdmax_rlcx_rb_rptr);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_WPTR_HI,
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_WPTR_HI,
m->sdmax_rlcx_rb_rptr_hi);
}
- WREG32(sdma_base_addr + mmSDMA0_RLC0_MINOR_PTR_UPDATE, 0);
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_MINOR_PTR_UPDATE, 0);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_BASE, m->sdmax_rlcx_rb_base);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_BASE_HI,
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_BASE, m->sdmax_rlcx_rb_base);
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_BASE_HI,
m->sdmax_rlcx_rb_base_hi);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_RPTR_ADDR_LO,
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_RPTR_ADDR_LO,
m->sdmax_rlcx_rb_rptr_addr_lo);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_RPTR_ADDR_HI,
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_RPTR_ADDR_HI,
m->sdmax_rlcx_rb_rptr_addr_hi);
data = REG_SET_FIELD(m->sdmax_rlcx_rb_cntl, SDMA0_RLC0_RB_CNTL,
RB_ENABLE, 1);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_CNTL, data);
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_CNTL, data);
return 0;
}
@@ -567,28 +461,26 @@ static int kgd_hqd_sdma_dump(struct kgd_dev *kgd,
uint32_t (**dump)[2], uint32_t *n_regs)
{
struct amdgpu_device *adev = get_amdgpu_device(kgd);
- uint32_t sdma_base_addr = get_sdma_base_addr(adev, engine_id, queue_id);
+ uint32_t sdma_rlc_reg_offset = get_sdma_rlc_reg_offset(adev,
+ engine_id, queue_id);
uint32_t i = 0, reg;
#undef HQD_N_REGS
#define HQD_N_REGS (19+6+7+10)
- pr_debug("sdma dump engine id %d queue_id %d\n", engine_id, queue_id);
- pr_debug("sdma base addr %x\n", sdma_base_addr);
-
*dump = kmalloc(HQD_N_REGS*2*sizeof(uint32_t), GFP_KERNEL);
if (*dump == NULL)
return -ENOMEM;
for (reg = mmSDMA0_RLC0_RB_CNTL; reg <= mmSDMA0_RLC0_DOORBELL; reg++)
- DUMP_REG(sdma_base_addr + reg);
+ DUMP_REG(sdma_rlc_reg_offset + reg);
for (reg = mmSDMA0_RLC0_STATUS; reg <= mmSDMA0_RLC0_CSA_ADDR_HI; reg++)
- DUMP_REG(sdma_base_addr + reg);
+ DUMP_REG(sdma_rlc_reg_offset + reg);
for (reg = mmSDMA0_RLC0_IB_SUB_REMAIN;
reg <= mmSDMA0_RLC0_MINOR_PTR_UPDATE; reg++)
- DUMP_REG(sdma_base_addr + reg);
+ DUMP_REG(sdma_rlc_reg_offset + reg);
for (reg = mmSDMA0_RLC0_MIDCMD_DATA0;
reg <= mmSDMA0_RLC0_MIDCMD_CNTL; reg++)
- DUMP_REG(sdma_base_addr + reg);
+ DUMP_REG(sdma_rlc_reg_offset + reg);
WARN_ON_ONCE(i != HQD_N_REGS);
*n_regs = i;
@@ -622,14 +514,14 @@ static bool kgd_hqd_sdma_is_occupied(struct kgd_dev *kgd, void *mqd)
{
struct amdgpu_device *adev = get_amdgpu_device(kgd);
struct v10_sdma_mqd *m;
- uint32_t sdma_base_addr;
+ uint32_t sdma_rlc_reg_offset;
uint32_t sdma_rlc_rb_cntl;
m = get_sdma_mqd(mqd);
- sdma_base_addr = get_sdma_base_addr(adev, m->sdma_engine_id,
+ sdma_rlc_reg_offset = get_sdma_rlc_reg_offset(adev, m->sdma_engine_id,
m->sdma_queue_id);
- sdma_rlc_rb_cntl = RREG32(sdma_base_addr + mmSDMA0_RLC0_RB_CNTL);
+ sdma_rlc_rb_cntl = RREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_CNTL);
if (sdma_rlc_rb_cntl & SDMA0_RLC0_RB_CNTL__RB_ENABLE_MASK)
return true;
@@ -750,20 +642,20 @@ static int kgd_hqd_sdma_destroy(struct kgd_dev *kgd, void *mqd,
{
struct amdgpu_device *adev = get_amdgpu_device(kgd);
struct v10_sdma_mqd *m;
- uint32_t sdma_base_addr;
+ uint32_t sdma_rlc_reg_offset;
uint32_t temp;
unsigned long end_jiffies = (utimeout * HZ / 1000) + jiffies;
m = get_sdma_mqd(mqd);
- sdma_base_addr = get_sdma_base_addr(adev, m->sdma_engine_id,
+ sdma_rlc_reg_offset = get_sdma_rlc_reg_offset(adev, m->sdma_engine_id,
m->sdma_queue_id);
- temp = RREG32(sdma_base_addr + mmSDMA0_RLC0_RB_CNTL);
+ temp = RREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_CNTL);
temp = temp & ~SDMA0_RLC0_RB_CNTL__RB_ENABLE_MASK;
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_CNTL, temp);
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_CNTL, temp);
while (true) {
- temp = RREG32(sdma_base_addr + mmSDMA0_RLC0_CONTEXT_STATUS);
+ temp = RREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_CONTEXT_STATUS);
if (temp & SDMA0_RLC0_CONTEXT_STATUS__IDLE_MASK)
break;
if (time_after(jiffies, end_jiffies)) {
@@ -773,38 +665,29 @@ static int kgd_hqd_sdma_destroy(struct kgd_dev *kgd, void *mqd,
usleep_range(500, 1000);
}
- WREG32(sdma_base_addr + mmSDMA0_RLC0_DOORBELL, 0);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_CNTL,
- RREG32(sdma_base_addr + mmSDMA0_RLC0_RB_CNTL) |
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_DOORBELL, 0);
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_CNTL,
+ RREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_CNTL) |
SDMA0_RLC0_RB_CNTL__RB_ENABLE_MASK);
- m->sdmax_rlcx_rb_rptr = RREG32(sdma_base_addr + mmSDMA0_RLC0_RB_RPTR);
+ m->sdmax_rlcx_rb_rptr = RREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_RPTR);
m->sdmax_rlcx_rb_rptr_hi =
- RREG32(sdma_base_addr + mmSDMA0_RLC0_RB_RPTR_HI);
+ RREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_RPTR_HI);
return 0;
}
-static bool get_atc_vmid_pasid_mapping_valid(struct kgd_dev *kgd,
- uint8_t vmid)
+static bool get_atc_vmid_pasid_mapping_info(struct kgd_dev *kgd,
+ uint8_t vmid, uint16_t *p_pasid)
{
- uint32_t reg;
+ uint32_t value;
struct amdgpu_device *adev = (struct amdgpu_device *) kgd;
- reg = RREG32(SOC15_REG_OFFSET(ATHUB, 0, mmATC_VMID0_PASID_MAPPING)
+ value = RREG32(SOC15_REG_OFFSET(ATHUB, 0, mmATC_VMID0_PASID_MAPPING)
+ vmid);
- return reg & ATC_VMID0_PASID_MAPPING__VALID_MASK;
-}
-
-static uint16_t get_atc_vmid_pasid_mapping_pasid(struct kgd_dev *kgd,
- uint8_t vmid)
-{
- uint32_t reg;
- struct amdgpu_device *adev = (struct amdgpu_device *) kgd;
+ *p_pasid = value & ATC_VMID0_PASID_MAPPING__PASID_MASK;
- reg = RREG32(SOC15_REG_OFFSET(ATHUB, 0, mmATC_VMID0_PASID_MAPPING)
- + vmid);
- return reg & ATC_VMID0_PASID_MAPPING__PASID_MASK;
+ return !!(value & ATC_VMID0_PASID_MAPPING__VALID_MASK);
}
static int invalidate_tlbs_with_kiq(struct amdgpu_device *adev, uint16_t pasid)
@@ -836,6 +719,8 @@ static int invalidate_tlbs(struct kgd_dev *kgd, uint16_t pasid)
{
struct amdgpu_device *adev = (struct amdgpu_device *) kgd;
int vmid;
+ uint16_t queried_pasid;
+ bool ret;
struct amdgpu_ring *ring = &adev->gfx.kiq.ring;
if (amdgpu_emu_mode == 0 && ring->sched.ready)
@@ -844,13 +729,13 @@ static int invalidate_tlbs(struct kgd_dev *kgd, uint16_t pasid)
for (vmid = 0; vmid < 16; vmid++) {
if (!amdgpu_amdkfd_is_kfd_vmid(adev, vmid))
continue;
- if (get_atc_vmid_pasid_mapping_valid(kgd, vmid)) {
- if (get_atc_vmid_pasid_mapping_pasid(kgd, vmid)
- == pasid) {
- amdgpu_gmc_flush_gpu_tlb(adev, vmid,
- AMDGPU_GFXHUB_0, 0);
- break;
- }
+
+ ret = get_atc_vmid_pasid_mapping_info(kgd, vmid,
+ &queried_pasid);
+ if (ret && queried_pasid == pasid) {
+ amdgpu_gmc_flush_gpu_tlb(adev, vmid,
+ AMDGPU_GFXHUB_0, 0);
+ break;
}
}
@@ -920,7 +805,6 @@ static void set_vm_context_page_table_base(struct kgd_dev *kgd, uint32_t vmid,
uint64_t page_table_base)
{
struct amdgpu_device *adev = get_amdgpu_device(kgd);
- uint64_t base = page_table_base | AMDGPU_PTE_VALID;
if (!amdgpu_amdkfd_is_kfd_vmid(adev, vmid)) {
pr_err("trying to set page table base for wrong VMID %u\n",
@@ -928,18 +812,31 @@ static void set_vm_context_page_table_base(struct kgd_dev *kgd, uint32_t vmid,
return;
}
- /* TODO: take advantage of per-process address space size. For
- * now, all processes share the same address space size, like
- * on GFX8 and older.
- */
- WREG32(SOC15_REG_OFFSET(GC, 0, mmGCVM_CONTEXT0_PAGE_TABLE_START_ADDR_LO32) + (vmid*2), 0);
- WREG32(SOC15_REG_OFFSET(GC, 0, mmGCVM_CONTEXT0_PAGE_TABLE_START_ADDR_HI32) + (vmid*2), 0);
-
- WREG32(SOC15_REG_OFFSET(GC, 0, mmGCVM_CONTEXT0_PAGE_TABLE_END_ADDR_LO32) + (vmid*2),
- lower_32_bits(adev->vm_manager.max_pfn - 1));
- WREG32(SOC15_REG_OFFSET(GC, 0, mmGCVM_CONTEXT0_PAGE_TABLE_END_ADDR_HI32) + (vmid*2),
- upper_32_bits(adev->vm_manager.max_pfn - 1));
-
- WREG32(SOC15_REG_OFFSET(GC, 0, mmGCVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32) + (vmid*2), lower_32_bits(base));
- WREG32(SOC15_REG_OFFSET(GC, 0, mmGCVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_HI32) + (vmid*2), upper_32_bits(base));
+ /* SDMA is on gfxhub as well for Navi1* series */
+ gfxhub_v2_0_setup_vm_pt_regs(adev, vmid, page_table_base);
}
+
+const struct kfd2kgd_calls gfx_v10_kfd2kgd = {
+ .program_sh_mem_settings = kgd_program_sh_mem_settings,
+ .set_pasid_vmid_mapping = kgd_set_pasid_vmid_mapping,
+ .init_interrupts = kgd_init_interrupts,
+ .hqd_load = kgd_hqd_load,
+ .hqd_sdma_load = kgd_hqd_sdma_load,
+ .hqd_dump = kgd_hqd_dump,
+ .hqd_sdma_dump = kgd_hqd_sdma_dump,
+ .hqd_is_occupied = kgd_hqd_is_occupied,
+ .hqd_sdma_is_occupied = kgd_hqd_sdma_is_occupied,
+ .hqd_destroy = kgd_hqd_destroy,
+ .hqd_sdma_destroy = kgd_hqd_sdma_destroy,
+ .address_watch_disable = kgd_address_watch_disable,
+ .address_watch_execute = kgd_address_watch_execute,
+ .wave_control_execute = kgd_wave_control_execute,
+ .address_watch_get_offset = kgd_address_watch_get_offset,
+ .get_atc_vmid_pasid_mapping_info =
+ get_atc_vmid_pasid_mapping_info,
+ .get_tile_config = amdgpu_amdkfd_get_tile_config,
+ .set_vm_context_page_table_base = set_vm_context_page_table_base,
+ .invalidate_tlbs = invalidate_tlbs,
+ .invalidate_tlbs_vmid = invalidate_tlbs_vmid,
+ .get_hive_id = amdgpu_amdkfd_get_hive_id,
+};
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c
index 7eedb5c9fd1d..495b15ed28cd 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c
@@ -23,8 +23,6 @@
#undef pr_fmt
#define pr_fmt(fmt) "kfd2kgd: " fmt
-#include <linux/fdtable.h>
-#include <linux/uaccess.h>
#include <linux/mmu_context.h>
#include <drm/drmP.h>
#include "amdgpu.h"
@@ -89,65 +87,6 @@ union TCP_WATCH_CNTL_BITS {
float f32All;
};
-/*
- * Register access functions
- */
-
-static void kgd_program_sh_mem_settings(struct kgd_dev *kgd, uint32_t vmid,
- uint32_t sh_mem_config, uint32_t sh_mem_ape1_base,
- uint32_t sh_mem_ape1_limit, uint32_t sh_mem_bases);
-
-static int kgd_set_pasid_vmid_mapping(struct kgd_dev *kgd, unsigned int pasid,
- unsigned int vmid);
-
-static int kgd_init_interrupts(struct kgd_dev *kgd, uint32_t pipe_id);
-static int kgd_hqd_load(struct kgd_dev *kgd, void *mqd, uint32_t pipe_id,
- uint32_t queue_id, uint32_t __user *wptr,
- uint32_t wptr_shift, uint32_t wptr_mask,
- struct mm_struct *mm);
-static int kgd_hqd_dump(struct kgd_dev *kgd,
- uint32_t pipe_id, uint32_t queue_id,
- uint32_t (**dump)[2], uint32_t *n_regs);
-static int kgd_hqd_sdma_load(struct kgd_dev *kgd, void *mqd,
- uint32_t __user *wptr, struct mm_struct *mm);
-static int kgd_hqd_sdma_dump(struct kgd_dev *kgd,
- uint32_t engine_id, uint32_t queue_id,
- uint32_t (**dump)[2], uint32_t *n_regs);
-static bool kgd_hqd_is_occupied(struct kgd_dev *kgd, uint64_t queue_address,
- uint32_t pipe_id, uint32_t queue_id);
-
-static int kgd_hqd_destroy(struct kgd_dev *kgd, void *mqd,
- enum kfd_preempt_type reset_type,
- unsigned int utimeout, uint32_t pipe_id,
- uint32_t queue_id);
-static bool kgd_hqd_sdma_is_occupied(struct kgd_dev *kgd, void *mqd);
-static int kgd_hqd_sdma_destroy(struct kgd_dev *kgd, void *mqd,
- unsigned int utimeout);
-static int kgd_address_watch_disable(struct kgd_dev *kgd);
-static int kgd_address_watch_execute(struct kgd_dev *kgd,
- unsigned int watch_point_id,
- uint32_t cntl_val,
- uint32_t addr_hi,
- uint32_t addr_lo);
-static int kgd_wave_control_execute(struct kgd_dev *kgd,
- uint32_t gfx_index_val,
- uint32_t sq_cmd);
-static uint32_t kgd_address_watch_get_offset(struct kgd_dev *kgd,
- unsigned int watch_point_id,
- unsigned int reg_offset);
-
-static bool get_atc_vmid_pasid_mapping_valid(struct kgd_dev *kgd, uint8_t vmid);
-static uint16_t get_atc_vmid_pasid_mapping_pasid(struct kgd_dev *kgd,
- uint8_t vmid);
-
-static void set_scratch_backing_va(struct kgd_dev *kgd,
- uint64_t va, uint32_t vmid);
-static void set_vm_context_page_table_base(struct kgd_dev *kgd, uint32_t vmid,
- uint64_t page_table_base);
-static int invalidate_tlbs(struct kgd_dev *kgd, uint16_t pasid);
-static int invalidate_tlbs_vmid(struct kgd_dev *kgd, uint16_t vmid);
-static uint32_t read_vmid_from_vmfault_reg(struct kgd_dev *kgd);
-
/* Because of REG_GET_FIELD() being used, we put this function in the
* asic specific file.
*/
@@ -173,37 +112,6 @@ static int get_tile_config(struct kgd_dev *kgd,
return 0;
}
-static const struct kfd2kgd_calls kfd2kgd = {
- .program_sh_mem_settings = kgd_program_sh_mem_settings,
- .set_pasid_vmid_mapping = kgd_set_pasid_vmid_mapping,
- .init_interrupts = kgd_init_interrupts,
- .hqd_load = kgd_hqd_load,
- .hqd_sdma_load = kgd_hqd_sdma_load,
- .hqd_dump = kgd_hqd_dump,
- .hqd_sdma_dump = kgd_hqd_sdma_dump,
- .hqd_is_occupied = kgd_hqd_is_occupied,
- .hqd_sdma_is_occupied = kgd_hqd_sdma_is_occupied,
- .hqd_destroy = kgd_hqd_destroy,
- .hqd_sdma_destroy = kgd_hqd_sdma_destroy,
- .address_watch_disable = kgd_address_watch_disable,
- .address_watch_execute = kgd_address_watch_execute,
- .wave_control_execute = kgd_wave_control_execute,
- .address_watch_get_offset = kgd_address_watch_get_offset,
- .get_atc_vmid_pasid_mapping_pasid = get_atc_vmid_pasid_mapping_pasid,
- .get_atc_vmid_pasid_mapping_valid = get_atc_vmid_pasid_mapping_valid,
- .set_scratch_backing_va = set_scratch_backing_va,
- .get_tile_config = get_tile_config,
- .set_vm_context_page_table_base = set_vm_context_page_table_base,
- .invalidate_tlbs = invalidate_tlbs,
- .invalidate_tlbs_vmid = invalidate_tlbs_vmid,
- .read_vmid_from_vmfault_reg = read_vmid_from_vmfault_reg,
-};
-
-struct kfd2kgd_calls *amdgpu_amdkfd_gfx_7_get_functions(void)
-{
- return (struct kfd2kgd_calls *)&kfd2kgd;
-}
-
static inline struct amdgpu_device *get_amdgpu_device(struct kgd_dev *kgd)
{
return (struct amdgpu_device *)kgd;
@@ -306,13 +214,15 @@ static int kgd_init_interrupts(struct kgd_dev *kgd, uint32_t pipe_id)
return 0;
}
-static inline uint32_t get_sdma_base_addr(struct cik_sdma_rlc_registers *m)
+static inline uint32_t get_sdma_rlc_reg_offset(struct cik_sdma_rlc_registers *m)
{
uint32_t retval;
retval = m->sdma_engine_id * SDMA1_REGISTER_OFFSET +
m->sdma_queue_id * KFD_CIK_SDMA_QUEUE_OFFSET;
- pr_debug("sdma base address: 0x%x\n", retval);
+
+ pr_debug("RLC register offset for SDMA%d RLC%d: 0x%x\n",
+ m->sdma_engine_id, m->sdma_queue_id, retval);
return retval;
}
@@ -415,18 +325,18 @@ static int kgd_hqd_sdma_load(struct kgd_dev *kgd, void *mqd,
struct amdgpu_device *adev = get_amdgpu_device(kgd);
struct cik_sdma_rlc_registers *m;
unsigned long end_jiffies;
- uint32_t sdma_base_addr;
+ uint32_t sdma_rlc_reg_offset;
uint32_t data;
m = get_sdma_mqd(mqd);
- sdma_base_addr = get_sdma_base_addr(m);
+ sdma_rlc_reg_offset = get_sdma_rlc_reg_offset(m);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_CNTL,
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_CNTL,
m->sdma_rlc_rb_cntl & (~SDMA0_RLC0_RB_CNTL__RB_ENABLE_MASK));
end_jiffies = msecs_to_jiffies(2000) + jiffies;
while (true) {
- data = RREG32(sdma_base_addr + mmSDMA0_RLC0_CONTEXT_STATUS);
+ data = RREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_CONTEXT_STATUS);
if (data & SDMA0_RLC0_CONTEXT_STATUS__IDLE_MASK)
break;
if (time_after(jiffies, end_jiffies)) {
@@ -435,42 +345,32 @@ static int kgd_hqd_sdma_load(struct kgd_dev *kgd, void *mqd,
}
usleep_range(500, 1000);
}
- if (m->sdma_engine_id) {
- data = RREG32(mmSDMA1_GFX_CONTEXT_CNTL);
- data = REG_SET_FIELD(data, SDMA1_GFX_CONTEXT_CNTL,
- RESUME_CTX, 0);
- WREG32(mmSDMA1_GFX_CONTEXT_CNTL, data);
- } else {
- data = RREG32(mmSDMA0_GFX_CONTEXT_CNTL);
- data = REG_SET_FIELD(data, SDMA0_GFX_CONTEXT_CNTL,
- RESUME_CTX, 0);
- WREG32(mmSDMA0_GFX_CONTEXT_CNTL, data);
- }
data = REG_SET_FIELD(m->sdma_rlc_doorbell, SDMA0_RLC0_DOORBELL,
ENABLE, 1);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_DOORBELL, data);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_RPTR, m->sdma_rlc_rb_rptr);
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_DOORBELL, data);
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_RPTR,
+ m->sdma_rlc_rb_rptr);
if (read_user_wptr(mm, wptr, data))
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_WPTR, data);
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_WPTR, data);
else
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_WPTR,
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_WPTR,
m->sdma_rlc_rb_rptr);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_VIRTUAL_ADDR,
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_VIRTUAL_ADDR,
m->sdma_rlc_virtual_addr);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_BASE, m->sdma_rlc_rb_base);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_BASE_HI,
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_BASE, m->sdma_rlc_rb_base);
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_BASE_HI,
m->sdma_rlc_rb_base_hi);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_RPTR_ADDR_LO,
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_RPTR_ADDR_LO,
m->sdma_rlc_rb_rptr_addr_lo);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_RPTR_ADDR_HI,
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_RPTR_ADDR_HI,
m->sdma_rlc_rb_rptr_addr_hi);
data = REG_SET_FIELD(m->sdma_rlc_rb_cntl, SDMA0_RLC0_RB_CNTL,
RB_ENABLE, 1);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_CNTL, data);
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_CNTL, data);
return 0;
}
@@ -528,13 +428,13 @@ static bool kgd_hqd_sdma_is_occupied(struct kgd_dev *kgd, void *mqd)
{
struct amdgpu_device *adev = get_amdgpu_device(kgd);
struct cik_sdma_rlc_registers *m;
- uint32_t sdma_base_addr;
+ uint32_t sdma_rlc_reg_offset;
uint32_t sdma_rlc_rb_cntl;
m = get_sdma_mqd(mqd);
- sdma_base_addr = get_sdma_base_addr(m);
+ sdma_rlc_reg_offset = get_sdma_rlc_reg_offset(m);
- sdma_rlc_rb_cntl = RREG32(sdma_base_addr + mmSDMA0_RLC0_RB_CNTL);
+ sdma_rlc_rb_cntl = RREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_CNTL);
if (sdma_rlc_rb_cntl & SDMA0_RLC0_RB_CNTL__RB_ENABLE_MASK)
return true;
@@ -649,19 +549,19 @@ static int kgd_hqd_sdma_destroy(struct kgd_dev *kgd, void *mqd,
{
struct amdgpu_device *adev = get_amdgpu_device(kgd);
struct cik_sdma_rlc_registers *m;
- uint32_t sdma_base_addr;
+ uint32_t sdma_rlc_reg_offset;
uint32_t temp;
unsigned long end_jiffies = (utimeout * HZ / 1000) + jiffies;
m = get_sdma_mqd(mqd);
- sdma_base_addr = get_sdma_base_addr(m);
+ sdma_rlc_reg_offset = get_sdma_rlc_reg_offset(m);
- temp = RREG32(sdma_base_addr + mmSDMA0_RLC0_RB_CNTL);
+ temp = RREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_CNTL);
temp = temp & ~SDMA0_RLC0_RB_CNTL__RB_ENABLE_MASK;
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_CNTL, temp);
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_CNTL, temp);
while (true) {
- temp = RREG32(sdma_base_addr + mmSDMA0_RLC0_CONTEXT_STATUS);
+ temp = RREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_CONTEXT_STATUS);
if (temp & SDMA0_RLC0_CONTEXT_STATUS__IDLE_MASK)
break;
if (time_after(jiffies, end_jiffies)) {
@@ -671,12 +571,12 @@ static int kgd_hqd_sdma_destroy(struct kgd_dev *kgd, void *mqd,
usleep_range(500, 1000);
}
- WREG32(sdma_base_addr + mmSDMA0_RLC0_DOORBELL, 0);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_CNTL,
- RREG32(sdma_base_addr + mmSDMA0_RLC0_RB_CNTL) |
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_DOORBELL, 0);
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_CNTL,
+ RREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_CNTL) |
SDMA0_RLC0_RB_CNTL__RB_ENABLE_MASK);
- m->sdma_rlc_rb_rptr = RREG32(sdma_base_addr + mmSDMA0_RLC0_RB_RPTR);
+ m->sdma_rlc_rb_rptr = RREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_RPTR);
return 0;
}
@@ -764,24 +664,16 @@ static uint32_t kgd_address_watch_get_offset(struct kgd_dev *kgd,
return watchRegs[watch_point_id * ADDRESS_WATCH_REG_MAX + reg_offset];
}
-static bool get_atc_vmid_pasid_mapping_valid(struct kgd_dev *kgd,
- uint8_t vmid)
+static bool get_atc_vmid_pasid_mapping_info(struct kgd_dev *kgd,
+ uint8_t vmid, uint16_t *p_pasid)
{
- uint32_t reg;
+ uint32_t value;
struct amdgpu_device *adev = (struct amdgpu_device *) kgd;
- reg = RREG32(mmATC_VMID0_PASID_MAPPING + vmid);
- return reg & ATC_VMID0_PASID_MAPPING__VALID_MASK;
-}
+ value = RREG32(mmATC_VMID0_PASID_MAPPING + vmid);
+ *p_pasid = value & ATC_VMID0_PASID_MAPPING__PASID_MASK;
-static uint16_t get_atc_vmid_pasid_mapping_pasid(struct kgd_dev *kgd,
- uint8_t vmid)
-{
- uint32_t reg;
- struct amdgpu_device *adev = (struct amdgpu_device *) kgd;
-
- reg = RREG32(mmATC_VMID0_PASID_MAPPING + vmid);
- return reg & ATC_VMID0_PASID_MAPPING__PASID_MASK;
+ return !!(value & ATC_VMID0_PASID_MAPPING__VALID_MASK);
}
static void set_scratch_backing_va(struct kgd_dev *kgd,
@@ -861,3 +753,28 @@ static uint32_t read_vmid_from_vmfault_reg(struct kgd_dev *kgd)
return REG_GET_FIELD(status, VM_CONTEXT1_PROTECTION_FAULT_STATUS, VMID);
}
+
+const struct kfd2kgd_calls gfx_v7_kfd2kgd = {
+ .program_sh_mem_settings = kgd_program_sh_mem_settings,
+ .set_pasid_vmid_mapping = kgd_set_pasid_vmid_mapping,
+ .init_interrupts = kgd_init_interrupts,
+ .hqd_load = kgd_hqd_load,
+ .hqd_sdma_load = kgd_hqd_sdma_load,
+ .hqd_dump = kgd_hqd_dump,
+ .hqd_sdma_dump = kgd_hqd_sdma_dump,
+ .hqd_is_occupied = kgd_hqd_is_occupied,
+ .hqd_sdma_is_occupied = kgd_hqd_sdma_is_occupied,
+ .hqd_destroy = kgd_hqd_destroy,
+ .hqd_sdma_destroy = kgd_hqd_sdma_destroy,
+ .address_watch_disable = kgd_address_watch_disable,
+ .address_watch_execute = kgd_address_watch_execute,
+ .wave_control_execute = kgd_wave_control_execute,
+ .address_watch_get_offset = kgd_address_watch_get_offset,
+ .get_atc_vmid_pasid_mapping_info = get_atc_vmid_pasid_mapping_info,
+ .set_scratch_backing_va = set_scratch_backing_va,
+ .get_tile_config = get_tile_config,
+ .set_vm_context_page_table_base = set_vm_context_page_table_base,
+ .invalidate_tlbs = invalidate_tlbs,
+ .invalidate_tlbs_vmid = invalidate_tlbs_vmid,
+ .read_vmid_from_vmfault_reg = read_vmid_from_vmfault_reg,
+};
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c
index fafb42175656..0118e6f18355 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c
@@ -23,9 +23,6 @@
#undef pr_fmt
#define pr_fmt(fmt) "kfd2kgd: " fmt
-#include <linux/module.h>
-#include <linux/fdtable.h>
-#include <linux/uaccess.h>
#include <linux/mmu_context.h>
#include <drm/drmP.h>
#include "amdgpu.h"
@@ -54,62 +51,6 @@ static const uint32_t watchRegs[MAX_WATCH_ADDRESSES * ADDRESS_WATCH_REG_MAX] = {
mmTCP_WATCH3_ADDR_H, mmTCP_WATCH3_ADDR_L, mmTCP_WATCH3_CNTL
};
-/*
- * Register access functions
- */
-
-static void kgd_program_sh_mem_settings(struct kgd_dev *kgd, uint32_t vmid,
- uint32_t sh_mem_config,
- uint32_t sh_mem_ape1_base, uint32_t sh_mem_ape1_limit,
- uint32_t sh_mem_bases);
-static int kgd_set_pasid_vmid_mapping(struct kgd_dev *kgd, unsigned int pasid,
- unsigned int vmid);
-static int kgd_init_interrupts(struct kgd_dev *kgd, uint32_t pipe_id);
-static int kgd_hqd_load(struct kgd_dev *kgd, void *mqd, uint32_t pipe_id,
- uint32_t queue_id, uint32_t __user *wptr,
- uint32_t wptr_shift, uint32_t wptr_mask,
- struct mm_struct *mm);
-static int kgd_hqd_dump(struct kgd_dev *kgd,
- uint32_t pipe_id, uint32_t queue_id,
- uint32_t (**dump)[2], uint32_t *n_regs);
-static int kgd_hqd_sdma_load(struct kgd_dev *kgd, void *mqd,
- uint32_t __user *wptr, struct mm_struct *mm);
-static int kgd_hqd_sdma_dump(struct kgd_dev *kgd,
- uint32_t engine_id, uint32_t queue_id,
- uint32_t (**dump)[2], uint32_t *n_regs);
-static bool kgd_hqd_is_occupied(struct kgd_dev *kgd, uint64_t queue_address,
- uint32_t pipe_id, uint32_t queue_id);
-static bool kgd_hqd_sdma_is_occupied(struct kgd_dev *kgd, void *mqd);
-static int kgd_hqd_destroy(struct kgd_dev *kgd, void *mqd,
- enum kfd_preempt_type reset_type,
- unsigned int utimeout, uint32_t pipe_id,
- uint32_t queue_id);
-static int kgd_hqd_sdma_destroy(struct kgd_dev *kgd, void *mqd,
- unsigned int utimeout);
-static int kgd_address_watch_disable(struct kgd_dev *kgd);
-static int kgd_address_watch_execute(struct kgd_dev *kgd,
- unsigned int watch_point_id,
- uint32_t cntl_val,
- uint32_t addr_hi,
- uint32_t addr_lo);
-static int kgd_wave_control_execute(struct kgd_dev *kgd,
- uint32_t gfx_index_val,
- uint32_t sq_cmd);
-static uint32_t kgd_address_watch_get_offset(struct kgd_dev *kgd,
- unsigned int watch_point_id,
- unsigned int reg_offset);
-
-static bool get_atc_vmid_pasid_mapping_valid(struct kgd_dev *kgd,
- uint8_t vmid);
-static uint16_t get_atc_vmid_pasid_mapping_pasid(struct kgd_dev *kgd,
- uint8_t vmid);
-static void set_scratch_backing_va(struct kgd_dev *kgd,
- uint64_t va, uint32_t vmid);
-static void set_vm_context_page_table_base(struct kgd_dev *kgd, uint32_t vmid,
- uint64_t page_table_base);
-static int invalidate_tlbs(struct kgd_dev *kgd, uint16_t pasid);
-static int invalidate_tlbs_vmid(struct kgd_dev *kgd, uint16_t vmid);
-
/* Because of REG_GET_FIELD() being used, we put this function in the
* asic specific file.
*/
@@ -135,38 +76,6 @@ static int get_tile_config(struct kgd_dev *kgd,
return 0;
}
-static const struct kfd2kgd_calls kfd2kgd = {
- .program_sh_mem_settings = kgd_program_sh_mem_settings,
- .set_pasid_vmid_mapping = kgd_set_pasid_vmid_mapping,
- .init_interrupts = kgd_init_interrupts,
- .hqd_load = kgd_hqd_load,
- .hqd_sdma_load = kgd_hqd_sdma_load,
- .hqd_dump = kgd_hqd_dump,
- .hqd_sdma_dump = kgd_hqd_sdma_dump,
- .hqd_is_occupied = kgd_hqd_is_occupied,
- .hqd_sdma_is_occupied = kgd_hqd_sdma_is_occupied,
- .hqd_destroy = kgd_hqd_destroy,
- .hqd_sdma_destroy = kgd_hqd_sdma_destroy,
- .address_watch_disable = kgd_address_watch_disable,
- .address_watch_execute = kgd_address_watch_execute,
- .wave_control_execute = kgd_wave_control_execute,
- .address_watch_get_offset = kgd_address_watch_get_offset,
- .get_atc_vmid_pasid_mapping_pasid =
- get_atc_vmid_pasid_mapping_pasid,
- .get_atc_vmid_pasid_mapping_valid =
- get_atc_vmid_pasid_mapping_valid,
- .set_scratch_backing_va = set_scratch_backing_va,
- .get_tile_config = get_tile_config,
- .set_vm_context_page_table_base = set_vm_context_page_table_base,
- .invalidate_tlbs = invalidate_tlbs,
- .invalidate_tlbs_vmid = invalidate_tlbs_vmid,
-};
-
-struct kfd2kgd_calls *amdgpu_amdkfd_gfx_8_0_get_functions(void)
-{
- return (struct kfd2kgd_calls *)&kfd2kgd;
-}
-
static inline struct amdgpu_device *get_amdgpu_device(struct kgd_dev *kgd)
{
return (struct amdgpu_device *)kgd;
@@ -270,13 +179,15 @@ static int kgd_init_interrupts(struct kgd_dev *kgd, uint32_t pipe_id)
return 0;
}
-static inline uint32_t get_sdma_base_addr(struct vi_sdma_mqd *m)
+static inline uint32_t get_sdma_rlc_reg_offset(struct vi_sdma_mqd *m)
{
uint32_t retval;
retval = m->sdma_engine_id * SDMA1_REGISTER_OFFSET +
m->sdma_queue_id * KFD_VI_SDMA_QUEUE_OFFSET;
- pr_debug("sdma base address: 0x%x\n", retval);
+
+ pr_debug("RLC register offset for SDMA%d RLC%d: 0x%x\n",
+ m->sdma_engine_id, m->sdma_queue_id, retval);
return retval;
}
@@ -408,17 +319,17 @@ static int kgd_hqd_sdma_load(struct kgd_dev *kgd, void *mqd,
struct amdgpu_device *adev = get_amdgpu_device(kgd);
struct vi_sdma_mqd *m;
unsigned long end_jiffies;
- uint32_t sdma_base_addr;
+ uint32_t sdma_rlc_reg_offset;
uint32_t data;
m = get_sdma_mqd(mqd);
- sdma_base_addr = get_sdma_base_addr(m);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_CNTL,
+ sdma_rlc_reg_offset = get_sdma_rlc_reg_offset(m);
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_CNTL,
m->sdmax_rlcx_rb_cntl & (~SDMA0_RLC0_RB_CNTL__RB_ENABLE_MASK));
end_jiffies = msecs_to_jiffies(2000) + jiffies;
while (true) {
- data = RREG32(sdma_base_addr + mmSDMA0_RLC0_CONTEXT_STATUS);
+ data = RREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_CONTEXT_STATUS);
if (data & SDMA0_RLC0_CONTEXT_STATUS__IDLE_MASK)
break;
if (time_after(jiffies, end_jiffies)) {
@@ -427,42 +338,32 @@ static int kgd_hqd_sdma_load(struct kgd_dev *kgd, void *mqd,
}
usleep_range(500, 1000);
}
- if (m->sdma_engine_id) {
- data = RREG32(mmSDMA1_GFX_CONTEXT_CNTL);
- data = REG_SET_FIELD(data, SDMA1_GFX_CONTEXT_CNTL,
- RESUME_CTX, 0);
- WREG32(mmSDMA1_GFX_CONTEXT_CNTL, data);
- } else {
- data = RREG32(mmSDMA0_GFX_CONTEXT_CNTL);
- data = REG_SET_FIELD(data, SDMA0_GFX_CONTEXT_CNTL,
- RESUME_CTX, 0);
- WREG32(mmSDMA0_GFX_CONTEXT_CNTL, data);
- }
data = REG_SET_FIELD(m->sdmax_rlcx_doorbell, SDMA0_RLC0_DOORBELL,
ENABLE, 1);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_DOORBELL, data);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_RPTR, m->sdmax_rlcx_rb_rptr);
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_DOORBELL, data);
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_RPTR,
+ m->sdmax_rlcx_rb_rptr);
if (read_user_wptr(mm, wptr, data))
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_WPTR, data);
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_WPTR, data);
else
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_WPTR,
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_WPTR,
m->sdmax_rlcx_rb_rptr);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_VIRTUAL_ADDR,
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_VIRTUAL_ADDR,
m->sdmax_rlcx_virtual_addr);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_BASE, m->sdmax_rlcx_rb_base);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_BASE_HI,
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_BASE, m->sdmax_rlcx_rb_base);
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_BASE_HI,
m->sdmax_rlcx_rb_base_hi);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_RPTR_ADDR_LO,
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_RPTR_ADDR_LO,
m->sdmax_rlcx_rb_rptr_addr_lo);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_RPTR_ADDR_HI,
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_RPTR_ADDR_HI,
m->sdmax_rlcx_rb_rptr_addr_hi);
data = REG_SET_FIELD(m->sdmax_rlcx_rb_cntl, SDMA0_RLC0_RB_CNTL,
RB_ENABLE, 1);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_CNTL, data);
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_CNTL, data);
return 0;
}
@@ -529,13 +430,13 @@ static bool kgd_hqd_sdma_is_occupied(struct kgd_dev *kgd, void *mqd)
{
struct amdgpu_device *adev = get_amdgpu_device(kgd);
struct vi_sdma_mqd *m;
- uint32_t sdma_base_addr;
+ uint32_t sdma_rlc_reg_offset;
uint32_t sdma_rlc_rb_cntl;
m = get_sdma_mqd(mqd);
- sdma_base_addr = get_sdma_base_addr(m);
+ sdma_rlc_reg_offset = get_sdma_rlc_reg_offset(m);
- sdma_rlc_rb_cntl = RREG32(sdma_base_addr + mmSDMA0_RLC0_RB_CNTL);
+ sdma_rlc_rb_cntl = RREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_CNTL);
if (sdma_rlc_rb_cntl & SDMA0_RLC0_RB_CNTL__RB_ENABLE_MASK)
return true;
@@ -653,19 +554,19 @@ static int kgd_hqd_sdma_destroy(struct kgd_dev *kgd, void *mqd,
{
struct amdgpu_device *adev = get_amdgpu_device(kgd);
struct vi_sdma_mqd *m;
- uint32_t sdma_base_addr;
+ uint32_t sdma_rlc_reg_offset;
uint32_t temp;
unsigned long end_jiffies = (utimeout * HZ / 1000) + jiffies;
m = get_sdma_mqd(mqd);
- sdma_base_addr = get_sdma_base_addr(m);
+ sdma_rlc_reg_offset = get_sdma_rlc_reg_offset(m);
- temp = RREG32(sdma_base_addr + mmSDMA0_RLC0_RB_CNTL);
+ temp = RREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_CNTL);
temp = temp & ~SDMA0_RLC0_RB_CNTL__RB_ENABLE_MASK;
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_CNTL, temp);
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_CNTL, temp);
while (true) {
- temp = RREG32(sdma_base_addr + mmSDMA0_RLC0_CONTEXT_STATUS);
+ temp = RREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_CONTEXT_STATUS);
if (temp & SDMA0_RLC0_CONTEXT_STATUS__IDLE_MASK)
break;
if (time_after(jiffies, end_jiffies)) {
@@ -675,34 +576,26 @@ static int kgd_hqd_sdma_destroy(struct kgd_dev *kgd, void *mqd,
usleep_range(500, 1000);
}
- WREG32(sdma_base_addr + mmSDMA0_RLC0_DOORBELL, 0);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_CNTL,
- RREG32(sdma_base_addr + mmSDMA0_RLC0_RB_CNTL) |
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_DOORBELL, 0);
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_CNTL,
+ RREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_CNTL) |
SDMA0_RLC0_RB_CNTL__RB_ENABLE_MASK);
- m->sdmax_rlcx_rb_rptr = RREG32(sdma_base_addr + mmSDMA0_RLC0_RB_RPTR);
+ m->sdmax_rlcx_rb_rptr = RREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_RPTR);
return 0;
}
-static bool get_atc_vmid_pasid_mapping_valid(struct kgd_dev *kgd,
- uint8_t vmid)
+static bool get_atc_vmid_pasid_mapping_info(struct kgd_dev *kgd,
+ uint8_t vmid, uint16_t *p_pasid)
{
- uint32_t reg;
+ uint32_t value;
struct amdgpu_device *adev = (struct amdgpu_device *) kgd;
- reg = RREG32(mmATC_VMID0_PASID_MAPPING + vmid);
- return reg & ATC_VMID0_PASID_MAPPING__VALID_MASK;
-}
-
-static uint16_t get_atc_vmid_pasid_mapping_pasid(struct kgd_dev *kgd,
- uint8_t vmid)
-{
- uint32_t reg;
- struct amdgpu_device *adev = (struct amdgpu_device *) kgd;
+ value = RREG32(mmATC_VMID0_PASID_MAPPING + vmid);
+ *p_pasid = value & ATC_VMID0_PASID_MAPPING__PASID_MASK;
- reg = RREG32(mmATC_VMID0_PASID_MAPPING + vmid);
- return reg & ATC_VMID0_PASID_MAPPING__PASID_MASK;
+ return !!(value & ATC_VMID0_PASID_MAPPING__VALID_MASK);
}
static int kgd_address_watch_disable(struct kgd_dev *kgd)
@@ -893,3 +786,28 @@ static int invalidate_tlbs_vmid(struct kgd_dev *kgd, uint16_t vmid)
RREG32(mmVM_INVALIDATE_RESPONSE);
return 0;
}
+
+const struct kfd2kgd_calls gfx_v8_kfd2kgd = {
+ .program_sh_mem_settings = kgd_program_sh_mem_settings,
+ .set_pasid_vmid_mapping = kgd_set_pasid_vmid_mapping,
+ .init_interrupts = kgd_init_interrupts,
+ .hqd_load = kgd_hqd_load,
+ .hqd_sdma_load = kgd_hqd_sdma_load,
+ .hqd_dump = kgd_hqd_dump,
+ .hqd_sdma_dump = kgd_hqd_sdma_dump,
+ .hqd_is_occupied = kgd_hqd_is_occupied,
+ .hqd_sdma_is_occupied = kgd_hqd_sdma_is_occupied,
+ .hqd_destroy = kgd_hqd_destroy,
+ .hqd_sdma_destroy = kgd_hqd_sdma_destroy,
+ .address_watch_disable = kgd_address_watch_disable,
+ .address_watch_execute = kgd_address_watch_execute,
+ .wave_control_execute = kgd_wave_control_execute,
+ .address_watch_get_offset = kgd_address_watch_get_offset,
+ .get_atc_vmid_pasid_mapping_info =
+ get_atc_vmid_pasid_mapping_info,
+ .set_scratch_backing_va = set_scratch_backing_va,
+ .get_tile_config = get_tile_config,
+ .set_vm_context_page_table_base = set_vm_context_page_table_base,
+ .invalidate_tlbs = invalidate_tlbs,
+ .invalidate_tlbs_vmid = invalidate_tlbs_vmid,
+};
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 f83b5741ed1b..4c524398f556 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
@@ -23,15 +23,11 @@
#undef pr_fmt
#define pr_fmt(fmt) "kfd2kgd: " fmt
-#include <linux/module.h>
-#include <linux/fdtable.h>
-#include <linux/uaccess.h>
#include <linux/mmu_context.h>
#include <drm/drmP.h>
#include "amdgpu.h"
#include "amdgpu_amdkfd.h"
#include "amdgpu_amdkfd_gfx_v8.h"
-#include "soc15_hw_ip.h"
#include "gc/gc_9_0_offset.h"
#include "gc/gc_9_0_sh_mask.h"
#include "vega10_enum.h"
@@ -52,9 +48,6 @@
#include "gmc_v9_0.h"
-#define V9_PIPE_PER_MEC (4)
-#define V9_QUEUES_PER_PIPE_MEC (8)
-
enum hqd_dequeue_request_type {
NO_ACTION = 0,
DRAIN_PIPE,
@@ -234,22 +227,21 @@ int kgd_gfx_v9_init_interrupts(struct kgd_dev *kgd, uint32_t pipe_id)
return 0;
}
-static uint32_t get_sdma_base_addr(struct amdgpu_device *adev,
+static uint32_t get_sdma_rlc_reg_offset(struct amdgpu_device *adev,
unsigned int engine_id,
unsigned int queue_id)
{
- uint32_t base[2] = {
+ uint32_t sdma_engine_reg_base[2] = {
SOC15_REG_OFFSET(SDMA0, 0,
mmSDMA0_RLC0_RB_CNTL) - mmSDMA0_RLC0_RB_CNTL,
SOC15_REG_OFFSET(SDMA1, 0,
mmSDMA1_RLC0_RB_CNTL) - mmSDMA1_RLC0_RB_CNTL
};
- uint32_t retval;
-
- retval = base[engine_id] + queue_id * (mmSDMA0_RLC1_RB_CNTL -
- mmSDMA0_RLC0_RB_CNTL);
+ uint32_t retval = sdma_engine_reg_base[engine_id]
+ + queue_id * (mmSDMA0_RLC1_RB_CNTL - mmSDMA0_RLC0_RB_CNTL);
- pr_debug("sdma base address: 0x%x\n", retval);
+ pr_debug("RLC register offset for SDMA%d RLC%d: 0x%x\n", engine_id,
+ queue_id, retval);
return retval;
}
@@ -406,25 +398,22 @@ static int kgd_hqd_sdma_load(struct kgd_dev *kgd, void *mqd,
{
struct amdgpu_device *adev = get_amdgpu_device(kgd);
struct v9_sdma_mqd *m;
- uint32_t sdma_base_addr, sdmax_gfx_context_cntl;
+ uint32_t sdma_rlc_reg_offset;
unsigned long end_jiffies;
uint32_t data;
uint64_t data64;
uint64_t __user *wptr64 = (uint64_t __user *)wptr;
m = get_sdma_mqd(mqd);
- sdma_base_addr = get_sdma_base_addr(adev, m->sdma_engine_id,
+ sdma_rlc_reg_offset = get_sdma_rlc_reg_offset(adev, m->sdma_engine_id,
m->sdma_queue_id);
- sdmax_gfx_context_cntl = m->sdma_engine_id ?
- SOC15_REG_OFFSET(SDMA1, 0, mmSDMA1_GFX_CONTEXT_CNTL) :
- SOC15_REG_OFFSET(SDMA0, 0, mmSDMA0_GFX_CONTEXT_CNTL);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_CNTL,
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_CNTL,
m->sdmax_rlcx_rb_cntl & (~SDMA0_RLC0_RB_CNTL__RB_ENABLE_MASK));
end_jiffies = msecs_to_jiffies(2000) + jiffies;
while (true) {
- data = RREG32(sdma_base_addr + mmSDMA0_RLC0_CONTEXT_STATUS);
+ data = RREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_CONTEXT_STATUS);
if (data & SDMA0_RLC0_CONTEXT_STATUS__IDLE_MASK)
break;
if (time_after(jiffies, end_jiffies)) {
@@ -433,46 +422,43 @@ static int kgd_hqd_sdma_load(struct kgd_dev *kgd, void *mqd,
}
usleep_range(500, 1000);
}
- data = RREG32(sdmax_gfx_context_cntl);
- data = REG_SET_FIELD(data, SDMA0_GFX_CONTEXT_CNTL,
- RESUME_CTX, 0);
- WREG32(sdmax_gfx_context_cntl, data);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_DOORBELL_OFFSET,
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_DOORBELL_OFFSET,
m->sdmax_rlcx_doorbell_offset);
data = REG_SET_FIELD(m->sdmax_rlcx_doorbell, SDMA0_RLC0_DOORBELL,
ENABLE, 1);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_DOORBELL, data);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_RPTR, m->sdmax_rlcx_rb_rptr);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_RPTR_HI,
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_DOORBELL, data);
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_RPTR,
+ m->sdmax_rlcx_rb_rptr);
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_RPTR_HI,
m->sdmax_rlcx_rb_rptr_hi);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_MINOR_PTR_UPDATE, 1);
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_MINOR_PTR_UPDATE, 1);
if (read_user_wptr(mm, wptr64, data64)) {
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_WPTR,
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_WPTR,
lower_32_bits(data64));
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_WPTR_HI,
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_WPTR_HI,
upper_32_bits(data64));
} else {
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_WPTR,
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_WPTR,
m->sdmax_rlcx_rb_rptr);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_WPTR_HI,
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_WPTR_HI,
m->sdmax_rlcx_rb_rptr_hi);
}
- WREG32(sdma_base_addr + mmSDMA0_RLC0_MINOR_PTR_UPDATE, 0);
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_MINOR_PTR_UPDATE, 0);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_BASE, m->sdmax_rlcx_rb_base);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_BASE_HI,
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_BASE, m->sdmax_rlcx_rb_base);
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_BASE_HI,
m->sdmax_rlcx_rb_base_hi);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_RPTR_ADDR_LO,
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_RPTR_ADDR_LO,
m->sdmax_rlcx_rb_rptr_addr_lo);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_RPTR_ADDR_HI,
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_RPTR_ADDR_HI,
m->sdmax_rlcx_rb_rptr_addr_hi);
data = REG_SET_FIELD(m->sdmax_rlcx_rb_cntl, SDMA0_RLC0_RB_CNTL,
RB_ENABLE, 1);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_CNTL, data);
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_CNTL, data);
return 0;
}
@@ -482,7 +468,8 @@ static int kgd_hqd_sdma_dump(struct kgd_dev *kgd,
uint32_t (**dump)[2], uint32_t *n_regs)
{
struct amdgpu_device *adev = get_amdgpu_device(kgd);
- uint32_t sdma_base_addr = get_sdma_base_addr(adev, engine_id, queue_id);
+ uint32_t sdma_rlc_reg_offset = get_sdma_rlc_reg_offset(adev,
+ engine_id, queue_id);
uint32_t i = 0, reg;
#undef HQD_N_REGS
#define HQD_N_REGS (19+6+7+10)
@@ -492,15 +479,15 @@ static int kgd_hqd_sdma_dump(struct kgd_dev *kgd,
return -ENOMEM;
for (reg = mmSDMA0_RLC0_RB_CNTL; reg <= mmSDMA0_RLC0_DOORBELL; reg++)
- DUMP_REG(sdma_base_addr + reg);
+ DUMP_REG(sdma_rlc_reg_offset + reg);
for (reg = mmSDMA0_RLC0_STATUS; reg <= mmSDMA0_RLC0_CSA_ADDR_HI; reg++)
- DUMP_REG(sdma_base_addr + reg);
+ DUMP_REG(sdma_rlc_reg_offset + reg);
for (reg = mmSDMA0_RLC0_IB_SUB_REMAIN;
reg <= mmSDMA0_RLC0_MINOR_PTR_UPDATE; reg++)
- DUMP_REG(sdma_base_addr + reg);
+ DUMP_REG(sdma_rlc_reg_offset + reg);
for (reg = mmSDMA0_RLC0_MIDCMD_DATA0;
reg <= mmSDMA0_RLC0_MIDCMD_CNTL; reg++)
- DUMP_REG(sdma_base_addr + reg);
+ DUMP_REG(sdma_rlc_reg_offset + reg);
WARN_ON_ONCE(i != HQD_N_REGS);
*n_regs = i;
@@ -534,14 +521,14 @@ static bool kgd_hqd_sdma_is_occupied(struct kgd_dev *kgd, void *mqd)
{
struct amdgpu_device *adev = get_amdgpu_device(kgd);
struct v9_sdma_mqd *m;
- uint32_t sdma_base_addr;
+ uint32_t sdma_rlc_reg_offset;
uint32_t sdma_rlc_rb_cntl;
m = get_sdma_mqd(mqd);
- sdma_base_addr = get_sdma_base_addr(adev, m->sdma_engine_id,
+ sdma_rlc_reg_offset = get_sdma_rlc_reg_offset(adev, m->sdma_engine_id,
m->sdma_queue_id);
- sdma_rlc_rb_cntl = RREG32(sdma_base_addr + mmSDMA0_RLC0_RB_CNTL);
+ sdma_rlc_rb_cntl = RREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_CNTL);
if (sdma_rlc_rb_cntl & SDMA0_RLC0_RB_CNTL__RB_ENABLE_MASK)
return true;
@@ -604,20 +591,20 @@ static int kgd_hqd_sdma_destroy(struct kgd_dev *kgd, void *mqd,
{
struct amdgpu_device *adev = get_amdgpu_device(kgd);
struct v9_sdma_mqd *m;
- uint32_t sdma_base_addr;
+ uint32_t sdma_rlc_reg_offset;
uint32_t temp;
unsigned long end_jiffies = (utimeout * HZ / 1000) + jiffies;
m = get_sdma_mqd(mqd);
- sdma_base_addr = get_sdma_base_addr(adev, m->sdma_engine_id,
+ sdma_rlc_reg_offset = get_sdma_rlc_reg_offset(adev, m->sdma_engine_id,
m->sdma_queue_id);
- temp = RREG32(sdma_base_addr + mmSDMA0_RLC0_RB_CNTL);
+ temp = RREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_CNTL);
temp = temp & ~SDMA0_RLC0_RB_CNTL__RB_ENABLE_MASK;
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_CNTL, temp);
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_CNTL, temp);
while (true) {
- temp = RREG32(sdma_base_addr + mmSDMA0_RLC0_CONTEXT_STATUS);
+ temp = RREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_CONTEXT_STATUS);
if (temp & SDMA0_RLC0_CONTEXT_STATUS__IDLE_MASK)
break;
if (time_after(jiffies, end_jiffies)) {
@@ -627,38 +614,29 @@ static int kgd_hqd_sdma_destroy(struct kgd_dev *kgd, void *mqd,
usleep_range(500, 1000);
}
- WREG32(sdma_base_addr + mmSDMA0_RLC0_DOORBELL, 0);
- WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_CNTL,
- RREG32(sdma_base_addr + mmSDMA0_RLC0_RB_CNTL) |
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_DOORBELL, 0);
+ WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_CNTL,
+ RREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_CNTL) |
SDMA0_RLC0_RB_CNTL__RB_ENABLE_MASK);
- m->sdmax_rlcx_rb_rptr = RREG32(sdma_base_addr + mmSDMA0_RLC0_RB_RPTR);
+ m->sdmax_rlcx_rb_rptr = RREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_RPTR);
m->sdmax_rlcx_rb_rptr_hi =
- RREG32(sdma_base_addr + mmSDMA0_RLC0_RB_RPTR_HI);
+ RREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_RPTR_HI);
return 0;
}
-bool kgd_gfx_v9_get_atc_vmid_pasid_mapping_valid(struct kgd_dev *kgd,
- uint8_t vmid)
+bool kgd_gfx_v9_get_atc_vmid_pasid_mapping_info(struct kgd_dev *kgd,
+ uint8_t vmid, uint16_t *p_pasid)
{
- uint32_t reg;
+ uint32_t value;
struct amdgpu_device *adev = (struct amdgpu_device *) kgd;
- reg = RREG32(SOC15_REG_OFFSET(ATHUB, 0, mmATC_VMID0_PASID_MAPPING)
+ value = RREG32(SOC15_REG_OFFSET(ATHUB, 0, mmATC_VMID0_PASID_MAPPING)
+ vmid);
- return reg & ATC_VMID0_PASID_MAPPING__VALID_MASK;
-}
-
-uint16_t kgd_gfx_v9_get_atc_vmid_pasid_mapping_pasid(struct kgd_dev *kgd,
- uint8_t vmid)
-{
- uint32_t reg;
- struct amdgpu_device *adev = (struct amdgpu_device *) kgd;
+ *p_pasid = value & ATC_VMID0_PASID_MAPPING__PASID_MASK;
- reg = RREG32(SOC15_REG_OFFSET(ATHUB, 0, mmATC_VMID0_PASID_MAPPING)
- + vmid);
- return reg & ATC_VMID0_PASID_MAPPING__PASID_MASK;
+ return !!(value & ATC_VMID0_PASID_MAPPING__VALID_MASK);
}
static int invalidate_tlbs_with_kiq(struct amdgpu_device *adev, uint16_t pasid,
@@ -693,6 +671,8 @@ int kgd_gfx_v9_invalidate_tlbs(struct kgd_dev *kgd, uint16_t pasid)
{
struct amdgpu_device *adev = (struct amdgpu_device *) kgd;
int vmid, i;
+ uint16_t queried_pasid;
+ bool ret;
struct amdgpu_ring *ring = &adev->gfx.kiq.ring;
uint32_t flush_type = 0;
@@ -708,14 +688,14 @@ int kgd_gfx_v9_invalidate_tlbs(struct kgd_dev *kgd, uint16_t pasid)
for (vmid = 0; vmid < 16; vmid++) {
if (!amdgpu_amdkfd_is_kfd_vmid(adev, vmid))
continue;
- if (kgd_gfx_v9_get_atc_vmid_pasid_mapping_valid(kgd, vmid)) {
- if (kgd_gfx_v9_get_atc_vmid_pasid_mapping_pasid(kgd, vmid)
- == pasid) {
- for (i = 0; i < adev->num_vmhubs; i++)
- amdgpu_gmc_flush_gpu_tlb(adev, vmid,
- i, flush_type);
- break;
- }
+
+ ret = kgd_gfx_v9_get_atc_vmid_pasid_mapping_info(kgd, vmid,
+ &queried_pasid);
+ if (ret && queried_pasid == pasid) {
+ for (i = 0; i < adev->num_vmhubs; i++)
+ amdgpu_gmc_flush_gpu_tlb(adev, vmid,
+ i, flush_type);
+ break;
}
}
@@ -1033,7 +1013,7 @@ void kgd_gfx_v9_build_grace_period_packet_info(struct kgd_dev *kgd,
*reg_offset = mmCP_IQ_WAIT_TIME2;
}
-static const struct kfd2kgd_calls kfd2kgd = {
+const struct kfd2kgd_calls gfx_v9_kfd2kgd = {
.program_sh_mem_settings = kgd_gfx_v9_program_sh_mem_settings,
.set_pasid_vmid_mapping = kgd_gfx_v9_set_pasid_vmid_mapping,
.init_interrupts = kgd_gfx_v9_init_interrupts,
@@ -1049,10 +1029,8 @@ static const struct kfd2kgd_calls kfd2kgd = {
.address_watch_execute = kgd_gfx_v9_address_watch_execute,
.wave_control_execute = kgd_gfx_v9_wave_control_execute,
.address_watch_get_offset = kgd_gfx_v9_address_watch_get_offset,
- .get_atc_vmid_pasid_mapping_pasid =
- kgd_gfx_v9_get_atc_vmid_pasid_mapping_pasid,
- .get_atc_vmid_pasid_mapping_valid =
- kgd_gfx_v9_get_atc_vmid_pasid_mapping_valid,
+ .get_atc_vmid_pasid_mapping_info =
+ kgd_gfx_v9_get_atc_vmid_pasid_mapping_info,
.get_tile_config = kgd_gfx_v9_get_tile_config,
.set_vm_context_page_table_base = kgd_gfx_v9_set_vm_context_page_table_base,
.invalidate_tlbs = kgd_gfx_v9_invalidate_tlbs,
@@ -1066,8 +1044,3 @@ static const struct kfd2kgd_calls kfd2kgd = {
.get_iq_wait_times = kgd_gfx_v9_get_iq_wait_times,
.build_grace_period_packet_info = kgd_gfx_v9_build_grace_period_packet_info,
};
-
-struct kfd2kgd_calls *amdgpu_amdkfd_gfx_9_0_get_functions(void)
-{
- return (struct kfd2kgd_calls *)&kfd2kgd;
-}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h
index a30e36341502..7611ba466aa4 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h
@@ -55,10 +55,8 @@ uint32_t kgd_gfx_v9_address_watch_get_offset(struct kgd_dev *kgd,
unsigned int watch_point_id,
unsigned int reg_offset);
-bool kgd_gfx_v9_get_atc_vmid_pasid_mapping_valid(struct kgd_dev *kgd,
- uint8_t vmid);
-uint16_t kgd_gfx_v9_get_atc_vmid_pasid_mapping_pasid(struct kgd_dev *kgd,
- uint8_t vmid);
+bool kgd_gfx_v9_get_atc_vmid_pasid_mapping_info(struct kgd_dev *kgd,
+ uint8_t vmid, uint16_t *p_pasid);
void kgd_gfx_v9_set_vm_context_page_table_base(struct kgd_dev *kgd, uint32_t vmid,
uint64_t page_table_base);
int kgd_gfx_v9_invalidate_tlbs(struct kgd_dev *kgd, uint16_t pasid);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c
index fbff37bd2b03..a253a554f41f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c
@@ -169,8 +169,11 @@ static int convert_atom_mem_type_to_vram_type(struct amdgpu_device *adev,
return vram_type;
}
-int amdgpu_atomfirmware_get_vram_info(struct amdgpu_device *adev,
- int *vram_width, int *vram_type)
+
+int
+amdgpu_atomfirmware_get_vram_info(struct amdgpu_device *adev,
+ int *vram_width, int *vram_type,
+ int *vram_vendor)
{
struct amdgpu_mode_info *mode_info = &adev->mode_info;
int index, i = 0;
@@ -180,6 +183,7 @@ int amdgpu_atomfirmware_get_vram_info(struct amdgpu_device *adev,
union vram_module *vram_module;
u8 frev, crev;
u8 mem_type;
+ u8 mem_vendor;
u32 mem_channel_number;
u32 mem_channel_width;
u32 module_id;
@@ -231,6 +235,9 @@ int amdgpu_atomfirmware_get_vram_info(struct amdgpu_device *adev,
mem_channel_width = vram_module->v9.channel_width;
if (vram_width)
*vram_width = mem_channel_number * (1 << mem_channel_width);
+ mem_vendor = (vram_module->v9.vender_rev_id) & 0xF;
+ if (vram_vendor)
+ *vram_vendor = mem_vendor;
break;
case 4:
if (module_id > vram_info->v24.vram_module_num)
@@ -248,6 +255,9 @@ int amdgpu_atomfirmware_get_vram_info(struct amdgpu_device *adev,
mem_channel_width = vram_module->v10.channel_width;
if (vram_width)
*vram_width = mem_channel_number * (1 << mem_channel_width);
+ mem_vendor = (vram_module->v10.vender_rev_id) & 0xF;
+ if (vram_vendor)
+ *vram_vendor = mem_vendor;
break;
default:
return -EINVAL;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.h
index 82819f03e444..53449fc7baf4 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.h
@@ -30,7 +30,7 @@ bool amdgpu_atomfirmware_gpu_supports_virtualization(struct amdgpu_device *adev)
void amdgpu_atomfirmware_scratch_regs_init(struct amdgpu_device *adev);
int amdgpu_atomfirmware_allocate_fb_scratch(struct amdgpu_device *adev);
int amdgpu_atomfirmware_get_vram_info(struct amdgpu_device *adev,
- int *vram_width, int *vram_type);
+ int *vram_width, int *vram_type, int *vram_vendor);
int amdgpu_atomfirmware_get_clock_info(struct amdgpu_device *adev);
int amdgpu_atomfirmware_get_gfx_info(struct amdgpu_device *adev);
bool amdgpu_atomfirmware_mem_ecc_supported(struct amdgpu_device *adev);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c
index 0416abfb15ed..9b0f71eaae5b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c
@@ -139,7 +139,12 @@ int amdgpu_bo_list_create(struct amdgpu_device *adev, struct drm_file *filp,
return 0;
error_free:
- while (i--) {
+ for (i = 0; i < last_entry; ++i) {
+ struct amdgpu_bo *bo = ttm_to_amdgpu_bo(array[i].tv.bo);
+
+ amdgpu_bo_unref(&bo);
+ }
+ for (i = first_userptr; i < num_entries; ++i) {
struct amdgpu_bo *bo = ttm_to_amdgpu_bo(array[i].tv.bo);
amdgpu_bo_unref(&bo);
@@ -283,7 +288,7 @@ int amdgpu_bo_list_ioctl(struct drm_device *dev, void *data,
r = amdgpu_bo_create_list_entry_array(&args->in, &info);
if (r)
- goto error_free;
+ return r;
switch (args->in.operation) {
case AMDGPU_BO_LIST_OP_CREATE:
@@ -296,8 +301,7 @@ int amdgpu_bo_list_ioctl(struct drm_device *dev, void *data,
r = idr_alloc(&fpriv->bo_list_handles, list, 1, 0, GFP_KERNEL);
mutex_unlock(&fpriv->bo_list_lock);
if (r < 0) {
- amdgpu_bo_list_put(list);
- return r;
+ goto error_put_list;
}
handle = r;
@@ -319,9 +323,8 @@ int amdgpu_bo_list_ioctl(struct drm_device *dev, void *data,
mutex_unlock(&fpriv->bo_list_lock);
if (IS_ERR(old)) {
- amdgpu_bo_list_put(list);
r = PTR_ERR(old);
- goto error_free;
+ goto error_put_list;
}
amdgpu_bo_list_put(old);
@@ -338,8 +341,10 @@ int amdgpu_bo_list_ioctl(struct drm_device *dev, void *data,
return 0;
+error_put_list:
+ amdgpu_bo_list_put(list);
+
error_free:
- if (info)
- kvfree(info);
+ kvfree(info);
return r;
}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c
index 359d9610f7db..09d4619129b0 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c
@@ -1043,8 +1043,12 @@ amdgpu_connector_dvi_detect(struct drm_connector *connector, bool force)
*/
if (amdgpu_connector->shared_ddc && (ret == connector_status_connected)) {
struct drm_connector *list_connector;
+ struct drm_connector_list_iter iter;
struct amdgpu_connector *list_amdgpu_connector;
- list_for_each_entry(list_connector, &dev->mode_config.connector_list, head) {
+
+ drm_connector_list_iter_begin(dev, &iter);
+ drm_for_each_connector_iter(list_connector,
+ &iter) {
if (connector == list_connector)
continue;
list_amdgpu_connector = to_amdgpu_connector(list_connector);
@@ -1061,6 +1065,7 @@ amdgpu_connector_dvi_detect(struct drm_connector *connector, bool force)
}
}
}
+ drm_connector_list_iter_end(&iter);
}
}
}
@@ -1528,6 +1533,7 @@ amdgpu_connector_add(struct amdgpu_device *adev,
{
struct drm_device *dev = adev->ddev;
struct drm_connector *connector;
+ struct drm_connector_list_iter iter;
struct amdgpu_connector *amdgpu_connector;
struct amdgpu_connector_atom_dig *amdgpu_dig_connector;
struct drm_encoder *encoder;
@@ -1541,10 +1547,12 @@ amdgpu_connector_add(struct amdgpu_device *adev,
return;
/* see if we already added it */
- list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
+ drm_connector_list_iter_begin(dev, &iter);
+ drm_for_each_connector_iter(connector, &iter) {
amdgpu_connector = to_amdgpu_connector(connector);
if (amdgpu_connector->connector_id == connector_id) {
amdgpu_connector->devices |= supported_device;
+ drm_connector_list_iter_end(&iter);
return;
}
if (amdgpu_connector->ddc_bus && i2c_bus->valid) {
@@ -1559,6 +1567,7 @@ amdgpu_connector_add(struct amdgpu_device *adev,
}
}
}
+ drm_connector_list_iter_end(&iter);
/* check if it's a dp bridge */
list_for_each_entry(encoder, &dev->mode_config.encoder_list, head) {
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
index 46e67499efaf..1778feae6e35 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
@@ -241,6 +241,8 @@ static int amdgpu_cs_parser_init(struct amdgpu_cs_parser *p, union drm_amdgpu_cs
if (ret)
goto free_all_kdata;
+ p->job->secure = cs->in.flags & AMDGPU_CS_FLAGS_SECURE;
+
if (p->ctx->vram_lost_counter != p->job->vram_lost_counter) {
ret = -ECANCELED;
goto free_all_kdata;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index 80a687d41625..2c8b8afd5125 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -64,6 +64,7 @@
#include "amdgpu_xgmi.h"
#include "amdgpu_ras.h"
#include "amdgpu_pmu.h"
+#include "amdgpu_tmz.h"
#include <linux/suspend.h>
@@ -1030,6 +1031,8 @@ static int amdgpu_device_check_arguments(struct amdgpu_device *adev)
adev->firmware.load_type = amdgpu_ucode_get_load_type(adev, amdgpu_fw_load_type);
amdgpu_direct_gma_size = min(amdgpu_direct_gma_size, 96);
+ adev->tmz.enabled = amdgpu_is_tmz(adev);
+
return ret;
}
@@ -2562,6 +2565,73 @@ static void amdgpu_device_xgmi_reset_func(struct work_struct *__work)
adev->asic_reset_res, adev->ddev->unique);
}
+static int amdgpu_device_get_job_timeout_settings(struct amdgpu_device *adev)
+{
+ char *input = amdgpu_lockup_timeout;
+ char *timeout_setting = NULL;
+ int index = 0;
+ long timeout;
+ int ret = 0;
+
+ /*
+ * By default timeout for non compute jobs is 10000.
+ * And there is no timeout enforced on compute jobs.
+ * In SR-IOV or passthrough mode, timeout for compute
+ * jobs are 10000 by default.
+ */
+ adev->gfx_timeout = msecs_to_jiffies(10000);
+ adev->sdma_timeout = adev->video_timeout = adev->gfx_timeout;
+ if (amdgpu_sriov_vf(adev) || amdgpu_passthrough(adev))
+ adev->compute_timeout = adev->gfx_timeout;
+ else
+ adev->compute_timeout = MAX_SCHEDULE_TIMEOUT;
+
+ if (strnlen(input, AMDGPU_MAX_TIMEOUT_PARAM_LENTH)) {
+ while ((timeout_setting = strsep(&input, ",")) &&
+ strnlen(timeout_setting, AMDGPU_MAX_TIMEOUT_PARAM_LENTH)) {
+ ret = kstrtol(timeout_setting, 0, &timeout);
+ if (ret)
+ return ret;
+
+ if (timeout == 0) {
+ index++;
+ continue;
+ } else if (timeout < 0) {
+ timeout = MAX_SCHEDULE_TIMEOUT;
+ } else {
+ timeout = msecs_to_jiffies(timeout);
+ }
+
+ switch (index++) {
+ case 0:
+ adev->gfx_timeout = timeout;
+ break;
+ case 1:
+ adev->compute_timeout = timeout;
+ break;
+ case 2:
+ adev->sdma_timeout = timeout;
+ break;
+ case 3:
+ adev->video_timeout = timeout;
+ break;
+ default:
+ break;
+ }
+ }
+ /*
+ * There is only one value specified and
+ * it should apply to all non-compute jobs.
+ */
+ if (index == 1) {
+ adev->sdma_timeout = adev->video_timeout = adev->gfx_timeout;
+ if (amdgpu_sriov_vf(adev) || amdgpu_passthrough(adev))
+ adev->compute_timeout = adev->gfx_timeout;
+ }
+ }
+
+ return ret;
+}
/**
* amdgpu_device_init - initialize the driver
@@ -3038,6 +3108,7 @@ int amdgpu_device_suspend(struct drm_device *dev, bool suspend, bool fbcon)
struct amdgpu_device *adev;
struct drm_crtc *crtc;
struct drm_connector *connector;
+ struct drm_connector_list_iter iter;
int r;
if (dev == NULL || dev->dev_private == NULL) {
@@ -3060,9 +3131,11 @@ int amdgpu_device_suspend(struct drm_device *dev, bool suspend, bool fbcon)
if (!amdgpu_device_has_dc_support(adev)) {
/* turn off display hw */
drm_modeset_lock_all(dev);
- list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
- drm_helper_connector_dpms(connector, DRM_MODE_DPMS_OFF);
- }
+ drm_connector_list_iter_begin(dev, &iter);
+ drm_for_each_connector_iter(connector, &iter)
+ drm_helper_connector_dpms(connector,
+ DRM_MODE_DPMS_OFF);
+ drm_connector_list_iter_end(&iter);
drm_modeset_unlock_all(dev);
/* unpin the front buffers and cursors */
list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
@@ -3141,6 +3214,7 @@ int amdgpu_device_suspend(struct drm_device *dev, bool suspend, bool fbcon)
int amdgpu_device_resume(struct drm_device *dev, bool resume, bool fbcon)
{
struct drm_connector *connector;
+ struct drm_connector_list_iter iter;
struct amdgpu_device *adev = dev->dev_private;
struct drm_crtc *crtc;
int r = 0;
@@ -3211,9 +3285,13 @@ int amdgpu_device_resume(struct drm_device *dev, bool resume, bool fbcon)
/* turn on display hw */
drm_modeset_lock_all(dev);
- list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
- drm_helper_connector_dpms(connector, DRM_MODE_DPMS_ON);
- }
+
+ drm_connector_list_iter_begin(dev, &iter);
+ drm_for_each_connector_iter(connector, &iter)
+ drm_helper_connector_dpms(connector,
+ DRM_MODE_DPMS_ON);
+ drm_connector_list_iter_end(&iter);
+
drm_modeset_unlock_all(dev);
}
amdgpu_fbdev_set_suspend_unlocked(adev, 0);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c
index b2907b460240..88981221c518 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c
@@ -622,11 +622,13 @@ void amdgpu_display_print_display_setup(struct drm_device *dev)
struct amdgpu_connector *amdgpu_connector;
struct drm_encoder *encoder;
struct amdgpu_encoder *amdgpu_encoder;
+ struct drm_connector_list_iter iter;
uint32_t devices;
int i = 0;
+ drm_connector_list_iter_begin(dev, &iter);
DRM_INFO("AMDGPU Display Connectors\n");
- list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
+ drm_for_each_connector_iter(connector, &iter) {
amdgpu_connector = to_amdgpu_connector(connector);
DRM_INFO("Connector %d:\n", i);
DRM_INFO(" %s\n", connector->name);
@@ -690,6 +692,7 @@ void amdgpu_display_print_display_setup(struct drm_device *dev)
}
i++;
}
+ drm_connector_list_iter_end(&iter);
}
/**
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index 3ab7a2bc4317..3fc51071df47 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@ -80,13 +80,13 @@
* - 3.32.0 - Add syncobj timeline support to AMDGPU_CS.
* - 3.33.0 - Fixes for GDS ENOMEM failures in AMDGPU_CS.
* - 3.34.0 - Non-DC can flip correctly between buffers with different pitches
+ * - 3.35.0 - Add drm_amdgpu_info_device::tcc_disabled_mask
*/
#define KMS_DRIVER_MAJOR 3
-#define KMS_DRIVER_MINOR 34
+#define KMS_DRIVER_MINOR 35
#define KMS_DRIVER_PATCHLEVEL 0
#define AMDGPU_VERSION "5.0.87"
-#define AMDGPU_MAX_TIMEOUT_PARAM_LENTH 256
int amdgpu_vram_limit = 0;
int amdgpu_vis_vram_limit = 0;
@@ -144,6 +144,7 @@ int amdgpu_discovery = -1;
int amdgpu_mes = 0;
int amdgpu_noretry = 1;
int amdgpu_force_asic_type = -1;
+int amdgpu_tmz = 0;
struct amdgpu_mgpu_info mgpu_info = {
.mutex = __MUTEX_INITIALIZER(mgpu_info.mutex),
@@ -252,9 +253,11 @@ module_param_named(msi, amdgpu_msi, int, 0444);
* By default(with no lockup_timeout settings), the timeout for all non-compute(GFX, SDMA and Video)
* jobs is 10000. And there is no timeout enforced on compute jobs.
*/
-MODULE_PARM_DESC(lockup_timeout, "GPU lockup timeout in ms (default: 10000 for non-compute jobs and infinity timeout for compute jobs."
+MODULE_PARM_DESC(lockup_timeout, "GPU lockup timeout in ms (default: for bare metal 10000 for non-compute jobs and infinity timeout for compute jobs; "
+ "for passthrough or sriov, 10000 for all jobs."
" 0: keep default value. negative: infinity timeout), "
- "format is [Non-Compute] or [GFX,Compute,SDMA,Video]");
+ "format: for bare metal [Non-Compute] or [GFX,Compute,SDMA,Video]; "
+ "for passthrough or sriov [all jobs] or [GFX,Compute,SDMA,Video].");
module_param_string(lockup_timeout, amdgpu_lockup_timeout, sizeof(amdgpu_lockup_timeout), 0444);
/**
@@ -749,6 +752,16 @@ uint amdgpu_dm_abm_level = 0;
MODULE_PARM_DESC(abmlevel, "ABM level (0 = off (default), 1-4 = backlight reduction level) ");
module_param_named(abmlevel, amdgpu_dm_abm_level, uint, 0444);
+/**
+ * DOC: tmz (int)
+ * Trusted Memory Zone (TMZ) is a method to protect data being written
+ * to or read from memory.
+ *
+ * The default value: 0 (off). TODO: change to auto till it is completed.
+ */
+MODULE_PARM_DESC(tmz, "Enable TMZ feature (-1 = auto, 0 = off (default), 1 = on)");
+module_param_named(tmz, amdgpu_tmz, int, 0444);
+
static const struct pci_device_id pciidlist[] = {
#ifdef CONFIG_DRM_AMDGPU_SI
{0x1002, 0x6780, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_TAHITI},
@@ -1055,6 +1068,41 @@ static int amdgpu_pci_probe(struct pci_dev *pdev,
return -ENODEV;
}
+#ifdef CONFIG_DRM_AMDGPU_SI
+ if (!amdgpu_si_support) {
+ switch (flags & AMD_ASIC_MASK) {
+ case CHIP_TAHITI:
+ case CHIP_PITCAIRN:
+ case CHIP_VERDE:
+ case CHIP_OLAND:
+ case CHIP_HAINAN:
+ dev_info(&pdev->dev,
+ "SI support provided by radeon.\n");
+ dev_info(&pdev->dev,
+ "Use radeon.si_support=0 amdgpu.si_support=1 to override.\n"
+ );
+ return -ENODEV;
+ }
+ }
+#endif
+#ifdef CONFIG_DRM_AMDGPU_CIK
+ if (!amdgpu_cik_support) {
+ switch (flags & AMD_ASIC_MASK) {
+ case CHIP_KAVERI:
+ case CHIP_BONAIRE:
+ case CHIP_HAWAII:
+ case CHIP_KABINI:
+ case CHIP_MULLINS:
+ dev_info(&pdev->dev,
+ "CIK support provided by radeon.\n");
+ dev_info(&pdev->dev,
+ "Use radeon.cik_support=0 amdgpu.cik_support=1 to override.\n"
+ );
+ return -ENODEV;
+ }
+ }
+#endif
+
/* Get rid of things like offb */
ret = drm_fb_helper_remove_conflicting_pci_framebuffers(pdev, 0, "amdgpudrmfb");
if (ret)
@@ -1349,71 +1397,6 @@ int amdgpu_file_to_fpriv(struct file *filp, struct amdgpu_fpriv **fpriv)
return 0;
}
-int amdgpu_device_get_job_timeout_settings(struct amdgpu_device *adev)
-{
- char *input = amdgpu_lockup_timeout;
- char *timeout_setting = NULL;
- int index = 0;
- long timeout;
- int ret = 0;
-
- /*
- * By default timeout for non compute jobs is 10000.
- * And there is no timeout enforced on compute jobs.
- * In SR-IOV or passthrough mode, timeout for compute
- * jobs are 10000 by default.
- */
- adev->gfx_timeout = msecs_to_jiffies(10000);
- adev->sdma_timeout = adev->video_timeout = adev->gfx_timeout;
- if (amdgpu_sriov_vf(adev) || amdgpu_passthrough(adev))
- adev->compute_timeout = adev->gfx_timeout;
- else
- adev->compute_timeout = MAX_SCHEDULE_TIMEOUT;
-
- if (strnlen(input, AMDGPU_MAX_TIMEOUT_PARAM_LENTH)) {
- while ((timeout_setting = strsep(&input, ",")) &&
- strnlen(timeout_setting, AMDGPU_MAX_TIMEOUT_PARAM_LENTH)) {
- ret = kstrtol(timeout_setting, 0, &timeout);
- if (ret)
- return ret;
-
- if (timeout == 0) {
- index++;
- continue;
- } else if (timeout < 0) {
- timeout = MAX_SCHEDULE_TIMEOUT;
- } else {
- timeout = msecs_to_jiffies(timeout);
- }
-
- switch (index++) {
- case 0:
- adev->gfx_timeout = timeout;
- break;
- case 1:
- adev->compute_timeout = timeout;
- break;
- case 2:
- adev->sdma_timeout = timeout;
- break;
- case 3:
- adev->video_timeout = timeout;
- break;
- default:
- break;
- }
- }
- /*
- * There is only one value specified and
- * it should apply to all non-compute jobs.
- */
- if (index == 1)
- adev->sdma_timeout = adev->video_timeout = adev->gfx_timeout;
- }
-
- return ret;
-}
-
static struct drm_driver kms_driver = {
.driver_features =
DRIVER_USE_AGP |
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_encoders.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_encoders.c
index ec78e2b2015c..02a42e908b1d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_encoders.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_encoders.c
@@ -37,12 +37,14 @@ amdgpu_link_encoder_connector(struct drm_device *dev)
{
struct amdgpu_device *adev = dev->dev_private;
struct drm_connector *connector;
+ struct drm_connector_list_iter iter;
struct amdgpu_connector *amdgpu_connector;
struct drm_encoder *encoder;
struct amdgpu_encoder *amdgpu_encoder;
+ drm_connector_list_iter_begin(dev, &iter);
/* walk the list and link encoders to connectors */
- list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
+ drm_for_each_connector_iter(connector, &iter) {
amdgpu_connector = to_amdgpu_connector(connector);
list_for_each_entry(encoder, &dev->mode_config.encoder_list, head) {
amdgpu_encoder = to_amdgpu_encoder(encoder);
@@ -55,6 +57,7 @@ amdgpu_link_encoder_connector(struct drm_device *dev)
}
}
}
+ drm_connector_list_iter_end(&iter);
}
void amdgpu_encoder_set_active_device(struct drm_encoder *encoder)
@@ -62,8 +65,10 @@ void amdgpu_encoder_set_active_device(struct drm_encoder *encoder)
struct drm_device *dev = encoder->dev;
struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
struct drm_connector *connector;
+ struct drm_connector_list_iter iter;
- list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
+ drm_connector_list_iter_begin(dev, &iter);
+ drm_for_each_connector_iter(connector, &iter) {
if (connector->encoder == encoder) {
struct amdgpu_connector *amdgpu_connector = to_amdgpu_connector(connector);
amdgpu_encoder->active_device = amdgpu_encoder->devices & amdgpu_connector->devices;
@@ -72,6 +77,7 @@ void amdgpu_encoder_set_active_device(struct drm_encoder *encoder)
amdgpu_connector->devices, encoder->encoder_type);
}
}
+ drm_connector_list_iter_end(&iter);
}
struct drm_connector *
@@ -79,15 +85,20 @@ amdgpu_get_connector_for_encoder(struct drm_encoder *encoder)
{
struct drm_device *dev = encoder->dev;
struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
- struct drm_connector *connector;
+ struct drm_connector *connector, *found = NULL;
+ struct drm_connector_list_iter iter;
struct amdgpu_connector *amdgpu_connector;
- list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
+ drm_connector_list_iter_begin(dev, &iter);
+ drm_for_each_connector_iter(connector, &iter) {
amdgpu_connector = to_amdgpu_connector(connector);
- if (amdgpu_encoder->active_device & amdgpu_connector->devices)
- return connector;
+ if (amdgpu_encoder->active_device & amdgpu_connector->devices) {
+ found = connector;
+ break;
+ }
}
- return NULL;
+ drm_connector_list_iter_end(&iter);
+ return found;
}
struct drm_connector *
@@ -95,15 +106,20 @@ amdgpu_get_connector_for_encoder_init(struct drm_encoder *encoder)
{
struct drm_device *dev = encoder->dev;
struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
- struct drm_connector *connector;
+ struct drm_connector *connector, *found = NULL;
+ struct drm_connector_list_iter iter;
struct amdgpu_connector *amdgpu_connector;
- list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
+ drm_connector_list_iter_begin(dev, &iter);
+ drm_for_each_connector_iter(connector, &iter) {
amdgpu_connector = to_amdgpu_connector(connector);
- if (amdgpu_encoder->devices & amdgpu_connector->devices)
- return connector;
+ if (amdgpu_encoder->devices & amdgpu_connector->devices) {
+ found = connector;
+ break;
+ }
}
- return NULL;
+ drm_connector_list_iter_end(&iter);
+ return found;
}
struct drm_encoder *amdgpu_get_external_encoder(struct drm_encoder *encoder)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
index faf11b5858a2..1612375d5bb7 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
@@ -265,6 +265,7 @@ int amdgpu_gem_create_ioctl(struct drm_device *dev, void *data,
AMDGPU_GEM_CREATE_VRAM_CLEARED |
AMDGPU_GEM_CREATE_VM_ALWAYS_VALID |
AMDGPU_GEM_CREATE_EXPLICIT_SYNC |
+ AMDGPU_GEM_CREATE_ENCRYPTED |
AMDGPU_GEM_CREATE_NO_EVICT))
return -EINVAL;
@@ -273,6 +274,11 @@ int amdgpu_gem_create_ioctl(struct drm_device *dev, void *data,
if (args->in.domains & ~AMDGPU_GEM_DOMAIN_MASK)
return -EINVAL;
+ if (!adev->tmz.enabled && (flags & AMDGPU_GEM_CREATE_ENCRYPTED)) {
+ DRM_ERROR("Cannot allocate secure buffer while tmz is disabled\n");
+ return -EINVAL;
+ }
+
/* create a gem object to contain this object in */
if (args->in.domains & (AMDGPU_GEM_DOMAIN_GDS |
AMDGPU_GEM_DOMAIN_GWS | AMDGPU_GEM_DOMAIN_OA)) {
@@ -294,6 +300,10 @@ int amdgpu_gem_create_ioctl(struct drm_device *dev, void *data,
resv = vm->root.base.bo->tbo.resv;
}
+ if (flags & AMDGPU_GEM_CREATE_ENCRYPTED) {
+ /* XXX: pad out alignment to meet TMZ requirements */
+ }
+
r = amdgpu_gem_object_create(adev, size, args->in.alignment,
(u32)(0xffffffff & args->in.domains),
flags, ttm_bo_type_device, resv, &gobj);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
index fdcb7057093f..56b31668d551 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
@@ -232,12 +232,10 @@ void amdgpu_gfx_compute_queue_acquire(struct amdgpu_device *adev)
void amdgpu_gfx_graphics_queue_acquire(struct amdgpu_device *adev)
{
- int i, queue, pipe, me;
+ int i, queue, me;
for (i = 0; i < AMDGPU_MAX_GFX_QUEUES; ++i) {
queue = i % adev->gfx.me.num_queue_per_pipe;
- pipe = (i / adev->gfx.me.num_queue_per_pipe)
- % adev->gfx.me.num_pipe_per_me;
me = (i / adev->gfx.me.num_queue_per_pipe)
/ adev->gfx.me.num_pipe_per_me;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
index c5179a807a04..35eff9e6ce16 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
@@ -165,6 +165,7 @@ struct amdgpu_gfx_config {
uint32_t num_sc_per_sh;
uint32_t num_packer_per_sc;
uint32_t pa_sc_tile_steering_override;
+ uint64_t tcc_disabled_mask;
};
struct amdgpu_cu_info {
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h
index d591b375d1df..555d8e57fae9 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h
@@ -157,6 +157,7 @@ struct amdgpu_gmc {
uint32_t fw_version;
struct amdgpu_irq_src vm_fault;
uint32_t vram_type;
+ uint8_t vram_vendor;
uint32_t srbm_soft_reset;
bool prt_warning;
uint64_t stolen_size;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
index 0515f7a98a11..d121bbdf0ca7 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
@@ -219,7 +219,7 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs,
if (job && ring->funcs->emit_cntxcntl) {
status |= job->preamble_status;
status |= job->preemption_status;
- amdgpu_ring_emit_cntxcntl(ring, status);
+ amdgpu_ring_emit_cntxcntl(ring, status, job->secure);
}
for (i = 0; i < num_ibs; ++i) {
@@ -238,7 +238,7 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs,
}
if (ring->funcs->emit_tmz)
- amdgpu_ring_emit_tmz(ring, false);
+ amdgpu_ring_emit_tmz(ring, false, job ? job->secure : false);
#ifdef CONFIG_X86_64
if (!(adev->flags & AMD_IS_APU))
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
index 30bdb02b75c9..edd02044cf47 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
@@ -84,10 +84,13 @@ static void amdgpu_hotplug_work_func(struct work_struct *work)
struct drm_device *dev = adev->ddev;
struct drm_mode_config *mode_config = &dev->mode_config;
struct drm_connector *connector;
+ struct drm_connector_list_iter iter;
mutex_lock(&mode_config->mutex);
- list_for_each_entry(connector, &mode_config->connector_list, head)
+ drm_connector_list_iter_begin(dev, &iter);
+ drm_for_each_connector_iter(connector, &iter)
amdgpu_connector_hotplug(connector);
+ drm_connector_list_iter_end(&iter);
mutex_unlock(&mode_config->mutex);
/* Just fire off a uevent and let userspace tell us what to do */
drm_helper_hpd_irq_event(dev);
@@ -239,10 +242,19 @@ int amdgpu_irq_init(struct amdgpu_device *adev)
adev->irq.msi_enabled = false;
if (amdgpu_msi_ok(adev)) {
- int ret = pci_enable_msi(adev->pdev);
- if (!ret) {
+ int nvec = pci_msix_vec_count(adev->pdev);
+ unsigned int flags;
+
+ if (nvec <= 0) {
+ flags = PCI_IRQ_MSI;
+ } else {
+ flags = PCI_IRQ_MSI | PCI_IRQ_MSIX;
+ }
+ /* we only need one vector */
+ nvec = pci_alloc_irq_vectors(adev->pdev, 1, 1, flags);
+ if (nvec > 0) {
adev->irq.msi_enabled = true;
- dev_dbg(adev->dev, "amdgpu: using MSI.\n");
+ dev_dbg(adev->dev, "amdgpu: using MSI/MSI-X.\n");
}
}
@@ -265,7 +277,8 @@ int amdgpu_irq_init(struct amdgpu_device *adev)
INIT_WORK(&adev->irq.ih2_work, amdgpu_irq_handle_ih2);
adev->irq.installed = true;
- r = drm_irq_install(adev->ddev, adev->ddev->pdev->irq);
+ /* Use vector 0 for MSI-X */
+ r = drm_irq_install(adev->ddev, pci_irq_vector(adev->pdev, 0));
if (r) {
adev->irq.installed = false;
if (!amdgpu_device_has_dc_support(adev))
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.h
index dc7ee9358dcd..aa0e375062f2 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.h
@@ -63,6 +63,8 @@ struct amdgpu_job {
uint64_t uf_addr;
uint64_t uf_sequence;
+ /* the job is due to a secure command submission */
+ bool secure;
};
int amdgpu_job_alloc(struct amdgpu_device *adev, unsigned num_ibs,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
index 311e0806c9d8..793f41f39d03 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
@@ -152,38 +152,6 @@ int amdgpu_driver_load_kms(struct drm_device *dev, unsigned long flags)
struct amdgpu_device *adev;
int r, acpi_status;
-#ifdef CONFIG_DRM_AMDGPU_SI
- if (!amdgpu_si_support) {
- switch (flags & AMD_ASIC_MASK) {
- case CHIP_TAHITI:
- case CHIP_PITCAIRN:
- case CHIP_VERDE:
- case CHIP_OLAND:
- case CHIP_HAINAN:
- dev_info(dev->dev,
- "SI support provided by radeon.\n");
- dev_info(dev->dev,
- "Use radeon.si_support=0 amdgpu.si_support=1 to override.\n"
- );
- return -ENODEV;
- }
- }
-#endif
-#ifdef CONFIG_DRM_AMDGPU_CIK
- if (!amdgpu_cik_support) {
- switch (flags & AMD_ASIC_MASK) {
- case CHIP_KAVERI:
- case CHIP_BONAIRE:
- case CHIP_HAWAII:
- case CHIP_KABINI:
- case CHIP_MULLINS:
- dev_info(dev->dev,
- "CIK support disabled by module param\n");
- return -ENODEV;
- }
- }
-#endif
-
adev = kzalloc(sizeof(struct amdgpu_device), GFP_KERNEL);
if (adev == NULL) {
return -ENOMEM;
@@ -810,6 +778,8 @@ static int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file
dev_info.pa_sc_tile_steering_override =
adev->gfx.config.pa_sc_tile_steering_override;
+ dev_info.tcc_disabled_mask = adev->gfx.config.tcc_disabled_mask;
+
return copy_to_user(out, &dev_info,
min((size_t)size, sizeof(dev_info))) ? -EFAULT : 0;
}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h
index 1f26a17e6561..919bd566ba3c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h
@@ -67,6 +67,8 @@ struct amdgpu_nbio_funcs {
bool enable);
void (*ih_doorbell_range)(struct amdgpu_device *adev,
bool use_doorbell, int doorbell_index);
+ void (*enable_doorbell_interrupt)(struct amdgpu_device *adev,
+ bool enable);
void (*update_medium_grain_clock_gating)(struct amdgpu_device *adev,
bool enable);
void (*update_medium_grain_light_sleep)(struct amdgpu_device *adev,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
index ae07e69cac70..0c3d49e465c3 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
@@ -538,7 +538,8 @@ static int amdgpu_bo_do_create(struct amdgpu_device *adev,
.interruptible = (bp->type != ttm_bo_type_kernel),
.no_wait_gpu = bp->no_wait_gpu,
.resv = bp->resv,
- .flags = TTM_OPT_FLAG_ALLOW_RES_EVICT
+ .flags = bp->type != ttm_bo_type_kernel ?
+ TTM_OPT_FLAG_ALLOW_RES_EVICT : 0
};
struct amdgpu_bo *bo;
unsigned long page_align, size = bp->size;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h
index 57f41bed2898..38a6ba7ad9be 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h
@@ -229,6 +229,17 @@ static inline bool amdgpu_bo_explicit_sync(struct amdgpu_bo *bo)
return bo->flags & AMDGPU_GEM_CREATE_EXPLICIT_SYNC;
}
+/**
+ * amdgpu_bo_encrypted - test if the BO is encrypted
+ * @bo: pointer to a buffer object
+ *
+ * Return true if the buffer object is encrypted, false otherwise.
+ */
+static inline bool amdgpu_bo_encrypted(struct amdgpu_bo *bo)
+{
+ return bo->flags & AMDGPU_GEM_CREATE_ENCRYPTED;
+}
+
bool amdgpu_bo_is_amdgpu_bo(struct ttm_buffer_object *bo);
void amdgpu_bo_placement_from_domain(struct amdgpu_bo *abo, u32 domain);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
index 0e3857548163..628d6459ef03 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
@@ -69,6 +69,9 @@ const char *ras_block_string[] = {
atomic_t amdgpu_ras_in_intr = ATOMIC_INIT(0);
+static bool amdgpu_ras_check_bad_page(struct amdgpu_device *adev,
+ uint64_t addr);
+
static ssize_t amdgpu_ras_debugfs_read(struct file *f, char __user *buf,
size_t size, loff_t *pos)
{
@@ -213,11 +216,12 @@ static struct ras_manager *amdgpu_ras_find_obj(struct amdgpu_device *adev,
* value to the address.
*
* Second member: struct ras_debug_if::op.
- * It has three kinds of operations.
+ * It has four kinds of operations.
*
* - 0: disable RAS on the block. Take ::head as its data.
* - 1: enable RAS on the block. Take ::head as its data.
* - 2: inject errors on the block. Take ::inject as its data.
+ * - 3: reboot on unrecoverable error
*
* How to use the interface?
* programs:
@@ -226,13 +230,13 @@ static struct ras_manager *amdgpu_ras_find_obj(struct amdgpu_device *adev,
*
* .. code-block:: bash
*
- * echo op block [error [sub_blcok address value]] > .../ras/ras_ctrl
+ * echo op block [error [sub_block address value]] > .../ras/ras_ctrl
*
* op: disable, enable, inject
* disable: only block is needed
* enable: block and error are needed
* inject: error, address, value are needed
- * block: umc, smda, gfx, .........
+ * block: umc, sdma, gfx, .........
* see ras_block_string[] for details
* error: ue, ce
* ue: multi_uncorrectable
@@ -288,6 +292,14 @@ static ssize_t amdgpu_ras_debugfs_ctrl_write(struct file *f, const char __user *
break;
}
+ /* umc ce/ue error injection for a bad page is not allowed */
+ if ((data.head.block == AMDGPU_RAS_BLOCK__UMC) &&
+ amdgpu_ras_check_bad_page(adev, data.inject.address)) {
+ DRM_WARN("RAS WARN: 0x%llx has been marked as bad before error injection!\n",
+ data.inject.address);
+ break;
+ }
+
/* data.inject.address is offset instead of absolute gpu address */
ret = amdgpu_ras_error_inject(adev, &data.inject);
break;
@@ -1432,6 +1444,39 @@ out:
return ret;
}
+/*
+ * check if an address belongs to bad page
+ *
+ * Note: this check is only for umc block
+ */
+static bool amdgpu_ras_check_bad_page(struct amdgpu_device *adev,
+ uint64_t addr)
+{
+ struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
+ struct ras_err_handler_data *data;
+ int i;
+ bool ret = false;
+
+ if (!con || !con->eh_data)
+ return ret;
+
+ mutex_lock(&con->recovery_lock);
+ data = con->eh_data;
+ if (!data)
+ goto out;
+
+ addr >>= AMDGPU_GPU_PAGE_SHIFT;
+ for (i = 0; i < data->count; i++)
+ if (addr == data->bps[i].retired_page) {
+ ret = true;
+ goto out;
+ }
+
+out:
+ mutex_unlock(&con->recovery_lock);
+ return ret;
+}
+
/* called in gpu recovery/init */
int amdgpu_ras_reserve_bad_pages(struct amdgpu_device *adev)
{
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
index 9aa89eba983c..193758fd2c93 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
@@ -160,7 +160,8 @@ struct amdgpu_ring_funcs {
void (*begin_use)(struct amdgpu_ring *ring);
void (*end_use)(struct amdgpu_ring *ring);
void (*emit_switch_buffer) (struct amdgpu_ring *ring);
- void (*emit_cntxcntl) (struct amdgpu_ring *ring, uint32_t flags);
+ void (*emit_cntxcntl) (struct amdgpu_ring *ring, uint32_t flags,
+ bool trusted);
void (*emit_rreg)(struct amdgpu_ring *ring, uint32_t reg);
void (*emit_wreg)(struct amdgpu_ring *ring, uint32_t reg, uint32_t val);
void (*emit_reg_wait)(struct amdgpu_ring *ring, uint32_t reg,
@@ -168,7 +169,7 @@ struct amdgpu_ring_funcs {
void (*emit_reg_write_reg_wait)(struct amdgpu_ring *ring,
uint32_t reg0, uint32_t reg1,
uint32_t ref, uint32_t mask);
- void (*emit_tmz)(struct amdgpu_ring *ring, bool start);
+ void (*emit_tmz)(struct amdgpu_ring *ring, bool start, bool trusted);
/* priority functions */
void (*set_priority) (struct amdgpu_ring *ring,
enum drm_sched_priority priority);
@@ -244,12 +245,12 @@ struct amdgpu_ring {
#define amdgpu_ring_emit_gds_switch(r, v, db, ds, wb, ws, ab, as) (r)->funcs->emit_gds_switch((r), (v), (db), (ds), (wb), (ws), (ab), (as))
#define amdgpu_ring_emit_hdp_flush(r) (r)->funcs->emit_hdp_flush((r))
#define amdgpu_ring_emit_switch_buffer(r) (r)->funcs->emit_switch_buffer((r))
-#define amdgpu_ring_emit_cntxcntl(r, d) (r)->funcs->emit_cntxcntl((r), (d))
+#define amdgpu_ring_emit_cntxcntl(r, d, s) (r)->funcs->emit_cntxcntl((r), (d), (s))
#define amdgpu_ring_emit_rreg(r, d) (r)->funcs->emit_rreg((r), (d))
#define amdgpu_ring_emit_wreg(r, d, v) (r)->funcs->emit_wreg((r), (d), (v))
#define amdgpu_ring_emit_reg_wait(r, d, v, m) (r)->funcs->emit_reg_wait((r), (d), (v), (m))
#define amdgpu_ring_emit_reg_write_reg_wait(r, d0, d1, v, m) (r)->funcs->emit_reg_write_reg_wait((r), (d0), (d1), (v), (m))
-#define amdgpu_ring_emit_tmz(r, b) (r)->funcs->emit_tmz((r), (b))
+#define amdgpu_ring_emit_tmz(r, b, s) (r)->funcs->emit_tmz((r), (b), (s))
#define amdgpu_ring_pad_ib(r, ib) ((r)->funcs->pad_ib((r), (ib)))
#define amdgpu_ring_init_cond_exec(r) (r)->funcs->init_cond_exec((r))
#define amdgpu_ring_patch_cond_exec(r,o) (r)->funcs->patch_cond_exec((r),(o))
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_tmz.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_tmz.c
new file mode 100644
index 000000000000..14a55003dd81
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_tmz.c
@@ -0,0 +1,49 @@
+/*
+ * Copyright 2019 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 <drm/drmP.h>
+#include "amdgpu.h"
+#include "amdgpu_tmz.h"
+
+
+/**
+ * amdgpu_is_tmz - validate trust memory zone
+ *
+ * @adev: amdgpu_device pointer
+ *
+ * Return true if @dev supports trusted memory zones (TMZ), and return false if
+ * @dev does not support TMZ.
+ */
+bool amdgpu_is_tmz(struct amdgpu_device *adev)
+{
+ if (!amdgpu_tmz)
+ return false;
+
+ if (adev->asic_type < CHIP_RAVEN || adev->asic_type == CHIP_ARCTURUS) {
+ dev_warn(adev->dev, "doesn't support trusted memory zones (TMZ)\n");
+ return false;
+ }
+
+ dev_info(adev->dev, "TMZ feature is enabled\n");
+
+ return true;
+}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_tmz.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_tmz.h
new file mode 100644
index 000000000000..28e05177fb89
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_tmz.h
@@ -0,0 +1,39 @@
+/*
+ * Copyright 2019 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.
+ *
+ */
+
+#ifndef __AMDGPU_TMZ_H__
+#define __AMDGPU_TMZ_H__
+
+#include "amdgpu.h"
+
+/*
+ * Trust memory zone stuff
+ */
+struct amdgpu_tmz {
+ bool enabled;
+};
+
+
+extern bool amdgpu_is_tmz(struct amdgpu_device *adev);
+
+#endif
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
index ed7b10e0848d..2a73f54c5dbd 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
@@ -1684,7 +1684,6 @@ static void amdgpu_ttm_fw_reserve_vram_fini(struct amdgpu_device *adev)
static int amdgpu_ttm_fw_reserve_vram_init(struct amdgpu_device *adev)
{
uint64_t vram_size = adev->gmc.visible_vram_size;
- int r;
adev->fw_vram_usage.va = NULL;
adev->fw_vram_usage.reserved_bo = NULL;
@@ -1699,7 +1698,6 @@ static int amdgpu_ttm_fw_reserve_vram_init(struct amdgpu_device *adev)
AMDGPU_GEM_DOMAIN_VRAM,
&adev->fw_vram_usage.reserved_bo,
&adev->fw_vram_usage.va);
- return r;
}
static int amdgpu_direct_gma_init(struct amdgpu_device *adev)
@@ -2150,10 +2148,7 @@ static int amdgpu_map_buffer(struct ttm_buffer_object *bo,
*addr += (u64)window * AMDGPU_GTT_MAX_TRANSFER_SIZE *
AMDGPU_GPU_PAGE_SIZE;
- num_dw = adev->mman.buffer_funcs->copy_num_dw;
- while (num_dw & 0x7)
- num_dw++;
-
+ num_dw = ALIGN(adev->mman.buffer_funcs->copy_num_dw, 8);
num_bytes = num_pages * 8;
r = amdgpu_job_alloc_with_ib(adev, num_dw * 4 + num_bytes, &job);
@@ -2213,11 +2208,7 @@ int amdgpu_copy_buffer(struct amdgpu_ring *ring, uint64_t src_offset,
max_bytes = adev->mman.buffer_funcs->copy_max_bytes;
num_loops = DIV_ROUND_UP(byte_count, max_bytes);
- num_dw = num_loops * adev->mman.buffer_funcs->copy_num_dw;
-
- /* for IB padding */
- while (num_dw & 0x7)
- num_dw++;
+ num_dw = ALIGN(num_loops * adev->mman.buffer_funcs->copy_num_dw, 8);
r = amdgpu_job_alloc_with_ib(adev, num_dw * 4, &job);
if (r)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
index c80198a10b92..3642b0ebc9e3 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
@@ -1060,7 +1060,7 @@ int amdgpu_vm_flush(struct amdgpu_ring *ring, struct amdgpu_job *job,
id->oa_size != job->oa_size);
bool vm_flush_needed = job->vm_needs_flush;
struct dma_fence *fence = NULL;
- bool pasid_mapping_needed;
+ bool pasid_mapping_needed = false;
unsigned patch_offset = 0;
int r;
@@ -1747,6 +1747,10 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev, struct amdgpu_bo_va *bo_va,
if (bo) {
flags = amdgpu_ttm_tt_pte_flags(adev, bo->tbo.ttm, mem);
+
+ if (amdgpu_bo_encrypted(bo))
+ flags |= AMDGPU_PTE_TMZ;
+
bo_adev = amdgpu_ttm_adev(bo->tbo.bdev);
} else {
flags = 0x0;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
index cffa01935fce..3325625d8e63 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
@@ -58,6 +58,9 @@ struct amdgpu_bo_list_entry;
#define AMDGPU_PTE_SYSTEM (1ULL << 1)
#define AMDGPU_PTE_SNOOPED (1ULL << 2)
+/* RV+ */
+#define AMDGPU_PTE_TMZ (1ULL << 3)
+
/* VI only */
#define AMDGPU_PTE_EXECUTABLE (1ULL << 4)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
index ebbc72ed368e..ce367d3ba791 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
@@ -25,6 +25,8 @@
#include <drm/drmP.h>
#include "amdgpu.h"
#include "amdgpu_vm.h"
+#include "amdgpu_atomfirmware.h"
+#include "atom.h"
struct amdgpu_vram_mgr {
struct drm_mm mm;
@@ -103,6 +105,39 @@ static ssize_t amdgpu_mem_info_vis_vram_used_show(struct device *dev,
amdgpu_vram_mgr_vis_usage(&adev->mman.bdev.man[TTM_PL_VRAM]));
}
+static ssize_t amdgpu_mem_info_vram_vendor(struct device *dev,
+ struct device_attribute *attr,
+ char *buf)
+{
+ struct drm_device *ddev = dev_get_drvdata(dev);
+ struct amdgpu_device *adev = ddev->dev_private;
+
+ switch (adev->gmc.vram_vendor) {
+ case SAMSUNG:
+ return snprintf(buf, PAGE_SIZE, "samsung\n");
+ case INFINEON:
+ return snprintf(buf, PAGE_SIZE, "infineon\n");
+ case ELPIDA:
+ return snprintf(buf, PAGE_SIZE, "elpida\n");
+ case ETRON:
+ return snprintf(buf, PAGE_SIZE, "etron\n");
+ case NANYA:
+ return snprintf(buf, PAGE_SIZE, "nanya\n");
+ case HYNIX:
+ return snprintf(buf, PAGE_SIZE, "hynix\n");
+ case MOSEL:
+ return snprintf(buf, PAGE_SIZE, "mosel\n");
+ case WINBOND:
+ return snprintf(buf, PAGE_SIZE, "winbond\n");
+ case ESMT:
+ return snprintf(buf, PAGE_SIZE, "esmt\n");
+ case MICRON:
+ return snprintf(buf, PAGE_SIZE, "micron\n");
+ default:
+ return snprintf(buf, PAGE_SIZE, "unknown\n");
+ }
+}
+
static DEVICE_ATTR(mem_info_vram_total, S_IRUGO,
amdgpu_mem_info_vram_total_show, NULL);
static DEVICE_ATTR(mem_info_vis_vram_total, S_IRUGO,
@@ -111,6 +146,8 @@ static DEVICE_ATTR(mem_info_vram_used, S_IRUGO,
amdgpu_mem_info_vram_used_show, NULL);
static DEVICE_ATTR(mem_info_vis_vram_used, S_IRUGO,
amdgpu_mem_info_vis_vram_used_show, NULL);
+static DEVICE_ATTR(mem_info_vram_vendor, S_IRUGO,
+ amdgpu_mem_info_vram_vendor, NULL);
/**
* amdgpu_vram_mgr_init - init VRAM manager and DRM MM
@@ -156,6 +193,11 @@ static int amdgpu_vram_mgr_init(struct ttm_mem_type_manager *man,
DRM_ERROR("Failed to create device file mem_info_vis_vram_used\n");
return ret;
}
+ ret = device_create_file(adev->dev, &dev_attr_mem_info_vram_vendor);
+ if (ret) {
+ DRM_ERROR("Failed to create device file mem_info_vram_vendor\n");
+ return ret;
+ }
return 0;
}
@@ -182,6 +224,7 @@ static int amdgpu_vram_mgr_fini(struct ttm_mem_type_manager *man)
device_remove_file(adev->dev, &dev_attr_mem_info_vis_vram_total);
device_remove_file(adev->dev, &dev_attr_mem_info_vram_used);
device_remove_file(adev->dev, &dev_attr_mem_info_vis_vram_used);
+ device_remove_file(adev->dev, &dev_attr_mem_info_vram_vendor);
return 0;
}
diff --git a/drivers/gpu/drm/amd/amdgpu/arct_reg_init.c b/drivers/gpu/drm/amd/amdgpu/arct_reg_init.c
index e62609d5126b..fda99c958c3b 100644
--- a/drivers/gpu/drm/amd/amdgpu/arct_reg_init.c
+++ b/drivers/gpu/drm/amd/amdgpu/arct_reg_init.c
@@ -24,7 +24,6 @@
#include "soc15.h"
#include "soc15_common.h"
-#include "soc15_hw_ip.h"
#include "arct_ip_offset.h"
int arct_reg_base_init(struct amdgpu_device *adev)
diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
index 4505cbfe95c1..f3a8caa59657 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
@@ -331,9 +331,11 @@ static void dce_v10_0_hpd_init(struct amdgpu_device *adev)
{
struct drm_device *dev = adev->ddev;
struct drm_connector *connector;
+ struct drm_connector_list_iter iter;
u32 tmp;
- list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
+ drm_connector_list_iter_begin(dev, &iter);
+ drm_for_each_connector_iter(connector, &iter) {
struct amdgpu_connector *amdgpu_connector = to_amdgpu_connector(connector);
if (amdgpu_connector->hpd.hpd >= adev->mode_info.num_hpd)
@@ -369,6 +371,7 @@ static void dce_v10_0_hpd_init(struct amdgpu_device *adev)
amdgpu_irq_get(adev, &adev->hpd_irq,
amdgpu_connector->hpd.hpd);
}
+ drm_connector_list_iter_end(&iter);
}
/**
@@ -383,9 +386,11 @@ static void dce_v10_0_hpd_fini(struct amdgpu_device *adev)
{
struct drm_device *dev = adev->ddev;
struct drm_connector *connector;
+ struct drm_connector_list_iter iter;
u32 tmp;
- list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
+ drm_connector_list_iter_begin(dev, &iter);
+ drm_for_each_connector_iter(connector, &iter) {
struct amdgpu_connector *amdgpu_connector = to_amdgpu_connector(connector);
if (amdgpu_connector->hpd.hpd >= adev->mode_info.num_hpd)
@@ -398,6 +403,7 @@ static void dce_v10_0_hpd_fini(struct amdgpu_device *adev)
amdgpu_irq_put(adev, &adev->hpd_irq,
amdgpu_connector->hpd.hpd);
}
+ drm_connector_list_iter_end(&iter);
}
static u32 dce_v10_0_hpd_get_gpio_reg(struct amdgpu_device *adev)
@@ -1220,10 +1226,12 @@ static void dce_v10_0_afmt_audio_select_pin(struct drm_encoder *encoder)
static void dce_v10_0_audio_write_latency_fields(struct drm_encoder *encoder,
struct drm_display_mode *mode)
{
- struct amdgpu_device *adev = encoder->dev->dev_private;
+ struct drm_device *dev = encoder->dev;
+ struct amdgpu_device *adev = dev->dev_private;
struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
struct drm_connector *connector;
+ struct drm_connector_list_iter iter;
struct amdgpu_connector *amdgpu_connector = NULL;
u32 tmp;
int interlace = 0;
@@ -1231,12 +1239,14 @@ static void dce_v10_0_audio_write_latency_fields(struct drm_encoder *encoder,
if (!dig || !dig->afmt || !dig->afmt->pin)
return;
- list_for_each_entry(connector, &encoder->dev->mode_config.connector_list, head) {
+ drm_connector_list_iter_begin(dev, &iter);
+ drm_for_each_connector_iter(connector, &iter) {
if (connector->encoder == encoder) {
amdgpu_connector = to_amdgpu_connector(connector);
break;
}
}
+ drm_connector_list_iter_end(&iter);
if (!amdgpu_connector) {
DRM_ERROR("Couldn't find encoder's connector\n");
@@ -1262,10 +1272,12 @@ static void dce_v10_0_audio_write_latency_fields(struct drm_encoder *encoder,
static void dce_v10_0_audio_write_speaker_allocation(struct drm_encoder *encoder)
{
- struct amdgpu_device *adev = encoder->dev->dev_private;
+ struct drm_device *dev = encoder->dev;
+ struct amdgpu_device *adev = dev->dev_private;
struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
struct drm_connector *connector;
+ struct drm_connector_list_iter iter;
struct amdgpu_connector *amdgpu_connector = NULL;
u32 tmp;
u8 *sadb = NULL;
@@ -1274,12 +1286,14 @@ static void dce_v10_0_audio_write_speaker_allocation(struct drm_encoder *encoder
if (!dig || !dig->afmt || !dig->afmt->pin)
return;
- list_for_each_entry(connector, &encoder->dev->mode_config.connector_list, head) {
+ drm_connector_list_iter_begin(dev, &iter);
+ drm_for_each_connector_iter(connector, &iter) {
if (connector->encoder == encoder) {
amdgpu_connector = to_amdgpu_connector(connector);
break;
}
}
+ drm_connector_list_iter_end(&iter);
if (!amdgpu_connector) {
DRM_ERROR("Couldn't find encoder's connector\n");
@@ -1314,10 +1328,12 @@ static void dce_v10_0_audio_write_speaker_allocation(struct drm_encoder *encoder
static void dce_v10_0_audio_write_sad_regs(struct drm_encoder *encoder)
{
- struct amdgpu_device *adev = encoder->dev->dev_private;
+ struct drm_device *dev = encoder->dev;
+ struct amdgpu_device *adev = dev->dev_private;
struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
struct drm_connector *connector;
+ struct drm_connector_list_iter iter;
struct amdgpu_connector *amdgpu_connector = NULL;
struct cea_sad *sads;
int i, sad_count;
@@ -1340,12 +1356,14 @@ static void dce_v10_0_audio_write_sad_regs(struct drm_encoder *encoder)
if (!dig || !dig->afmt || !dig->afmt->pin)
return;
- list_for_each_entry(connector, &encoder->dev->mode_config.connector_list, head) {
+ drm_connector_list_iter_begin(dev, &iter);
+ drm_for_each_connector_iter(connector, &iter) {
if (connector->encoder == encoder) {
amdgpu_connector = to_amdgpu_connector(connector);
break;
}
}
+ drm_connector_list_iter_end(&iter);
if (!amdgpu_connector) {
DRM_ERROR("Couldn't find encoder's connector\n");
diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
index c3eb728ec71c..6982f962c3d9 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
@@ -349,9 +349,11 @@ static void dce_v11_0_hpd_init(struct amdgpu_device *adev)
{
struct drm_device *dev = adev->ddev;
struct drm_connector *connector;
+ struct drm_connector_list_iter iter;
u32 tmp;
- list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
+ drm_connector_list_iter_begin(dev, &iter);
+ drm_for_each_connector_iter(connector, &iter) {
struct amdgpu_connector *amdgpu_connector = to_amdgpu_connector(connector);
if (amdgpu_connector->hpd.hpd >= adev->mode_info.num_hpd)
@@ -386,6 +388,7 @@ static void dce_v11_0_hpd_init(struct amdgpu_device *adev)
dce_v11_0_hpd_set_polarity(adev, amdgpu_connector->hpd.hpd);
amdgpu_irq_get(adev, &adev->hpd_irq, amdgpu_connector->hpd.hpd);
}
+ drm_connector_list_iter_end(&iter);
}
/**
@@ -400,9 +403,11 @@ static void dce_v11_0_hpd_fini(struct amdgpu_device *adev)
{
struct drm_device *dev = adev->ddev;
struct drm_connector *connector;
+ struct drm_connector_list_iter iter;
u32 tmp;
- list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
+ drm_connector_list_iter_begin(dev, &iter);
+ drm_for_each_connector_iter(connector, &iter) {
struct amdgpu_connector *amdgpu_connector = to_amdgpu_connector(connector);
if (amdgpu_connector->hpd.hpd >= adev->mode_info.num_hpd)
@@ -414,6 +419,7 @@ static void dce_v11_0_hpd_fini(struct amdgpu_device *adev)
amdgpu_irq_put(adev, &adev->hpd_irq, amdgpu_connector->hpd.hpd);
}
+ drm_connector_list_iter_end(&iter);
}
static u32 dce_v11_0_hpd_get_gpio_reg(struct amdgpu_device *adev)
@@ -1246,10 +1252,12 @@ static void dce_v11_0_afmt_audio_select_pin(struct drm_encoder *encoder)
static void dce_v11_0_audio_write_latency_fields(struct drm_encoder *encoder,
struct drm_display_mode *mode)
{
- struct amdgpu_device *adev = encoder->dev->dev_private;
+ struct drm_device *dev = encoder->dev;
+ struct amdgpu_device *adev = dev->dev_private;
struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
struct drm_connector *connector;
+ struct drm_connector_list_iter iter;
struct amdgpu_connector *amdgpu_connector = NULL;
u32 tmp;
int interlace = 0;
@@ -1257,12 +1265,14 @@ static void dce_v11_0_audio_write_latency_fields(struct drm_encoder *encoder,
if (!dig || !dig->afmt || !dig->afmt->pin)
return;
- list_for_each_entry(connector, &encoder->dev->mode_config.connector_list, head) {
+ drm_connector_list_iter_begin(dev, &iter);
+ drm_for_each_connector_iter(connector, &iter) {
if (connector->encoder == encoder) {
amdgpu_connector = to_amdgpu_connector(connector);
break;
}
}
+ drm_connector_list_iter_end(&iter);
if (!amdgpu_connector) {
DRM_ERROR("Couldn't find encoder's connector\n");
@@ -1288,10 +1298,12 @@ static void dce_v11_0_audio_write_latency_fields(struct drm_encoder *encoder,
static void dce_v11_0_audio_write_speaker_allocation(struct drm_encoder *encoder)
{
- struct amdgpu_device *adev = encoder->dev->dev_private;
+ struct drm_device *dev = encoder->dev;
+ struct amdgpu_device *adev = dev->dev_private;
struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
struct drm_connector *connector;
+ struct drm_connector_list_iter iter;
struct amdgpu_connector *amdgpu_connector = NULL;
u32 tmp;
u8 *sadb = NULL;
@@ -1300,12 +1312,14 @@ static void dce_v11_0_audio_write_speaker_allocation(struct drm_encoder *encoder
if (!dig || !dig->afmt || !dig->afmt->pin)
return;
- list_for_each_entry(connector, &encoder->dev->mode_config.connector_list, head) {
+ drm_connector_list_iter_begin(dev, &iter);
+ drm_for_each_connector_iter(connector, &iter) {
if (connector->encoder == encoder) {
amdgpu_connector = to_amdgpu_connector(connector);
break;
}
}
+ drm_connector_list_iter_end(&iter);
if (!amdgpu_connector) {
DRM_ERROR("Couldn't find encoder's connector\n");
@@ -1340,10 +1354,12 @@ static void dce_v11_0_audio_write_speaker_allocation(struct drm_encoder *encoder
static void dce_v11_0_audio_write_sad_regs(struct drm_encoder *encoder)
{
- struct amdgpu_device *adev = encoder->dev->dev_private;
+ struct drm_device *dev = encoder->dev;
+ struct amdgpu_device *adev = dev->dev_private;
struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
struct drm_connector *connector;
+ struct drm_connector_list_iter iter;
struct amdgpu_connector *amdgpu_connector = NULL;
struct cea_sad *sads;
int i, sad_count;
@@ -1366,12 +1382,14 @@ static void dce_v11_0_audio_write_sad_regs(struct drm_encoder *encoder)
if (!dig || !dig->afmt || !dig->afmt->pin)
return;
- list_for_each_entry(connector, &encoder->dev->mode_config.connector_list, head) {
+ drm_connector_list_iter_begin(dev, &iter);
+ drm_for_each_connector_iter(connector, &iter) {
if (connector->encoder == encoder) {
amdgpu_connector = to_amdgpu_connector(connector);
break;
}
}
+ drm_connector_list_iter_end(&iter);
if (!amdgpu_connector) {
DRM_ERROR("Couldn't find encoder's connector\n");
diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c
index e843bec703d3..f50876ef5c53 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c
@@ -280,9 +280,11 @@ static void dce_v6_0_hpd_init(struct amdgpu_device *adev)
{
struct drm_device *dev = adev->ddev;
struct drm_connector *connector;
+ struct drm_connector_list_iter iter;
u32 tmp;
- list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
+ drm_connector_list_iter_begin(dev, &iter);
+ drm_for_each_connector_iter(connector, &iter) {
struct amdgpu_connector *amdgpu_connector = to_amdgpu_connector(connector);
if (amdgpu_connector->hpd.hpd >= adev->mode_info.num_hpd)
@@ -308,7 +310,7 @@ static void dce_v6_0_hpd_init(struct amdgpu_device *adev)
dce_v6_0_hpd_set_polarity(adev, amdgpu_connector->hpd.hpd);
amdgpu_irq_get(adev, &adev->hpd_irq, amdgpu_connector->hpd.hpd);
}
-
+ drm_connector_list_iter_end(&iter);
}
/**
@@ -323,9 +325,11 @@ static void dce_v6_0_hpd_fini(struct amdgpu_device *adev)
{
struct drm_device *dev = adev->ddev;
struct drm_connector *connector;
+ struct drm_connector_list_iter iter;
u32 tmp;
- list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
+ drm_connector_list_iter_begin(dev, &iter);
+ drm_for_each_connector_iter(connector, &iter) {
struct amdgpu_connector *amdgpu_connector = to_amdgpu_connector(connector);
if (amdgpu_connector->hpd.hpd >= adev->mode_info.num_hpd)
@@ -337,6 +341,7 @@ static void dce_v6_0_hpd_fini(struct amdgpu_device *adev)
amdgpu_irq_put(adev, &adev->hpd_irq, amdgpu_connector->hpd.hpd);
}
+ drm_connector_list_iter_end(&iter);
}
static u32 dce_v6_0_hpd_get_gpio_reg(struct amdgpu_device *adev)
@@ -1123,20 +1128,24 @@ static void dce_v6_0_audio_select_pin(struct drm_encoder *encoder)
static void dce_v6_0_audio_write_latency_fields(struct drm_encoder *encoder,
struct drm_display_mode *mode)
{
- struct amdgpu_device *adev = encoder->dev->dev_private;
+ struct drm_device *dev = encoder->dev;
+ struct amdgpu_device *adev = dev->dev_private;
struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
struct drm_connector *connector;
+ struct drm_connector_list_iter iter;
struct amdgpu_connector *amdgpu_connector = NULL;
int interlace = 0;
u32 tmp;
- list_for_each_entry(connector, &encoder->dev->mode_config.connector_list, head) {
+ drm_connector_list_iter_begin(dev, &iter);
+ drm_for_each_connector_iter(connector, &iter) {
if (connector->encoder == encoder) {
amdgpu_connector = to_amdgpu_connector(connector);
break;
}
}
+ drm_connector_list_iter_end(&iter);
if (!amdgpu_connector) {
DRM_ERROR("Couldn't find encoder's connector\n");
@@ -1163,21 +1172,25 @@ static void dce_v6_0_audio_write_latency_fields(struct drm_encoder *encoder,
static void dce_v6_0_audio_write_speaker_allocation(struct drm_encoder *encoder)
{
- struct amdgpu_device *adev = encoder->dev->dev_private;
+ struct drm_device *dev = encoder->dev;
+ struct amdgpu_device *adev = dev->dev_private;
struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
struct drm_connector *connector;
+ struct drm_connector_list_iter iter;
struct amdgpu_connector *amdgpu_connector = NULL;
u8 *sadb = NULL;
int sad_count;
u32 tmp;
- list_for_each_entry(connector, &encoder->dev->mode_config.connector_list, head) {
+ drm_connector_list_iter_begin(dev, &iter);
+ drm_for_each_connector_iter(connector, &iter) {
if (connector->encoder == encoder) {
amdgpu_connector = to_amdgpu_connector(connector);
break;
}
}
+ drm_connector_list_iter_end(&iter);
if (!amdgpu_connector) {
DRM_ERROR("Couldn't find encoder's connector\n");
@@ -1220,10 +1233,12 @@ static void dce_v6_0_audio_write_speaker_allocation(struct drm_encoder *encoder)
static void dce_v6_0_audio_write_sad_regs(struct drm_encoder *encoder)
{
- struct amdgpu_device *adev = encoder->dev->dev_private;
+ struct drm_device *dev = encoder->dev;
+ struct amdgpu_device *adev = dev->dev_private;
struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
struct drm_connector *connector;
+ struct drm_connector_list_iter iter;
struct amdgpu_connector *amdgpu_connector = NULL;
struct cea_sad *sads;
int i, sad_count;
@@ -1243,12 +1258,14 @@ static void dce_v6_0_audio_write_sad_regs(struct drm_encoder *encoder)
{ ixAZALIA_F0_CODEC_PIN_CONTROL_AUDIO_DESCRIPTOR13, HDMI_AUDIO_CODING_TYPE_WMA_PRO },
};
- list_for_each_entry(connector, &encoder->dev->mode_config.connector_list, head) {
+ drm_connector_list_iter_begin(dev, &iter);
+ drm_for_each_connector_iter(connector, &iter) {
if (connector->encoder == encoder) {
amdgpu_connector = to_amdgpu_connector(connector);
break;
}
}
+ drm_connector_list_iter_end(&iter);
if (!amdgpu_connector) {
DRM_ERROR("Couldn't find encoder's connector\n");
@@ -1634,6 +1651,7 @@ static void dce_v6_0_afmt_setmode(struct drm_encoder *encoder,
struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
struct drm_connector *connector;
+ struct drm_connector_list_iter iter;
struct amdgpu_connector *amdgpu_connector = NULL;
int em = amdgpu_atombios_encoder_get_encoder_mode(encoder);
int bpc = 8;
@@ -1641,12 +1659,14 @@ static void dce_v6_0_afmt_setmode(struct drm_encoder *encoder,
if (!dig || !dig->afmt)
return;
- list_for_each_entry(connector, &encoder->dev->mode_config.connector_list, head) {
+ drm_connector_list_iter_begin(dev, &iter);
+ drm_for_each_connector_iter(connector, &iter) {
if (connector->encoder == encoder) {
amdgpu_connector = to_amdgpu_connector(connector);
break;
}
}
+ drm_connector_list_iter_end(&iter);
if (!amdgpu_connector) {
DRM_ERROR("Couldn't find encoder's connector\n");
diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c
index 627ce4f82918..c8c532fbd6d2 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c
@@ -276,9 +276,11 @@ static void dce_v8_0_hpd_init(struct amdgpu_device *adev)
{
struct drm_device *dev = adev->ddev;
struct drm_connector *connector;
+ struct drm_connector_list_iter iter;
u32 tmp;
- list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
+ drm_connector_list_iter_begin(dev, &iter);
+ drm_for_each_connector_iter(connector, &iter) {
struct amdgpu_connector *amdgpu_connector = to_amdgpu_connector(connector);
if (amdgpu_connector->hpd.hpd >= adev->mode_info.num_hpd)
@@ -304,6 +306,7 @@ static void dce_v8_0_hpd_init(struct amdgpu_device *adev)
dce_v8_0_hpd_set_polarity(adev, amdgpu_connector->hpd.hpd);
amdgpu_irq_get(adev, &adev->hpd_irq, amdgpu_connector->hpd.hpd);
}
+ drm_connector_list_iter_end(&iter);
}
/**
@@ -318,9 +321,11 @@ static void dce_v8_0_hpd_fini(struct amdgpu_device *adev)
{
struct drm_device *dev = adev->ddev;
struct drm_connector *connector;
+ struct drm_connector_list_iter iter;
u32 tmp;
- list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
+ drm_connector_list_iter_begin(dev, &iter);
+ drm_for_each_connector_iter(connector, &iter) {
struct amdgpu_connector *amdgpu_connector = to_amdgpu_connector(connector);
if (amdgpu_connector->hpd.hpd >= adev->mode_info.num_hpd)
@@ -332,6 +337,7 @@ static void dce_v8_0_hpd_fini(struct amdgpu_device *adev)
amdgpu_irq_put(adev, &adev->hpd_irq, amdgpu_connector->hpd.hpd);
}
+ drm_connector_list_iter_end(&iter);
}
static u32 dce_v8_0_hpd_get_gpio_reg(struct amdgpu_device *adev)
@@ -1158,10 +1164,12 @@ static void dce_v8_0_afmt_audio_select_pin(struct drm_encoder *encoder)
static void dce_v8_0_audio_write_latency_fields(struct drm_encoder *encoder,
struct drm_display_mode *mode)
{
- struct amdgpu_device *adev = encoder->dev->dev_private;
+ struct drm_device *dev = encoder->dev;
+ struct amdgpu_device *adev = dev->dev_private;
struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
struct drm_connector *connector;
+ struct drm_connector_list_iter iter;
struct amdgpu_connector *amdgpu_connector = NULL;
u32 tmp = 0, offset;
@@ -1170,12 +1178,14 @@ static void dce_v8_0_audio_write_latency_fields(struct drm_encoder *encoder,
offset = dig->afmt->pin->offset;
- list_for_each_entry(connector, &encoder->dev->mode_config.connector_list, head) {
+ drm_connector_list_iter_begin(dev, &iter);
+ drm_for_each_connector_iter(connector, &iter) {
if (connector->encoder == encoder) {
amdgpu_connector = to_amdgpu_connector(connector);
break;
}
}
+ drm_connector_list_iter_end(&iter);
if (!amdgpu_connector) {
DRM_ERROR("Couldn't find encoder's connector\n");
@@ -1215,10 +1225,12 @@ static void dce_v8_0_audio_write_latency_fields(struct drm_encoder *encoder,
static void dce_v8_0_audio_write_speaker_allocation(struct drm_encoder *encoder)
{
- struct amdgpu_device *adev = encoder->dev->dev_private;
+ struct drm_device *dev = encoder->dev;
+ struct amdgpu_device *adev = dev->dev_private;
struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
struct drm_connector *connector;
+ struct drm_connector_list_iter iter;
struct amdgpu_connector *amdgpu_connector = NULL;
u32 offset, tmp;
u8 *sadb = NULL;
@@ -1229,12 +1241,14 @@ static void dce_v8_0_audio_write_speaker_allocation(struct drm_encoder *encoder)
offset = dig->afmt->pin->offset;
- list_for_each_entry(connector, &encoder->dev->mode_config.connector_list, head) {
+ drm_connector_list_iter_begin(dev, &iter);
+ drm_for_each_connector_iter(connector, &iter) {
if (connector->encoder == encoder) {
amdgpu_connector = to_amdgpu_connector(connector);
break;
}
}
+ drm_connector_list_iter_end(&iter);
if (!amdgpu_connector) {
DRM_ERROR("Couldn't find encoder's connector\n");
@@ -1264,11 +1278,13 @@ static void dce_v8_0_audio_write_speaker_allocation(struct drm_encoder *encoder)
static void dce_v8_0_audio_write_sad_regs(struct drm_encoder *encoder)
{
- struct amdgpu_device *adev = encoder->dev->dev_private;
+ struct drm_device *dev = encoder->dev;
+ struct amdgpu_device *adev = dev->dev_private;
struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
u32 offset;
struct drm_connector *connector;
+ struct drm_connector_list_iter iter;
struct amdgpu_connector *amdgpu_connector = NULL;
struct cea_sad *sads;
int i, sad_count;
@@ -1293,12 +1309,14 @@ static void dce_v8_0_audio_write_sad_regs(struct drm_encoder *encoder)
offset = dig->afmt->pin->offset;
- list_for_each_entry(connector, &encoder->dev->mode_config.connector_list, head) {
+ drm_connector_list_iter_begin(dev, &iter);
+ drm_for_each_connector_iter(connector, &iter) {
if (connector->encoder == encoder) {
amdgpu_connector = to_amdgpu_connector(connector);
break;
}
}
+ drm_connector_list_iter_end(&iter);
if (!amdgpu_connector) {
DRM_ERROR("Couldn't find encoder's connector\n");
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
index afe94b027112..f93ac8f44a58 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
@@ -248,7 +248,8 @@ static int gfx_v10_0_rlc_backdoor_autoload_enable(struct amdgpu_device *adev);
static int gfx_v10_0_wait_for_rlc_autoload_complete(struct amdgpu_device *adev);
static void gfx_v10_0_ring_emit_ce_meta(struct amdgpu_ring *ring, bool resume);
static void gfx_v10_0_ring_emit_de_meta(struct amdgpu_ring *ring, bool resume);
-static void gfx_v10_0_ring_emit_tmz(struct amdgpu_ring *ring, bool start);
+static void gfx_v10_0_ring_emit_tmz(struct amdgpu_ring *ring, bool start,
+ bool trusted);
static void gfx10_kiq_set_resources(struct amdgpu_ring *kiq_ring, uint64_t queue_mask)
{
@@ -1678,6 +1679,17 @@ static void gfx_v10_0_tcp_harvest(struct amdgpu_device *adev)
}
}
+static void gfx_v10_0_get_tcc_info(struct amdgpu_device *adev)
+{
+ /* TCCs are global (not instanced). */
+ uint32_t tcc_disable = RREG32_SOC15(GC, 0, mmCGTS_TCC_DISABLE) |
+ RREG32_SOC15(GC, 0, mmCGTS_USER_TCC_DISABLE);
+
+ adev->gfx.config.tcc_disabled_mask =
+ REG_GET_FIELD(tcc_disable, CGTS_TCC_DISABLE, TCC_DISABLE) |
+ (REG_GET_FIELD(tcc_disable, CGTS_TCC_DISABLE, HI_TCC_DISABLE) << 16);
+}
+
static void gfx_v10_0_constants_init(struct amdgpu_device *adev)
{
u32 tmp;
@@ -1689,6 +1701,7 @@ static void gfx_v10_0_constants_init(struct amdgpu_device *adev)
gfx_v10_0_setup_rb(adev);
gfx_v10_0_get_cu_info(adev, &adev->gfx.cu_info);
+ gfx_v10_0_get_tcc_info(adev);
adev->gfx.config.pa_sc_tile_steering_override =
gfx_v10_0_init_pa_sc_tile_steering_override(adev);
@@ -4528,7 +4541,9 @@ static void gfx_v10_0_ring_emit_sb(struct amdgpu_ring *ring)
amdgpu_ring_write(ring, 0);
}
-static void gfx_v10_0_ring_emit_cntxcntl(struct amdgpu_ring *ring, uint32_t flags)
+static void gfx_v10_0_ring_emit_cntxcntl(struct amdgpu_ring *ring,
+ uint32_t flags,
+ bool trusted)
{
uint32_t dw2 = 0;
@@ -4536,7 +4551,7 @@ static void gfx_v10_0_ring_emit_cntxcntl(struct amdgpu_ring *ring, uint32_t flag
gfx_v10_0_ring_emit_ce_meta(ring,
flags & AMDGPU_IB_PREEMPTED ? true : false);
- gfx_v10_0_ring_emit_tmz(ring, true);
+ gfx_v10_0_ring_emit_tmz(ring, true, false);
dw2 |= 0x80000000; /* set load_enable otherwise this package is just NOPs */
if (flags & AMDGPU_HAVE_CTX_SWITCH) {
@@ -4694,10 +4709,17 @@ static void gfx_v10_0_ring_emit_de_meta(struct amdgpu_ring *ring, bool resume)
sizeof(de_payload) >> 2);
}
-static void gfx_v10_0_ring_emit_tmz(struct amdgpu_ring *ring, bool start)
+static void gfx_v10_0_ring_emit_tmz(struct amdgpu_ring *ring, bool start,
+ bool trusted)
{
amdgpu_ring_write(ring, PACKET3(PACKET3_FRAME_CONTROL, 0));
- amdgpu_ring_write(ring, FRAME_CMD(start ? 0 : 1)); /* frame_end */
+ /*
+ * cmd = 0: frame begin
+ * cmd = 1: frame end
+ */
+ amdgpu_ring_write(ring,
+ ((ring->adev->tmz.enabled && trusted) ? FRAME_TMZ : 0)
+ | FRAME_CMD(start ? 0 : 1));
}
static void gfx_v10_0_ring_emit_rreg(struct amdgpu_ring *ring, uint32_t reg)
@@ -5263,15 +5285,12 @@ static void gfx_v10_0_set_rlc_funcs(struct amdgpu_device *adev)
static void gfx_v10_0_set_gds_init(struct amdgpu_device *adev)
{
- /* init asic gds info */
- switch (adev->asic_type) {
- case CHIP_NAVI10:
- default:
- adev->gds.gds_size = 0x10000;
- adev->gds.gds_compute_max_wave_id = 0x4ff;
- break;
- }
+ unsigned total_cu = adev->gfx.config.max_cu_per_sh *
+ adev->gfx.config.max_sh_per_se *
+ adev->gfx.config.max_shader_engines;
+ adev->gds.gds_size = 0x10000;
+ adev->gds.gds_compute_max_wave_id = total_cu * 32 - 1;
adev->gds.gws_size = 64;
adev->gds.oa_size = 16;
}
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c
index 8c27c305e692..b4af1b55f852 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c
@@ -2972,7 +2972,8 @@ static uint64_t gfx_v6_0_get_gpu_clock_counter(struct amdgpu_device *adev)
return clock;
}
-static void gfx_v6_ring_emit_cntxcntl(struct amdgpu_ring *ring, uint32_t flags)
+static void gfx_v6_ring_emit_cntxcntl(struct amdgpu_ring *ring, uint32_t flags,
+ bool trusted)
{
if (flags & AMDGPU_HAVE_CTX_SWITCH)
gfx_v6_0_ring_emit_vgt_flush(ring);
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
index 48796b6824cf..c08f5c53dcb4 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
@@ -2309,7 +2309,8 @@ static void gfx_v7_0_ring_emit_ib_compute(struct amdgpu_ring *ring,
amdgpu_ring_write(ring, control);
}
-static void gfx_v7_ring_emit_cntxcntl(struct amdgpu_ring *ring, uint32_t flags)
+static void gfx_v7_ring_emit_cntxcntl(struct amdgpu_ring *ring, uint32_t flags,
+ bool trusted)
{
uint32_t dw2 = 0;
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
index 98e5aa8d6ac1..d3a23fdcc6c0 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
@@ -6393,7 +6393,8 @@ static void gfx_v8_ring_emit_sb(struct amdgpu_ring *ring)
amdgpu_ring_write(ring, 0);
}
-static void gfx_v8_ring_emit_cntxcntl(struct amdgpu_ring *ring, uint32_t flags)
+static void gfx_v8_ring_emit_cntxcntl(struct amdgpu_ring *ring, uint32_t flags,
+ bool trusted)
{
uint32_t dw2 = 0;
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
index 5f7956004627..16043b824f97 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
@@ -127,6 +127,18 @@ MODULE_FIRMWARE("amdgpu/renoir_rlc.bin");
#define mmTCP_CHAN_STEER_5_ARCT 0x0b0c
#define mmTCP_CHAN_STEER_5_ARCT_BASE_IDX 0
+struct ras_gfx_subblock_reg {
+ const char *name;
+ uint32_t hwip;
+ uint32_t inst;
+ uint32_t seg;
+ uint32_t reg_offset;
+ uint32_t sec_count_mask;
+ uint32_t sec_count_shift;
+ uint32_t ded_count_mask;
+ uint32_t ded_count_shift;
+};
+
enum ta_ras_gfx_subblock {
/*CPC*/
TA_RAS_BLOCK__GFX_CPC_INDEX_START = 0,
@@ -3976,6 +3988,7 @@ static const struct soc15_reg_entry sec_ded_counter_registers[] = {
{ SOC15_REG_ENTRY(GC, 0, mmTCC_EDC_CNT), 0, 1, 16},
{ SOC15_REG_ENTRY(GC, 0, mmTCP_ATC_EDC_GATCL1_CNT), 0, 4, 16},
{ SOC15_REG_ENTRY(GC, 0, mmTCP_EDC_CNT), 0, 4, 16},
+ { SOC15_REG_ENTRY(GC, 0, mmTCP_EDC_CNT_NEW), 0, 4, 16},
{ SOC15_REG_ENTRY(GC, 0, mmTD_EDC_CNT), 0, 4, 16},
{ SOC15_REG_ENTRY(GC, 0, mmSQC_EDC_CNT2), 0, 4, 6},
{ SOC15_REG_ENTRY(GC, 0, mmSQ_EDC_CNT), 0, 4, 16},
@@ -4272,16 +4285,14 @@ static void gfx_v9_0_update_gfx_cg_power_gating(struct amdgpu_device *adev,
{
amdgpu_gfx_rlc_enter_safe_mode(adev);
- if (is_support_sw_smu(adev) && !enable)
- smu_set_gfx_cgpg(&adev->smu, enable);
-
if ((adev->pg_flags & AMD_PG_SUPPORT_GFX_PG) && enable) {
gfx_v9_0_enable_gfx_cg_power_gating(adev, true);
if (adev->pg_flags & AMD_PG_SUPPORT_GFX_PIPELINE)
gfx_v9_0_enable_gfx_pipeline_powergating(adev, true);
} else {
gfx_v9_0_enable_gfx_cg_power_gating(adev, false);
- gfx_v9_0_enable_gfx_pipeline_powergating(adev, false);
+ if (adev->pg_flags & AMD_PG_SUPPORT_GFX_PIPELINE)
+ gfx_v9_0_enable_gfx_pipeline_powergating(adev, false);
}
amdgpu_gfx_rlc_exit_safe_mode(adev);
@@ -4550,8 +4561,6 @@ static int gfx_v9_0_set_powergating_state(void *handle,
gfx_v9_0_enable_cp_power_gating(adev, false);
/* update gfx cgpg state */
- if (is_support_sw_smu(adev) && enable)
- smu_set_gfx_cgpg(&adev->smu, enable);
gfx_v9_0_update_gfx_cg_power_gating(adev, enable);
/* update mgcg state */
@@ -5036,20 +5045,28 @@ static void gfx_v9_0_ring_emit_de_meta(struct amdgpu_ring *ring)
amdgpu_ring_write_multiple(ring, (void *)&de_payload, sizeof(de_payload) >> 2);
}
-static void gfx_v9_0_ring_emit_tmz(struct amdgpu_ring *ring, bool start)
+static void gfx_v9_0_ring_emit_tmz(struct amdgpu_ring *ring, bool start,
+ bool trusted)
{
amdgpu_ring_write(ring, PACKET3(PACKET3_FRAME_CONTROL, 0));
- amdgpu_ring_write(ring, FRAME_CMD(start ? 0 : 1)); /* frame_end */
+ /*
+ * cmd = 0: frame begin
+ * cmd = 1: frame end
+ */
+ amdgpu_ring_write(ring,
+ ((ring->adev->tmz.enabled && trusted) ? FRAME_TMZ : 0)
+ | FRAME_CMD(start ? 0 : 1));
}
-static void gfx_v9_ring_emit_cntxcntl(struct amdgpu_ring *ring, uint32_t flags)
+static void gfx_v9_ring_emit_cntxcntl(struct amdgpu_ring *ring, uint32_t flags,
+ bool trusted)
{
uint32_t dw2 = 0;
if (amdgpu_sriov_vf(ring->adev))
gfx_v9_0_ring_emit_ce_meta(ring);
- gfx_v9_0_ring_emit_tmz(ring, true);
+ gfx_v9_0_ring_emit_tmz(ring, true, trusted);
dw2 |= 0x80000000; /* set load_enable otherwise this package is just NOPs */
if (flags & AMDGPU_HAVE_CTX_SWITCH) {
@@ -5435,301 +5452,446 @@ static int gfx_v9_0_priv_inst_irq(struct amdgpu_device *adev,
return 0;
}
-static const struct {
- const char *name;
- uint32_t ip;
- uint32_t inst;
- uint32_t seg;
- uint32_t reg_offset;
- uint32_t per_se_instance;
- int32_t num_instance;
- uint32_t sec_count_mask;
- uint32_t ded_count_mask;
-} gfx_ras_edc_regs[] = {
- { "CPC_SCRATCH", SOC15_REG_ENTRY(GC, 0, mmCPC_EDC_SCRATCH_CNT), 0, 1,
- REG_FIELD_MASK(CPC_EDC_SCRATCH_CNT, SEC_COUNT),
- REG_FIELD_MASK(CPC_EDC_SCRATCH_CNT, DED_COUNT) },
- { "CPC_UCODE", SOC15_REG_ENTRY(GC, 0, mmCPC_EDC_UCODE_CNT), 0, 1,
- REG_FIELD_MASK(CPC_EDC_UCODE_CNT, SEC_COUNT),
- REG_FIELD_MASK(CPC_EDC_UCODE_CNT, DED_COUNT) },
- { "CPF_ROQ_ME1", SOC15_REG_ENTRY(GC, 0, mmCPF_EDC_ROQ_CNT), 0, 1,
- REG_FIELD_MASK(CPF_EDC_ROQ_CNT, COUNT_ME1), 0 },
- { "CPF_ROQ_ME2", SOC15_REG_ENTRY(GC, 0, mmCPF_EDC_ROQ_CNT), 0, 1,
- REG_FIELD_MASK(CPF_EDC_ROQ_CNT, COUNT_ME2), 0 },
- { "CPF_TAG", SOC15_REG_ENTRY(GC, 0, mmCPF_EDC_TAG_CNT), 0, 1,
- REG_FIELD_MASK(CPF_EDC_TAG_CNT, SEC_COUNT),
- REG_FIELD_MASK(CPF_EDC_TAG_CNT, DED_COUNT) },
- { "CPG_DMA_ROQ", SOC15_REG_ENTRY(GC, 0, mmCPG_EDC_DMA_CNT), 0, 1,
- REG_FIELD_MASK(CPG_EDC_DMA_CNT, ROQ_COUNT), 0 },
- { "CPG_DMA_TAG", SOC15_REG_ENTRY(GC, 0, mmCPG_EDC_DMA_CNT), 0, 1,
- REG_FIELD_MASK(CPG_EDC_DMA_CNT, TAG_SEC_COUNT),
- REG_FIELD_MASK(CPG_EDC_DMA_CNT, TAG_DED_COUNT) },
- { "CPG_TAG", SOC15_REG_ENTRY(GC, 0, mmCPG_EDC_TAG_CNT), 0, 1,
- REG_FIELD_MASK(CPG_EDC_TAG_CNT, SEC_COUNT),
- REG_FIELD_MASK(CPG_EDC_TAG_CNT, DED_COUNT) },
- { "DC_CSINVOC", SOC15_REG_ENTRY(GC, 0, mmDC_EDC_CSINVOC_CNT), 0, 1,
- REG_FIELD_MASK(DC_EDC_CSINVOC_CNT, COUNT_ME1), 0 },
- { "DC_RESTORE", SOC15_REG_ENTRY(GC, 0, mmDC_EDC_RESTORE_CNT), 0, 1,
- REG_FIELD_MASK(DC_EDC_RESTORE_CNT, COUNT_ME1), 0 },
- { "DC_STATE", SOC15_REG_ENTRY(GC, 0, mmDC_EDC_STATE_CNT), 0, 1,
- REG_FIELD_MASK(DC_EDC_STATE_CNT, COUNT_ME1), 0 },
- { "GDS_MEM", SOC15_REG_ENTRY(GC, 0, mmGDS_EDC_CNT), 0, 1,
- REG_FIELD_MASK(GDS_EDC_CNT, GDS_MEM_SEC),
- REG_FIELD_MASK(GDS_EDC_CNT, GDS_MEM_DED) },
- { "GDS_INPUT_QUEUE", SOC15_REG_ENTRY(GC, 0, mmGDS_EDC_CNT), 0, 1,
- REG_FIELD_MASK(GDS_EDC_CNT, GDS_INPUT_QUEUE_SED), 0 },
+
+static const struct ras_gfx_subblock_reg ras_subblock_regs[] = {
+ { "CPC_SCRATCH", SOC15_REG_ENTRY(GC, 0, mmCPC_EDC_SCRATCH_CNT),
+ SOC15_REG_FIELD(CPC_EDC_SCRATCH_CNT, SEC_COUNT),
+ SOC15_REG_FIELD(CPC_EDC_SCRATCH_CNT, DED_COUNT)
+ },
+ { "CPC_UCODE", SOC15_REG_ENTRY(GC, 0, mmCPC_EDC_UCODE_CNT),
+ SOC15_REG_FIELD(CPC_EDC_UCODE_CNT, SEC_COUNT),
+ SOC15_REG_FIELD(CPC_EDC_UCODE_CNT, DED_COUNT)
+ },
+ { "CPF_ROQ_ME1", SOC15_REG_ENTRY(GC, 0, mmCPF_EDC_ROQ_CNT),
+ SOC15_REG_FIELD(CPF_EDC_ROQ_CNT, COUNT_ME1),
+ 0, 0
+ },
+ { "CPF_ROQ_ME2", SOC15_REG_ENTRY(GC, 0, mmCPF_EDC_ROQ_CNT),
+ SOC15_REG_FIELD(CPF_EDC_ROQ_CNT, COUNT_ME2),
+ 0, 0
+ },
+ { "CPF_TAG", SOC15_REG_ENTRY(GC, 0, mmCPF_EDC_TAG_CNT),
+ SOC15_REG_FIELD(CPF_EDC_TAG_CNT, SEC_COUNT),
+ SOC15_REG_FIELD(CPF_EDC_TAG_CNT, DED_COUNT)
+ },
+ { "CPG_DMA_ROQ", SOC15_REG_ENTRY(GC, 0, mmCPG_EDC_DMA_CNT),
+ SOC15_REG_FIELD(CPG_EDC_DMA_CNT, ROQ_COUNT),
+ 0, 0
+ },
+ { "CPG_DMA_TAG", SOC15_REG_ENTRY(GC, 0, mmCPG_EDC_DMA_CNT),
+ SOC15_REG_FIELD(CPG_EDC_DMA_CNT, TAG_SEC_COUNT),
+ SOC15_REG_FIELD(CPG_EDC_DMA_CNT, TAG_DED_COUNT)
+ },
+ { "CPG_TAG", SOC15_REG_ENTRY(GC, 0, mmCPG_EDC_TAG_CNT),
+ SOC15_REG_FIELD(CPG_EDC_TAG_CNT, SEC_COUNT),
+ SOC15_REG_FIELD(CPG_EDC_TAG_CNT, DED_COUNT)
+ },
+ { "DC_CSINVOC", SOC15_REG_ENTRY(GC, 0, mmDC_EDC_CSINVOC_CNT),
+ SOC15_REG_FIELD(DC_EDC_CSINVOC_CNT, COUNT_ME1),
+ 0, 0
+ },
+ { "DC_RESTORE", SOC15_REG_ENTRY(GC, 0, mmDC_EDC_RESTORE_CNT),
+ SOC15_REG_FIELD(DC_EDC_RESTORE_CNT, COUNT_ME1),
+ 0, 0
+ },
+ { "DC_STATE", SOC15_REG_ENTRY(GC, 0, mmDC_EDC_STATE_CNT),
+ SOC15_REG_FIELD(DC_EDC_STATE_CNT, COUNT_ME1),
+ 0, 0
+ },
+ { "GDS_MEM", SOC15_REG_ENTRY(GC, 0, mmGDS_EDC_CNT),
+ SOC15_REG_FIELD(GDS_EDC_CNT, GDS_MEM_SEC),
+ SOC15_REG_FIELD(GDS_EDC_CNT, GDS_MEM_DED)
+ },
+ { "GDS_INPUT_QUEUE", SOC15_REG_ENTRY(GC, 0, mmGDS_EDC_CNT),
+ SOC15_REG_FIELD(GDS_EDC_CNT, GDS_INPUT_QUEUE_SED),
+ 0, 0
+ },
{ "GDS_ME0_CS_PIPE_MEM", SOC15_REG_ENTRY(GC, 0, mmGDS_EDC_OA_PHY_CNT),
- 0, 1, REG_FIELD_MASK(GDS_EDC_OA_PHY_CNT, ME0_CS_PIPE_MEM_SEC),
- REG_FIELD_MASK(GDS_EDC_OA_PHY_CNT, ME0_CS_PIPE_MEM_DED) },
+ SOC15_REG_FIELD(GDS_EDC_OA_PHY_CNT, ME0_CS_PIPE_MEM_SEC),
+ SOC15_REG_FIELD(GDS_EDC_OA_PHY_CNT, ME0_CS_PIPE_MEM_DED)
+ },
{ "GDS_OA_PHY_PHY_CMD_RAM_MEM",
- SOC15_REG_ENTRY(GC, 0, mmGDS_EDC_OA_PHY_CNT), 0, 1,
- REG_FIELD_MASK(GDS_EDC_OA_PHY_CNT, PHY_CMD_RAM_MEM_SEC),
- REG_FIELD_MASK(GDS_EDC_OA_PHY_CNT, PHY_CMD_RAM_MEM_DED) },
+ SOC15_REG_ENTRY(GC, 0, mmGDS_EDC_OA_PHY_CNT),
+ SOC15_REG_FIELD(GDS_EDC_OA_PHY_CNT, PHY_CMD_RAM_MEM_SEC),
+ SOC15_REG_FIELD(GDS_EDC_OA_PHY_CNT, PHY_CMD_RAM_MEM_DED)
+ },
{ "GDS_OA_PHY_PHY_DATA_RAM_MEM",
- SOC15_REG_ENTRY(GC, 0, mmGDS_EDC_OA_PHY_CNT), 0, 1,
- REG_FIELD_MASK(GDS_EDC_OA_PHY_CNT, PHY_DATA_RAM_MEM_SED), 0 },
+ SOC15_REG_ENTRY(GC, 0, mmGDS_EDC_OA_PHY_CNT),
+ SOC15_REG_FIELD(GDS_EDC_OA_PHY_CNT, PHY_DATA_RAM_MEM_SED),
+ 0, 0
+ },
{ "GDS_OA_PIPE_ME1_PIPE0_PIPE_MEM",
- SOC15_REG_ENTRY(GC, 0, mmGDS_EDC_OA_PIPE_CNT), 0, 1,
- REG_FIELD_MASK(GDS_EDC_OA_PIPE_CNT, ME1_PIPE0_PIPE_MEM_SEC),
- REG_FIELD_MASK(GDS_EDC_OA_PIPE_CNT, ME1_PIPE0_PIPE_MEM_DED) },
+ SOC15_REG_ENTRY(GC, 0, mmGDS_EDC_OA_PIPE_CNT),
+ SOC15_REG_FIELD(GDS_EDC_OA_PIPE_CNT, ME1_PIPE0_PIPE_MEM_SEC),
+ SOC15_REG_FIELD(GDS_EDC_OA_PIPE_CNT, ME1_PIPE0_PIPE_MEM_DED)
+ },
{ "GDS_OA_PIPE_ME1_PIPE1_PIPE_MEM",
- SOC15_REG_ENTRY(GC, 0, mmGDS_EDC_OA_PIPE_CNT), 0, 1,
- REG_FIELD_MASK(GDS_EDC_OA_PIPE_CNT, ME1_PIPE1_PIPE_MEM_SEC),
- REG_FIELD_MASK(GDS_EDC_OA_PIPE_CNT, ME1_PIPE1_PIPE_MEM_DED) },
+ SOC15_REG_ENTRY(GC, 0, mmGDS_EDC_OA_PIPE_CNT),
+ SOC15_REG_FIELD(GDS_EDC_OA_PIPE_CNT, ME1_PIPE1_PIPE_MEM_SEC),
+ SOC15_REG_FIELD(GDS_EDC_OA_PIPE_CNT, ME1_PIPE1_PIPE_MEM_DED)
+ },
{ "GDS_OA_PIPE_ME1_PIPE2_PIPE_MEM",
- SOC15_REG_ENTRY(GC, 0, mmGDS_EDC_OA_PIPE_CNT), 0, 1,
- REG_FIELD_MASK(GDS_EDC_OA_PIPE_CNT, ME1_PIPE2_PIPE_MEM_SEC),
- REG_FIELD_MASK(GDS_EDC_OA_PIPE_CNT, ME1_PIPE2_PIPE_MEM_DED) },
+ SOC15_REG_ENTRY(GC, 0, mmGDS_EDC_OA_PIPE_CNT),
+ SOC15_REG_FIELD(GDS_EDC_OA_PIPE_CNT, ME1_PIPE2_PIPE_MEM_SEC),
+ SOC15_REG_FIELD(GDS_EDC_OA_PIPE_CNT, ME1_PIPE2_PIPE_MEM_DED)
+ },
{ "GDS_OA_PIPE_ME1_PIPE3_PIPE_MEM",
- SOC15_REG_ENTRY(GC, 0, mmGDS_EDC_OA_PIPE_CNT), 0, 1,
- REG_FIELD_MASK(GDS_EDC_OA_PIPE_CNT, ME1_PIPE3_PIPE_MEM_SEC),
- REG_FIELD_MASK(GDS_EDC_OA_PIPE_CNT, ME1_PIPE3_PIPE_MEM_DED) },
- { "SPI_SR_MEM", SOC15_REG_ENTRY(GC, 0, mmSPI_EDC_CNT), 1, 1,
- REG_FIELD_MASK(SPI_EDC_CNT, SPI_SR_MEM_SED_COUNT), 0 },
- { "TA_FS_DFIFO", SOC15_REG_ENTRY(GC, 0, mmTA_EDC_CNT), 1, 16,
- REG_FIELD_MASK(TA_EDC_CNT, TA_FS_DFIFO_SEC_COUNT),
- REG_FIELD_MASK(TA_EDC_CNT, TA_FS_DFIFO_DED_COUNT) },
- { "TA_FS_AFIFO", SOC15_REG_ENTRY(GC, 0, mmTA_EDC_CNT), 1, 16,
- REG_FIELD_MASK(TA_EDC_CNT, TA_FS_AFIFO_SED_COUNT), 0 },
- { "TA_FL_LFIFO", SOC15_REG_ENTRY(GC, 0, mmTA_EDC_CNT), 1, 16,
- REG_FIELD_MASK(TA_EDC_CNT, TA_FL_LFIFO_SED_COUNT), 0 },
- { "TA_FX_LFIFO", SOC15_REG_ENTRY(GC, 0, mmTA_EDC_CNT), 1, 16,
- REG_FIELD_MASK(TA_EDC_CNT, TA_FX_LFIFO_SED_COUNT), 0 },
- { "TA_FS_CFIFO", SOC15_REG_ENTRY(GC, 0, mmTA_EDC_CNT), 1, 16,
- REG_FIELD_MASK(TA_EDC_CNT, TA_FS_CFIFO_SED_COUNT), 0 },
- { "TCA_HOLE_FIFO", SOC15_REG_ENTRY(GC, 0, mmTCA_EDC_CNT), 0, 2,
- REG_FIELD_MASK(TCA_EDC_CNT, HOLE_FIFO_SED_COUNT), 0 },
- { "TCA_REQ_FIFO", SOC15_REG_ENTRY(GC, 0, mmTCA_EDC_CNT), 0, 2,
- REG_FIELD_MASK(TCA_EDC_CNT, REQ_FIFO_SED_COUNT), 0 },
- { "TCC_CACHE_DATA", SOC15_REG_ENTRY(GC, 0, mmTCC_EDC_CNT), 0, 16,
- REG_FIELD_MASK(TCC_EDC_CNT, CACHE_DATA_SEC_COUNT),
- REG_FIELD_MASK(TCC_EDC_CNT, CACHE_DATA_DED_COUNT) },
- { "TCC_CACHE_DIRTY", SOC15_REG_ENTRY(GC, 0, mmTCC_EDC_CNT), 0, 16,
- REG_FIELD_MASK(TCC_EDC_CNT, CACHE_DIRTY_SEC_COUNT),
- REG_FIELD_MASK(TCC_EDC_CNT, CACHE_DIRTY_DED_COUNT) },
- { "TCC_HIGH_RATE_TAG", SOC15_REG_ENTRY(GC, 0, mmTCC_EDC_CNT), 0, 16,
- REG_FIELD_MASK(TCC_EDC_CNT, HIGH_RATE_TAG_SEC_COUNT),
- REG_FIELD_MASK(TCC_EDC_CNT, HIGH_RATE_TAG_DED_COUNT) },
- { "TCC_LOW_RATE_TAG", SOC15_REG_ENTRY(GC, 0, mmTCC_EDC_CNT), 0, 16,
- REG_FIELD_MASK(TCC_EDC_CNT, LOW_RATE_TAG_SEC_COUNT),
- REG_FIELD_MASK(TCC_EDC_CNT, LOW_RATE_TAG_DED_COUNT) },
- { "TCC_SRC_FIFO", SOC15_REG_ENTRY(GC, 0, mmTCC_EDC_CNT), 0, 16,
- REG_FIELD_MASK(TCC_EDC_CNT, SRC_FIFO_SEC_COUNT),
- REG_FIELD_MASK(TCC_EDC_CNT, SRC_FIFO_DED_COUNT) },
- { "TCC_IN_USE_DEC", SOC15_REG_ENTRY(GC, 0, mmTCC_EDC_CNT), 0, 16,
- REG_FIELD_MASK(TCC_EDC_CNT, IN_USE_DEC_SED_COUNT), 0 },
- { "TCC_IN_USE_TRANSFER", SOC15_REG_ENTRY(GC, 0, mmTCC_EDC_CNT), 0, 16,
- REG_FIELD_MASK(TCC_EDC_CNT, IN_USE_TRANSFER_SED_COUNT), 0 },
- { "TCC_LATENCY_FIFO", SOC15_REG_ENTRY(GC, 0, mmTCC_EDC_CNT), 0, 16,
- REG_FIELD_MASK(TCC_EDC_CNT, LATENCY_FIFO_SED_COUNT), 0 },
- { "TCC_RETURN_DATA", SOC15_REG_ENTRY(GC, 0, mmTCC_EDC_CNT), 0, 16,
- REG_FIELD_MASK(TCC_EDC_CNT, RETURN_DATA_SED_COUNT), 0 },
- { "TCC_RETURN_CONTROL", SOC15_REG_ENTRY(GC, 0, mmTCC_EDC_CNT), 0, 16,
- REG_FIELD_MASK(TCC_EDC_CNT, RETURN_CONTROL_SED_COUNT), 0 },
- { "TCC_UC_ATOMIC_FIFO", SOC15_REG_ENTRY(GC, 0, mmTCC_EDC_CNT), 0, 16,
- REG_FIELD_MASK(TCC_EDC_CNT, UC_ATOMIC_FIFO_SED_COUNT), 0 },
- { "TCC_WRITE_RETURN", SOC15_REG_ENTRY(GC, 0, mmTCC_EDC_CNT2), 0, 16,
- REG_FIELD_MASK(TCC_EDC_CNT2, WRITE_RETURN_SED_COUNT), 0 },
- { "TCC_WRITE_CACHE_READ", SOC15_REG_ENTRY(GC, 0, mmTCC_EDC_CNT2), 0, 16,
- REG_FIELD_MASK(TCC_EDC_CNT2, WRITE_CACHE_READ_SED_COUNT), 0 },
- { "TCC_SRC_FIFO_NEXT_RAM", SOC15_REG_ENTRY(GC, 0, mmTCC_EDC_CNT2), 0,
- 16, REG_FIELD_MASK(TCC_EDC_CNT2, SRC_FIFO_NEXT_RAM_SED_COUNT), 0 },
+ SOC15_REG_ENTRY(GC, 0, mmGDS_EDC_OA_PIPE_CNT),
+ SOC15_REG_FIELD(GDS_EDC_OA_PIPE_CNT, ME1_PIPE3_PIPE_MEM_SEC),
+ SOC15_REG_FIELD(GDS_EDC_OA_PIPE_CNT, ME1_PIPE3_PIPE_MEM_DED)
+ },
+ { "SPI_SR_MEM", SOC15_REG_ENTRY(GC, 0, mmSPI_EDC_CNT),
+ SOC15_REG_FIELD(SPI_EDC_CNT, SPI_SR_MEM_SED_COUNT),
+ 0, 0
+ },
+ { "TA_FS_DFIFO", SOC15_REG_ENTRY(GC, 0, mmTA_EDC_CNT),
+ SOC15_REG_FIELD(TA_EDC_CNT, TA_FS_DFIFO_SEC_COUNT),
+ SOC15_REG_FIELD(TA_EDC_CNT, TA_FS_DFIFO_DED_COUNT)
+ },
+ { "TA_FS_AFIFO", SOC15_REG_ENTRY(GC, 0, mmTA_EDC_CNT),
+ SOC15_REG_FIELD(TA_EDC_CNT, TA_FS_AFIFO_SED_COUNT),
+ 0, 0
+ },
+ { "TA_FL_LFIFO", SOC15_REG_ENTRY(GC, 0, mmTA_EDC_CNT),
+ SOC15_REG_FIELD(TA_EDC_CNT, TA_FL_LFIFO_SED_COUNT),
+ 0, 0
+ },
+ { "TA_FX_LFIFO", SOC15_REG_ENTRY(GC, 0, mmTA_EDC_CNT),
+ SOC15_REG_FIELD(TA_EDC_CNT, TA_FX_LFIFO_SED_COUNT),
+ 0, 0
+ },
+ { "TA_FS_CFIFO", SOC15_REG_ENTRY(GC, 0, mmTA_EDC_CNT),
+ SOC15_REG_FIELD(TA_EDC_CNT, TA_FS_CFIFO_SED_COUNT),
+ 0, 0
+ },
+ { "TCA_HOLE_FIFO", SOC15_REG_ENTRY(GC, 0, mmTCA_EDC_CNT),
+ SOC15_REG_FIELD(TCA_EDC_CNT, HOLE_FIFO_SED_COUNT),
+ 0, 0
+ },
+ { "TCA_REQ_FIFO", SOC15_REG_ENTRY(GC, 0, mmTCA_EDC_CNT),
+ SOC15_REG_FIELD(TCA_EDC_CNT, REQ_FIFO_SED_COUNT),
+ 0, 0
+ },
+ { "TCC_CACHE_DATA", SOC15_REG_ENTRY(GC, 0, mmTCC_EDC_CNT),
+ SOC15_REG_FIELD(TCC_EDC_CNT, CACHE_DATA_SEC_COUNT),
+ SOC15_REG_FIELD(TCC_EDC_CNT, CACHE_DATA_DED_COUNT)
+ },
+ { "TCC_CACHE_DIRTY", SOC15_REG_ENTRY(GC, 0, mmTCC_EDC_CNT),
+ SOC15_REG_FIELD(TCC_EDC_CNT, CACHE_DIRTY_SEC_COUNT),
+ SOC15_REG_FIELD(TCC_EDC_CNT, CACHE_DIRTY_DED_COUNT)
+ },
+ { "TCC_HIGH_RATE_TAG", SOC15_REG_ENTRY(GC, 0, mmTCC_EDC_CNT),
+ SOC15_REG_FIELD(TCC_EDC_CNT, HIGH_RATE_TAG_SEC_COUNT),
+ SOC15_REG_FIELD(TCC_EDC_CNT, HIGH_RATE_TAG_DED_COUNT)
+ },
+ { "TCC_LOW_RATE_TAG", SOC15_REG_ENTRY(GC, 0, mmTCC_EDC_CNT),
+ SOC15_REG_FIELD(TCC_EDC_CNT, LOW_RATE_TAG_SEC_COUNT),
+ SOC15_REG_FIELD(TCC_EDC_CNT, LOW_RATE_TAG_DED_COUNT)
+ },
+ { "TCC_SRC_FIFO", SOC15_REG_ENTRY(GC, 0, mmTCC_EDC_CNT),
+ SOC15_REG_FIELD(TCC_EDC_CNT, SRC_FIFO_SEC_COUNT),
+ SOC15_REG_FIELD(TCC_EDC_CNT, SRC_FIFO_DED_COUNT)
+ },
+ { "TCC_IN_USE_DEC", SOC15_REG_ENTRY(GC, 0, mmTCC_EDC_CNT),
+ SOC15_REG_FIELD(TCC_EDC_CNT, IN_USE_DEC_SED_COUNT),
+ 0, 0
+ },
+ { "TCC_IN_USE_TRANSFER", SOC15_REG_ENTRY(GC, 0, mmTCC_EDC_CNT),
+ SOC15_REG_FIELD(TCC_EDC_CNT, IN_USE_TRANSFER_SED_COUNT),
+ 0, 0
+ },
+ { "TCC_LATENCY_FIFO", SOC15_REG_ENTRY(GC, 0, mmTCC_EDC_CNT),
+ SOC15_REG_FIELD(TCC_EDC_CNT, LATENCY_FIFO_SED_COUNT),
+ 0, 0
+ },
+ { "TCC_RETURN_DATA", SOC15_REG_ENTRY(GC, 0, mmTCC_EDC_CNT),
+ SOC15_REG_FIELD(TCC_EDC_CNT, RETURN_DATA_SED_COUNT),
+ 0, 0
+ },
+ { "TCC_RETURN_CONTROL", SOC15_REG_ENTRY(GC, 0, mmTCC_EDC_CNT),
+ SOC15_REG_FIELD(TCC_EDC_CNT, RETURN_CONTROL_SED_COUNT),
+ 0, 0
+ },
+ { "TCC_UC_ATOMIC_FIFO", SOC15_REG_ENTRY(GC, 0, mmTCC_EDC_CNT),
+ SOC15_REG_FIELD(TCC_EDC_CNT, UC_ATOMIC_FIFO_SED_COUNT),
+ 0, 0
+ },
+ { "TCC_WRITE_RETURN", SOC15_REG_ENTRY(GC, 0, mmTCC_EDC_CNT2),
+ SOC15_REG_FIELD(TCC_EDC_CNT2, WRITE_RETURN_SED_COUNT),
+ 0, 0
+ },
+ { "TCC_WRITE_CACHE_READ", SOC15_REG_ENTRY(GC, 0, mmTCC_EDC_CNT2),
+ SOC15_REG_FIELD(TCC_EDC_CNT2, WRITE_CACHE_READ_SED_COUNT),
+ 0, 0
+ },
+ { "TCC_SRC_FIFO_NEXT_RAM", SOC15_REG_ENTRY(GC, 0, mmTCC_EDC_CNT2),
+ SOC15_REG_FIELD(TCC_EDC_CNT2, SRC_FIFO_NEXT_RAM_SED_COUNT),
+ 0, 0
+ },
{ "TCC_LATENCY_FIFO_NEXT_RAM", SOC15_REG_ENTRY(GC, 0, mmTCC_EDC_CNT2),
- 0, 16, REG_FIELD_MASK(TCC_EDC_CNT2, LATENCY_FIFO_NEXT_RAM_SED_COUNT),
- 0 },
- { "TCC_CACHE_TAG_PROBE_FIFO", SOC15_REG_ENTRY(GC, 0, mmTCC_EDC_CNT2), 0,
- 16, REG_FIELD_MASK(TCC_EDC_CNT2, CACHE_TAG_PROBE_FIFO_SED_COUNT), 0 },
+ SOC15_REG_FIELD(TCC_EDC_CNT2, LATENCY_FIFO_NEXT_RAM_SED_COUNT),
+ 0, 0
+ },
+ { "TCC_CACHE_TAG_PROBE_FIFO", SOC15_REG_ENTRY(GC, 0, mmTCC_EDC_CNT2),
+ SOC15_REG_FIELD(TCC_EDC_CNT2, CACHE_TAG_PROBE_FIFO_SED_COUNT),
+ 0, 0
+ },
{ "TCC_WRRET_TAG_WRITE_RETURN", SOC15_REG_ENTRY(GC, 0, mmTCC_EDC_CNT2),
- 0, 16, REG_FIELD_MASK(TCC_EDC_CNT2, WRRET_TAG_WRITE_RETURN_SED_COUNT),
- 0 },
- { "TCC_ATOMIC_RETURN_BUFFER", SOC15_REG_ENTRY(GC, 0, mmTCC_EDC_CNT2), 0,
- 16, REG_FIELD_MASK(TCC_EDC_CNT2, ATOMIC_RETURN_BUFFER_SED_COUNT), 0 },
- { "TCI_WRITE_RAM", SOC15_REG_ENTRY(GC, 0, mmTCI_EDC_CNT), 0, 72,
- REG_FIELD_MASK(TCI_EDC_CNT, WRITE_RAM_SED_COUNT), 0 },
- { "TCP_CACHE_RAM", SOC15_REG_ENTRY(GC, 0, mmTCP_EDC_CNT_NEW), 1, 16,
- REG_FIELD_MASK(TCP_EDC_CNT_NEW, CACHE_RAM_SEC_COUNT),
- REG_FIELD_MASK(TCP_EDC_CNT_NEW, CACHE_RAM_DED_COUNT) },
- { "TCP_LFIFO_RAM", SOC15_REG_ENTRY(GC, 0, mmTCP_EDC_CNT_NEW), 1, 16,
- REG_FIELD_MASK(TCP_EDC_CNT_NEW, LFIFO_RAM_SEC_COUNT),
- REG_FIELD_MASK(TCP_EDC_CNT_NEW, LFIFO_RAM_DED_COUNT) },
- { "TCP_CMD_FIFO", SOC15_REG_ENTRY(GC, 0, mmTCP_EDC_CNT_NEW), 1, 16,
- REG_FIELD_MASK(TCP_EDC_CNT_NEW, CMD_FIFO_SED_COUNT), 0 },
- { "TCP_VM_FIFO", SOC15_REG_ENTRY(GC, 0, mmTCP_EDC_CNT_NEW), 1, 16,
- REG_FIELD_MASK(TCP_EDC_CNT_NEW, VM_FIFO_SEC_COUNT), 0 },
- { "TCP_DB_RAM", SOC15_REG_ENTRY(GC, 0, mmTCP_EDC_CNT_NEW), 1, 16,
- REG_FIELD_MASK(TCP_EDC_CNT_NEW, DB_RAM_SED_COUNT), 0 },
- { "TCP_UTCL1_LFIFO0", SOC15_REG_ENTRY(GC, 0, mmTCP_EDC_CNT_NEW), 1, 16,
- REG_FIELD_MASK(TCP_EDC_CNT_NEW, UTCL1_LFIFO0_SEC_COUNT),
- REG_FIELD_MASK(TCP_EDC_CNT_NEW, UTCL1_LFIFO0_DED_COUNT) },
- { "TCP_UTCL1_LFIFO1", SOC15_REG_ENTRY(GC, 0, mmTCP_EDC_CNT_NEW), 1, 16,
- REG_FIELD_MASK(TCP_EDC_CNT_NEW, UTCL1_LFIFO1_SEC_COUNT),
- REG_FIELD_MASK(TCP_EDC_CNT_NEW, UTCL1_LFIFO1_DED_COUNT) },
- { "TD_SS_FIFO_LO", SOC15_REG_ENTRY(GC, 0, mmTD_EDC_CNT), 1, 16,
- REG_FIELD_MASK(TD_EDC_CNT, SS_FIFO_LO_SEC_COUNT),
- REG_FIELD_MASK(TD_EDC_CNT, SS_FIFO_LO_DED_COUNT) },
- { "TD_SS_FIFO_HI", SOC15_REG_ENTRY(GC, 0, mmTD_EDC_CNT), 1, 16,
- REG_FIELD_MASK(TD_EDC_CNT, SS_FIFO_HI_SEC_COUNT),
- REG_FIELD_MASK(TD_EDC_CNT, SS_FIFO_HI_DED_COUNT) },
- { "TD_CS_FIFO", SOC15_REG_ENTRY(GC, 0, mmTD_EDC_CNT), 1, 16,
- REG_FIELD_MASK(TD_EDC_CNT, CS_FIFO_SED_COUNT), 0 },
- { "SQ_LDS_D", SOC15_REG_ENTRY(GC, 0, mmSQ_EDC_CNT), 1, 16,
- REG_FIELD_MASK(SQ_EDC_CNT, LDS_D_SEC_COUNT),
- REG_FIELD_MASK(SQ_EDC_CNT, LDS_D_DED_COUNT) },
- { "SQ_LDS_I", SOC15_REG_ENTRY(GC, 0, mmSQ_EDC_CNT), 1, 16,
- REG_FIELD_MASK(SQ_EDC_CNT, LDS_I_SEC_COUNT),
- REG_FIELD_MASK(SQ_EDC_CNT, LDS_I_DED_COUNT) },
- { "SQ_SGPR", SOC15_REG_ENTRY(GC, 0, mmSQ_EDC_CNT), 1, 16,
- REG_FIELD_MASK(SQ_EDC_CNT, SGPR_SEC_COUNT),
- REG_FIELD_MASK(SQ_EDC_CNT, SGPR_DED_COUNT) },
- { "SQ_VGPR0", SOC15_REG_ENTRY(GC, 0, mmSQ_EDC_CNT), 1, 16,
- REG_FIELD_MASK(SQ_EDC_CNT, VGPR0_SEC_COUNT),
- REG_FIELD_MASK(SQ_EDC_CNT, VGPR0_DED_COUNT) },
- { "SQ_VGPR1", SOC15_REG_ENTRY(GC, 0, mmSQ_EDC_CNT), 1, 16,
- REG_FIELD_MASK(SQ_EDC_CNT, VGPR1_SEC_COUNT),
- REG_FIELD_MASK(SQ_EDC_CNT, VGPR1_DED_COUNT) },
- { "SQ_VGPR2", SOC15_REG_ENTRY(GC, 0, mmSQ_EDC_CNT), 1, 16,
- REG_FIELD_MASK(SQ_EDC_CNT, VGPR2_SEC_COUNT),
- REG_FIELD_MASK(SQ_EDC_CNT, VGPR2_DED_COUNT) },
- { "SQ_VGPR3", SOC15_REG_ENTRY(GC, 0, mmSQ_EDC_CNT), 1, 16,
- REG_FIELD_MASK(SQ_EDC_CNT, VGPR3_SEC_COUNT),
- REG_FIELD_MASK(SQ_EDC_CNT, VGPR3_DED_COUNT) },
+ SOC15_REG_FIELD(TCC_EDC_CNT2, WRRET_TAG_WRITE_RETURN_SED_COUNT),
+ 0, 0
+ },
+ { "TCC_ATOMIC_RETURN_BUFFER", SOC15_REG_ENTRY(GC, 0, mmTCC_EDC_CNT2),
+ SOC15_REG_FIELD(TCC_EDC_CNT2, ATOMIC_RETURN_BUFFER_SED_COUNT),
+ 0, 0
+ },
+ { "TCI_WRITE_RAM", SOC15_REG_ENTRY(GC, 0, mmTCI_EDC_CNT),
+ SOC15_REG_FIELD(TCI_EDC_CNT, WRITE_RAM_SED_COUNT),
+ 0, 0
+ },
+ { "TCP_CACHE_RAM", SOC15_REG_ENTRY(GC, 0, mmTCP_EDC_CNT_NEW),
+ SOC15_REG_FIELD(TCP_EDC_CNT_NEW, CACHE_RAM_SEC_COUNT),
+ SOC15_REG_FIELD(TCP_EDC_CNT_NEW, CACHE_RAM_DED_COUNT)
+ },
+ { "TCP_LFIFO_RAM", SOC15_REG_ENTRY(GC, 0, mmTCP_EDC_CNT_NEW),
+ SOC15_REG_FIELD(TCP_EDC_CNT_NEW, LFIFO_RAM_SEC_COUNT),
+ SOC15_REG_FIELD(TCP_EDC_CNT_NEW, LFIFO_RAM_DED_COUNT)
+ },
+ { "TCP_CMD_FIFO", SOC15_REG_ENTRY(GC, 0, mmTCP_EDC_CNT_NEW),
+ SOC15_REG_FIELD(TCP_EDC_CNT_NEW, CMD_FIFO_SED_COUNT),
+ 0, 0
+ },
+ { "TCP_VM_FIFO", SOC15_REG_ENTRY(GC, 0, mmTCP_EDC_CNT_NEW),
+ SOC15_REG_FIELD(TCP_EDC_CNT_NEW, VM_FIFO_SEC_COUNT),
+ 0, 0
+ },
+ { "TCP_DB_RAM", SOC15_REG_ENTRY(GC, 0, mmTCP_EDC_CNT_NEW),
+ SOC15_REG_FIELD(TCP_EDC_CNT_NEW, DB_RAM_SED_COUNT),
+ 0, 0
+ },
+ { "TCP_UTCL1_LFIFO0", SOC15_REG_ENTRY(GC, 0, mmTCP_EDC_CNT_NEW),
+ SOC15_REG_FIELD(TCP_EDC_CNT_NEW, UTCL1_LFIFO0_SEC_COUNT),
+ SOC15_REG_FIELD(TCP_EDC_CNT_NEW, UTCL1_LFIFO0_DED_COUNT)
+ },
+ { "TCP_UTCL1_LFIFO1", SOC15_REG_ENTRY(GC, 0, mmTCP_EDC_CNT_NEW),
+ SOC15_REG_FIELD(TCP_EDC_CNT_NEW, UTCL1_LFIFO1_SEC_COUNT),
+ SOC15_REG_FIELD(TCP_EDC_CNT_NEW, UTCL1_LFIFO1_DED_COUNT)
+ },
+ { "TD_SS_FIFO_LO", SOC15_REG_ENTRY(GC, 0, mmTD_EDC_CNT),
+ SOC15_REG_FIELD(TD_EDC_CNT, SS_FIFO_LO_SEC_COUNT),
+ SOC15_REG_FIELD(TD_EDC_CNT, SS_FIFO_LO_DED_COUNT)
+ },
+ { "TD_SS_FIFO_HI", SOC15_REG_ENTRY(GC, 0, mmTD_EDC_CNT),
+ SOC15_REG_FIELD(TD_EDC_CNT, SS_FIFO_HI_SEC_COUNT),
+ SOC15_REG_FIELD(TD_EDC_CNT, SS_FIFO_HI_DED_COUNT)
+ },
+ { "TD_CS_FIFO", SOC15_REG_ENTRY(GC, 0, mmTD_EDC_CNT),
+ SOC15_REG_FIELD(TD_EDC_CNT, CS_FIFO_SED_COUNT),
+ 0, 0
+ },
+ { "SQ_LDS_D", SOC15_REG_ENTRY(GC, 0, mmSQ_EDC_CNT),
+ SOC15_REG_FIELD(SQ_EDC_CNT, LDS_D_SEC_COUNT),
+ SOC15_REG_FIELD(SQ_EDC_CNT, LDS_D_DED_COUNT)
+ },
+ { "SQ_LDS_I", SOC15_REG_ENTRY(GC, 0, mmSQ_EDC_CNT),
+ SOC15_REG_FIELD(SQ_EDC_CNT, LDS_I_SEC_COUNT),
+ SOC15_REG_FIELD(SQ_EDC_CNT, LDS_I_DED_COUNT)
+ },
+ { "SQ_SGPR", SOC15_REG_ENTRY(GC, 0, mmSQ_EDC_CNT),
+ SOC15_REG_FIELD(SQ_EDC_CNT, SGPR_SEC_COUNT),
+ SOC15_REG_FIELD(SQ_EDC_CNT, SGPR_DED_COUNT)
+ },
+ { "SQ_VGPR0", SOC15_REG_ENTRY(GC, 0, mmSQ_EDC_CNT),
+ SOC15_REG_FIELD(SQ_EDC_CNT, VGPR0_SEC_COUNT),
+ SOC15_REG_FIELD(SQ_EDC_CNT, VGPR0_DED_COUNT)
+ },
+ { "SQ_VGPR1", SOC15_REG_ENTRY(GC, 0, mmSQ_EDC_CNT),
+ SOC15_REG_FIELD(SQ_EDC_CNT, VGPR1_SEC_COUNT),
+ SOC15_REG_FIELD(SQ_EDC_CNT, VGPR1_DED_COUNT)
+ },
+ { "SQ_VGPR2", SOC15_REG_ENTRY(GC, 0, mmSQ_EDC_CNT),
+ SOC15_REG_FIELD(SQ_EDC_CNT, VGPR2_SEC_COUNT),
+ SOC15_REG_FIELD(SQ_EDC_CNT, VGPR2_DED_COUNT)
+ },
+ { "SQ_VGPR3", SOC15_REG_ENTRY(GC, 0, mmSQ_EDC_CNT),
+ SOC15_REG_FIELD(SQ_EDC_CNT, VGPR3_SEC_COUNT),
+ SOC15_REG_FIELD(SQ_EDC_CNT, VGPR3_DED_COUNT)
+ },
{ "SQC_DATA_CU0_WRITE_DATA_BUF", SOC15_REG_ENTRY(GC, 0, mmSQC_EDC_CNT),
- 1, 6, REG_FIELD_MASK(SQC_EDC_CNT, DATA_CU0_WRITE_DATA_BUF_SEC_COUNT),
- REG_FIELD_MASK(SQC_EDC_CNT, DATA_CU0_WRITE_DATA_BUF_DED_COUNT) },
- { "SQC_DATA_CU0_UTCL1_LFIFO", SOC15_REG_ENTRY(GC, 0, mmSQC_EDC_CNT), 1,
- 6, REG_FIELD_MASK(SQC_EDC_CNT, DATA_CU0_UTCL1_LFIFO_SEC_COUNT),
- REG_FIELD_MASK(SQC_EDC_CNT, DATA_CU0_UTCL1_LFIFO_DED_COUNT) },
+ SOC15_REG_FIELD(SQC_EDC_CNT, DATA_CU0_WRITE_DATA_BUF_SEC_COUNT),
+ SOC15_REG_FIELD(SQC_EDC_CNT, DATA_CU0_WRITE_DATA_BUF_DED_COUNT)
+ },
+ { "SQC_DATA_CU0_UTCL1_LFIFO", SOC15_REG_ENTRY(GC, 0, mmSQC_EDC_CNT),
+ SOC15_REG_FIELD(SQC_EDC_CNT, DATA_CU0_UTCL1_LFIFO_SEC_COUNT),
+ SOC15_REG_FIELD(SQC_EDC_CNT, DATA_CU0_UTCL1_LFIFO_DED_COUNT)
+ },
{ "SQC_DATA_CU1_WRITE_DATA_BUF", SOC15_REG_ENTRY(GC, 0, mmSQC_EDC_CNT),
- 1, 6, REG_FIELD_MASK(SQC_EDC_CNT, DATA_CU1_WRITE_DATA_BUF_SEC_COUNT),
- REG_FIELD_MASK(SQC_EDC_CNT, DATA_CU1_WRITE_DATA_BUF_DED_COUNT) },
- { "SQC_DATA_CU1_UTCL1_LFIFO", SOC15_REG_ENTRY(GC, 0, mmSQC_EDC_CNT), 1,
- 6, REG_FIELD_MASK(SQC_EDC_CNT, DATA_CU1_UTCL1_LFIFO_SEC_COUNT),
- REG_FIELD_MASK(SQC_EDC_CNT, DATA_CU1_UTCL1_LFIFO_DED_COUNT) },
+ SOC15_REG_FIELD(SQC_EDC_CNT, DATA_CU1_WRITE_DATA_BUF_SEC_COUNT),
+ SOC15_REG_FIELD(SQC_EDC_CNT, DATA_CU1_WRITE_DATA_BUF_DED_COUNT)
+ },
+ { "SQC_DATA_CU1_UTCL1_LFIFO", SOC15_REG_ENTRY(GC, 0, mmSQC_EDC_CNT),
+ SOC15_REG_FIELD(SQC_EDC_CNT, DATA_CU1_UTCL1_LFIFO_SEC_COUNT),
+ SOC15_REG_FIELD(SQC_EDC_CNT, DATA_CU1_UTCL1_LFIFO_DED_COUNT)
+ },
{ "SQC_DATA_CU2_WRITE_DATA_BUF", SOC15_REG_ENTRY(GC, 0, mmSQC_EDC_CNT),
- 1, 6, REG_FIELD_MASK(SQC_EDC_CNT, DATA_CU2_WRITE_DATA_BUF_SEC_COUNT),
- REG_FIELD_MASK(SQC_EDC_CNT, DATA_CU2_WRITE_DATA_BUF_DED_COUNT) },
- { "SQC_DATA_CU2_UTCL1_LFIFO", SOC15_REG_ENTRY(GC, 0, mmSQC_EDC_CNT), 1,
- 6, REG_FIELD_MASK(SQC_EDC_CNT, DATA_CU2_UTCL1_LFIFO_SEC_COUNT),
- REG_FIELD_MASK(SQC_EDC_CNT, DATA_CU2_UTCL1_LFIFO_DED_COUNT) },
- { "SQC_INST_BANKA_TAG_RAM", SOC15_REG_ENTRY(GC, 0, mmSQC_EDC_CNT2), 1,
- 6, REG_FIELD_MASK(SQC_EDC_CNT2, INST_BANKA_TAG_RAM_SEC_COUNT),
- REG_FIELD_MASK(SQC_EDC_CNT2, INST_BANKA_TAG_RAM_DED_COUNT) },
- { "SQC_INST_BANKA_BANK_RAM", SOC15_REG_ENTRY(GC, 0, mmSQC_EDC_CNT2), 1,
- 6, REG_FIELD_MASK(SQC_EDC_CNT2, INST_BANKA_BANK_RAM_SEC_COUNT),
- REG_FIELD_MASK(SQC_EDC_CNT2, INST_BANKA_BANK_RAM_DED_COUNT) },
- { "SQC_DATA_BANKA_TAG_RAM", SOC15_REG_ENTRY(GC, 0, mmSQC_EDC_CNT2), 1,
- 6, REG_FIELD_MASK(SQC_EDC_CNT2, DATA_BANKA_TAG_RAM_SEC_COUNT),
- REG_FIELD_MASK(SQC_EDC_CNT2, DATA_BANKA_TAG_RAM_DED_COUNT) },
- { "SQC_DATA_BANKA_BANK_RAM", SOC15_REG_ENTRY(GC, 0, mmSQC_EDC_CNT2), 1,
- 6, REG_FIELD_MASK(SQC_EDC_CNT2, DATA_BANKA_BANK_RAM_SEC_COUNT),
- REG_FIELD_MASK(SQC_EDC_CNT2, DATA_BANKA_BANK_RAM_DED_COUNT) },
- { "SQC_INST_BANKA_UTCL1_MISS_FIFO",
- SOC15_REG_ENTRY(GC, 0, mmSQC_EDC_CNT2), 1, 6,
- REG_FIELD_MASK(SQC_EDC_CNT2, INST_BANKA_UTCL1_MISS_FIFO_SED_COUNT),
- 0 },
- { "SQC_INST_BANKA_MISS_FIFO", SOC15_REG_ENTRY(GC, 0, mmSQC_EDC_CNT2), 1,
- 6, REG_FIELD_MASK(SQC_EDC_CNT2, INST_BANKA_MISS_FIFO_SED_COUNT), 0 },
- { "SQC_DATA_BANKA_HIT_FIFO", SOC15_REG_ENTRY(GC, 0, mmSQC_EDC_CNT2), 1,
- 6, REG_FIELD_MASK(SQC_EDC_CNT2, DATA_BANKA_HIT_FIFO_SED_COUNT), 0 },
- { "SQC_DATA_BANKA_MISS_FIFO", SOC15_REG_ENTRY(GC, 0, mmSQC_EDC_CNT2), 1,
- 6, REG_FIELD_MASK(SQC_EDC_CNT2, DATA_BANKA_MISS_FIFO_SED_COUNT), 0 },
- { "SQC_DATA_BANKA_DIRTY_BIT_RAM",
- SOC15_REG_ENTRY(GC, 0, mmSQC_EDC_CNT2), 1, 6,
- REG_FIELD_MASK(SQC_EDC_CNT2, DATA_BANKA_DIRTY_BIT_RAM_SED_COUNT), 0 },
- { "SQC_INST_UTCL1_LFIFO", SOC15_REG_ENTRY(GC, 0, mmSQC_EDC_CNT2), 1, 6,
- REG_FIELD_MASK(SQC_EDC_CNT2, INST_UTCL1_LFIFO_SEC_COUNT),
- REG_FIELD_MASK(SQC_EDC_CNT2, INST_UTCL1_LFIFO_DED_COUNT) },
- { "SQC_INST_BANKB_TAG_RAM", SOC15_REG_ENTRY(GC, 0, mmSQC_EDC_CNT3), 1,
- 6, REG_FIELD_MASK(SQC_EDC_CNT3, INST_BANKB_TAG_RAM_SEC_COUNT),
- REG_FIELD_MASK(SQC_EDC_CNT3, INST_BANKB_TAG_RAM_DED_COUNT) },
- { "SQC_INST_BANKB_BANK_RAM", SOC15_REG_ENTRY(GC, 0, mmSQC_EDC_CNT3), 1,
- 6, REG_FIELD_MASK(SQC_EDC_CNT3, INST_BANKB_BANK_RAM_SEC_COUNT),
- REG_FIELD_MASK(SQC_EDC_CNT3, INST_BANKB_BANK_RAM_DED_COUNT) },
- { "SQC_DATA_BANKB_TAG_RAM", SOC15_REG_ENTRY(GC, 0, mmSQC_EDC_CNT3), 1,
- 6, REG_FIELD_MASK(SQC_EDC_CNT3, DATA_BANKB_TAG_RAM_SEC_COUNT),
- REG_FIELD_MASK(SQC_EDC_CNT3, DATA_BANKB_TAG_RAM_DED_COUNT) },
- { "SQC_DATA_BANKB_BANK_RAM", SOC15_REG_ENTRY(GC, 0, mmSQC_EDC_CNT3), 1,
- 6, REG_FIELD_MASK(SQC_EDC_CNT3, DATA_BANKB_BANK_RAM_SEC_COUNT),
- REG_FIELD_MASK(SQC_EDC_CNT3, DATA_BANKB_BANK_RAM_DED_COUNT) },
- { "SQC_INST_BANKB_UTCL1_MISS_FIFO",
- SOC15_REG_ENTRY(GC, 0, mmSQC_EDC_CNT3), 1, 6,
- REG_FIELD_MASK(SQC_EDC_CNT3, INST_BANKB_UTCL1_MISS_FIFO_SED_COUNT),
- 0 },
- { "SQC_INST_BANKB_MISS_FIFO", SOC15_REG_ENTRY(GC, 0, mmSQC_EDC_CNT3), 1,
- 6, REG_FIELD_MASK(SQC_EDC_CNT3, INST_BANKB_MISS_FIFO_SED_COUNT), 0 },
- { "SQC_DATA_BANKB_HIT_FIFO", SOC15_REG_ENTRY(GC, 0, mmSQC_EDC_CNT3), 1,
- 6, REG_FIELD_MASK(SQC_EDC_CNT3, DATA_BANKB_HIT_FIFO_SED_COUNT), 0 },
- { "SQC_DATA_BANKB_MISS_FIFO", SOC15_REG_ENTRY(GC, 0, mmSQC_EDC_CNT3), 1,
- 6, REG_FIELD_MASK(SQC_EDC_CNT3, DATA_BANKB_MISS_FIFO_SED_COUNT), 0 },
- { "SQC_DATA_BANKB_DIRTY_BIT_RAM",
- SOC15_REG_ENTRY(GC, 0, mmSQC_EDC_CNT3), 1, 6,
- REG_FIELD_MASK(SQC_EDC_CNT3, DATA_BANKB_DIRTY_BIT_RAM_SED_COUNT), 0 },
- { "EA_DRAMRD_CMDMEM", SOC15_REG_ENTRY(GC, 0, mmGCEA_EDC_CNT), 0, 32,
- REG_FIELD_MASK(GCEA_EDC_CNT, DRAMRD_CMDMEM_SEC_COUNT),
- REG_FIELD_MASK(GCEA_EDC_CNT, DRAMRD_CMDMEM_DED_COUNT) },
- { "EA_DRAMWR_CMDMEM", SOC15_REG_ENTRY(GC, 0, mmGCEA_EDC_CNT), 0, 32,
- REG_FIELD_MASK(GCEA_EDC_CNT, DRAMWR_CMDMEM_SEC_COUNT),
- REG_FIELD_MASK(GCEA_EDC_CNT, DRAMWR_CMDMEM_DED_COUNT) },
- { "EA_DRAMWR_DATAMEM", SOC15_REG_ENTRY(GC, 0, mmGCEA_EDC_CNT), 0, 32,
- REG_FIELD_MASK(GCEA_EDC_CNT, DRAMWR_DATAMEM_SEC_COUNT),
- REG_FIELD_MASK(GCEA_EDC_CNT, DRAMWR_DATAMEM_DED_COUNT) },
- { "EA_RRET_TAGMEM", SOC15_REG_ENTRY(GC, 0, mmGCEA_EDC_CNT), 0, 32,
- REG_FIELD_MASK(GCEA_EDC_CNT, RRET_TAGMEM_SEC_COUNT),
- REG_FIELD_MASK(GCEA_EDC_CNT, RRET_TAGMEM_DED_COUNT) },
- { "EA_WRET_TAGMEM", SOC15_REG_ENTRY(GC, 0, mmGCEA_EDC_CNT), 0, 32,
- REG_FIELD_MASK(GCEA_EDC_CNT, WRET_TAGMEM_SEC_COUNT),
- REG_FIELD_MASK(GCEA_EDC_CNT, WRET_TAGMEM_DED_COUNT) },
- { "EA_DRAMRD_PAGEMEM", SOC15_REG_ENTRY(GC, 0, mmGCEA_EDC_CNT), 0, 32,
- REG_FIELD_MASK(GCEA_EDC_CNT, DRAMRD_PAGEMEM_SED_COUNT), 0 },
- { "EA_DRAMWR_PAGEMEM", SOC15_REG_ENTRY(GC, 0, mmGCEA_EDC_CNT), 0, 32,
- REG_FIELD_MASK(GCEA_EDC_CNT, DRAMWR_PAGEMEM_SED_COUNT), 0 },
- { "EA_IORD_CMDMEM", SOC15_REG_ENTRY(GC, 0, mmGCEA_EDC_CNT), 0, 32,
- REG_FIELD_MASK(GCEA_EDC_CNT, IORD_CMDMEM_SED_COUNT), 0 },
- { "EA_IOWR_CMDMEM", SOC15_REG_ENTRY(GC, 0, mmGCEA_EDC_CNT), 0, 32,
- REG_FIELD_MASK(GCEA_EDC_CNT, IOWR_CMDMEM_SED_COUNT), 0 },
- { "EA_IOWR_DATAMEM", SOC15_REG_ENTRY(GC, 0, mmGCEA_EDC_CNT), 0, 32,
- REG_FIELD_MASK(GCEA_EDC_CNT, IOWR_DATAMEM_SED_COUNT), 0 },
- { "GMIRD_CMDMEM", SOC15_REG_ENTRY(GC, 0, mmGCEA_EDC_CNT2), 0, 32,
- REG_FIELD_MASK(GCEA_EDC_CNT2, GMIRD_CMDMEM_SEC_COUNT),
- REG_FIELD_MASK(GCEA_EDC_CNT2, GMIRD_CMDMEM_DED_COUNT) },
- { "GMIWR_CMDMEM", SOC15_REG_ENTRY(GC, 0, mmGCEA_EDC_CNT2), 0, 32,
- REG_FIELD_MASK(GCEA_EDC_CNT2, GMIWR_CMDMEM_SEC_COUNT),
- REG_FIELD_MASK(GCEA_EDC_CNT2, GMIWR_CMDMEM_DED_COUNT) },
- { "GMIWR_DATAMEM", SOC15_REG_ENTRY(GC, 0, mmGCEA_EDC_CNT2), 0, 32,
- REG_FIELD_MASK(GCEA_EDC_CNT2, GMIWR_DATAMEM_SEC_COUNT),
- REG_FIELD_MASK(GCEA_EDC_CNT2, GMIWR_DATAMEM_DED_COUNT) },
- { "GMIRD_PAGEMEM", SOC15_REG_ENTRY(GC, 0, mmGCEA_EDC_CNT2), 0, 32,
- REG_FIELD_MASK(GCEA_EDC_CNT2, GMIRD_PAGEMEM_SED_COUNT), 0 },
- { "GMIWR_PAGEMEM", SOC15_REG_ENTRY(GC, 0, mmGCEA_EDC_CNT2), 0, 32,
- REG_FIELD_MASK(GCEA_EDC_CNT2, GMIWR_PAGEMEM_SED_COUNT), 0 },
- { "MAM_D0MEM", SOC15_REG_ENTRY(GC, 0, mmGCEA_EDC_CNT2), 0, 32,
- REG_FIELD_MASK(GCEA_EDC_CNT2, MAM_D0MEM_SED_COUNT), 0 },
- { "MAM_D1MEM", SOC15_REG_ENTRY(GC, 0, mmGCEA_EDC_CNT2), 0, 32,
- REG_FIELD_MASK(GCEA_EDC_CNT2, MAM_D1MEM_SED_COUNT), 0 },
- { "MAM_D2MEM", SOC15_REG_ENTRY(GC, 0, mmGCEA_EDC_CNT2), 0, 32,
- REG_FIELD_MASK(GCEA_EDC_CNT2, MAM_D2MEM_SED_COUNT), 0 },
- { "MAM_D3MEM", SOC15_REG_ENTRY(GC, 0, mmGCEA_EDC_CNT2), 0, 32,
- REG_FIELD_MASK(GCEA_EDC_CNT2, MAM_D3MEM_SED_COUNT), 0 },
+ SOC15_REG_FIELD(SQC_EDC_CNT, DATA_CU2_WRITE_DATA_BUF_SEC_COUNT),
+ SOC15_REG_FIELD(SQC_EDC_CNT, DATA_CU2_WRITE_DATA_BUF_DED_COUNT)
+ },
+ { "SQC_DATA_CU2_UTCL1_LFIFO", SOC15_REG_ENTRY(GC, 0, mmSQC_EDC_CNT),
+ SOC15_REG_FIELD(SQC_EDC_CNT, DATA_CU2_UTCL1_LFIFO_SEC_COUNT),
+ SOC15_REG_FIELD(SQC_EDC_CNT, DATA_CU2_UTCL1_LFIFO_DED_COUNT)
+ },
+ { "SQC_INST_BANKA_TAG_RAM", SOC15_REG_ENTRY(GC, 0, mmSQC_EDC_CNT2),
+ SOC15_REG_FIELD(SQC_EDC_CNT2, INST_BANKA_TAG_RAM_SEC_COUNT),
+ SOC15_REG_FIELD(SQC_EDC_CNT2, INST_BANKA_TAG_RAM_DED_COUNT)
+ },
+ { "SQC_INST_BANKA_BANK_RAM", SOC15_REG_ENTRY(GC, 0, mmSQC_EDC_CNT2),
+ SOC15_REG_FIELD(SQC_EDC_CNT2, INST_BANKA_BANK_RAM_SEC_COUNT),
+ SOC15_REG_FIELD(SQC_EDC_CNT2, INST_BANKA_BANK_RAM_DED_COUNT)
+ },
+ { "SQC_DATA_BANKA_TAG_RAM", SOC15_REG_ENTRY(GC, 0, mmSQC_EDC_CNT2),
+ SOC15_REG_FIELD(SQC_EDC_CNT2, DATA_BANKA_TAG_RAM_SEC_COUNT),
+ SOC15_REG_FIELD(SQC_EDC_CNT2, DATA_BANKA_TAG_RAM_DED_COUNT)
+ },
+ { "SQC_DATA_BANKA_BANK_RAM", SOC15_REG_ENTRY(GC, 0, mmSQC_EDC_CNT2),
+ SOC15_REG_FIELD(SQC_EDC_CNT2, DATA_BANKA_BANK_RAM_SEC_COUNT),
+ SOC15_REG_FIELD(SQC_EDC_CNT2, DATA_BANKA_BANK_RAM_DED_COUNT)
+ },
+ { "SQC_INST_BANKA_UTCL1_MISS_FIFO", SOC15_REG_ENTRY(GC, 0, mmSQC_EDC_CNT2),
+ SOC15_REG_FIELD(SQC_EDC_CNT2, INST_BANKA_UTCL1_MISS_FIFO_SED_COUNT),
+ 0, 0
+ },
+ { "SQC_INST_BANKA_MISS_FIFO", SOC15_REG_ENTRY(GC, 0, mmSQC_EDC_CNT2),
+ SOC15_REG_FIELD(SQC_EDC_CNT2, INST_BANKA_MISS_FIFO_SED_COUNT),
+ 0, 0
+ },
+ { "SQC_DATA_BANKA_HIT_FIFO", SOC15_REG_ENTRY(GC, 0, mmSQC_EDC_CNT2),
+ SOC15_REG_FIELD(SQC_EDC_CNT2, DATA_BANKA_HIT_FIFO_SED_COUNT),
+ 0, 0
+ },
+ { "SQC_DATA_BANKA_MISS_FIFO", SOC15_REG_ENTRY(GC, 0, mmSQC_EDC_CNT2),
+ SOC15_REG_FIELD(SQC_EDC_CNT2, DATA_BANKA_MISS_FIFO_SED_COUNT),
+ 0, 0
+ },
+ { "SQC_DATA_BANKA_DIRTY_BIT_RAM", SOC15_REG_ENTRY(GC, 0, mmSQC_EDC_CNT2),
+ SOC15_REG_FIELD(SQC_EDC_CNT2, DATA_BANKA_DIRTY_BIT_RAM_SED_COUNT),
+ 0, 0
+ },
+ { "SQC_INST_UTCL1_LFIFO", SOC15_REG_ENTRY(GC, 0, mmSQC_EDC_CNT2),
+ SOC15_REG_FIELD(SQC_EDC_CNT2, INST_UTCL1_LFIFO_SEC_COUNT),
+ SOC15_REG_FIELD(SQC_EDC_CNT2, INST_UTCL1_LFIFO_DED_COUNT)
+ },
+ { "SQC_INST_BANKB_TAG_RAM", SOC15_REG_ENTRY(GC, 0, mmSQC_EDC_CNT3),
+ SOC15_REG_FIELD(SQC_EDC_CNT3, INST_BANKB_TAG_RAM_SEC_COUNT),
+ SOC15_REG_FIELD(SQC_EDC_CNT3, INST_BANKB_TAG_RAM_DED_COUNT)
+ },
+ { "SQC_INST_BANKB_BANK_RAM", SOC15_REG_ENTRY(GC, 0, mmSQC_EDC_CNT3),
+ SOC15_REG_FIELD(SQC_EDC_CNT3, INST_BANKB_BANK_RAM_SEC_COUNT),
+ SOC15_REG_FIELD(SQC_EDC_CNT3, INST_BANKB_BANK_RAM_DED_COUNT)
+ },
+ { "SQC_DATA_BANKB_TAG_RAM", SOC15_REG_ENTRY(GC, 0, mmSQC_EDC_CNT3),
+ SOC15_REG_FIELD(SQC_EDC_CNT3, DATA_BANKB_TAG_RAM_SEC_COUNT),
+ SOC15_REG_FIELD(SQC_EDC_CNT3, DATA_BANKB_TAG_RAM_DED_COUNT)
+ },
+ { "SQC_DATA_BANKB_BANK_RAM", SOC15_REG_ENTRY(GC, 0, mmSQC_EDC_CNT3),
+ SOC15_REG_FIELD(SQC_EDC_CNT3, DATA_BANKB_BANK_RAM_SEC_COUNT),
+ SOC15_REG_FIELD(SQC_EDC_CNT3, DATA_BANKB_BANK_RAM_DED_COUNT)
+ },
+ { "SQC_INST_BANKB_UTCL1_MISS_FIFO", SOC15_REG_ENTRY(GC, 0, mmSQC_EDC_CNT3),
+ SOC15_REG_FIELD(SQC_EDC_CNT3, INST_BANKB_UTCL1_MISS_FIFO_SED_COUNT),
+ 0, 0
+ },
+ { "SQC_INST_BANKB_MISS_FIFO", SOC15_REG_ENTRY(GC, 0, mmSQC_EDC_CNT3),
+ SOC15_REG_FIELD(SQC_EDC_CNT3, INST_BANKB_MISS_FIFO_SED_COUNT),
+ 0, 0
+ },
+ { "SQC_DATA_BANKB_HIT_FIFO", SOC15_REG_ENTRY(GC, 0, mmSQC_EDC_CNT3),
+ SOC15_REG_FIELD(SQC_EDC_CNT3, DATA_BANKB_HIT_FIFO_SED_COUNT),
+ 0, 0
+ },
+ { "SQC_DATA_BANKB_MISS_FIFO", SOC15_REG_ENTRY(GC, 0, mmSQC_EDC_CNT3),
+ SOC15_REG_FIELD(SQC_EDC_CNT3, DATA_BANKB_MISS_FIFO_SED_COUNT),
+ 0, 0
+ },
+ { "SQC_DATA_BANKB_DIRTY_BIT_RAM", SOC15_REG_ENTRY(GC, 0, mmSQC_EDC_CNT3),
+ SOC15_REG_FIELD(SQC_EDC_CNT3, DATA_BANKB_DIRTY_BIT_RAM_SED_COUNT),
+ 0, 0
+ },
+ { "EA_DRAMRD_CMDMEM", SOC15_REG_ENTRY(GC, 0, mmGCEA_EDC_CNT),
+ SOC15_REG_FIELD(GCEA_EDC_CNT, DRAMRD_CMDMEM_SEC_COUNT),
+ SOC15_REG_FIELD(GCEA_EDC_CNT, DRAMRD_CMDMEM_DED_COUNT)
+ },
+ { "EA_DRAMWR_CMDMEM", SOC15_REG_ENTRY(GC, 0, mmGCEA_EDC_CNT),
+ SOC15_REG_FIELD(GCEA_EDC_CNT, DRAMWR_CMDMEM_SEC_COUNT),
+ SOC15_REG_FIELD(GCEA_EDC_CNT, DRAMWR_CMDMEM_DED_COUNT)
+ },
+ { "EA_DRAMWR_DATAMEM", SOC15_REG_ENTRY(GC, 0, mmGCEA_EDC_CNT),
+ SOC15_REG_FIELD(GCEA_EDC_CNT, DRAMWR_DATAMEM_SEC_COUNT),
+ SOC15_REG_FIELD(GCEA_EDC_CNT, DRAMWR_DATAMEM_DED_COUNT)
+ },
+ { "EA_RRET_TAGMEM", SOC15_REG_ENTRY(GC, 0, mmGCEA_EDC_CNT),
+ SOC15_REG_FIELD(GCEA_EDC_CNT, RRET_TAGMEM_SEC_COUNT),
+ SOC15_REG_FIELD(GCEA_EDC_CNT, RRET_TAGMEM_DED_COUNT)
+ },
+ { "EA_WRET_TAGMEM", SOC15_REG_ENTRY(GC, 0, mmGCEA_EDC_CNT),
+ SOC15_REG_FIELD(GCEA_EDC_CNT, WRET_TAGMEM_SEC_COUNT),
+ SOC15_REG_FIELD(GCEA_EDC_CNT, WRET_TAGMEM_DED_COUNT)
+ },
+ { "EA_DRAMRD_PAGEMEM", SOC15_REG_ENTRY(GC, 0, mmGCEA_EDC_CNT),
+ SOC15_REG_FIELD(GCEA_EDC_CNT, DRAMRD_PAGEMEM_SED_COUNT),
+ 0, 0
+ },
+ { "EA_DRAMWR_PAGEMEM", SOC15_REG_ENTRY(GC, 0, mmGCEA_EDC_CNT),
+ SOC15_REG_FIELD(GCEA_EDC_CNT, DRAMWR_PAGEMEM_SED_COUNT),
+ 0, 0
+ },
+ { "EA_IORD_CMDMEM", SOC15_REG_ENTRY(GC, 0, mmGCEA_EDC_CNT),
+ SOC15_REG_FIELD(GCEA_EDC_CNT, IORD_CMDMEM_SED_COUNT),
+ 0, 0
+ },
+ { "EA_IOWR_CMDMEM", SOC15_REG_ENTRY(GC, 0, mmGCEA_EDC_CNT),
+ SOC15_REG_FIELD(GCEA_EDC_CNT, IOWR_CMDMEM_SED_COUNT),
+ 0, 0
+ },
+ { "EA_IOWR_DATAMEM", SOC15_REG_ENTRY(GC, 0, mmGCEA_EDC_CNT),
+ SOC15_REG_FIELD(GCEA_EDC_CNT, IOWR_DATAMEM_SED_COUNT),
+ 0, 0
+ },
+ { "GMIRD_CMDMEM", SOC15_REG_ENTRY(GC, 0, mmGCEA_EDC_CNT2),
+ SOC15_REG_FIELD(GCEA_EDC_CNT2, GMIRD_CMDMEM_SEC_COUNT),
+ SOC15_REG_FIELD(GCEA_EDC_CNT2, GMIRD_CMDMEM_DED_COUNT)
+ },
+ { "GMIWR_CMDMEM", SOC15_REG_ENTRY(GC, 0, mmGCEA_EDC_CNT2),
+ SOC15_REG_FIELD(GCEA_EDC_CNT2, GMIWR_CMDMEM_SEC_COUNT),
+ SOC15_REG_FIELD(GCEA_EDC_CNT2, GMIWR_CMDMEM_DED_COUNT)
+ },
+ { "GMIWR_DATAMEM", SOC15_REG_ENTRY(GC, 0, mmGCEA_EDC_CNT2),
+ SOC15_REG_FIELD(GCEA_EDC_CNT2, GMIWR_DATAMEM_SEC_COUNT),
+ SOC15_REG_FIELD(GCEA_EDC_CNT2, GMIWR_DATAMEM_DED_COUNT)
+ },
+ { "GMIRD_PAGEMEM", SOC15_REG_ENTRY(GC, 0, mmGCEA_EDC_CNT2),
+ SOC15_REG_FIELD(GCEA_EDC_CNT2, GMIRD_PAGEMEM_SED_COUNT),
+ 0, 0
+ },
+ { "GMIWR_PAGEMEM", SOC15_REG_ENTRY(GC, 0, mmGCEA_EDC_CNT2),
+ SOC15_REG_FIELD(GCEA_EDC_CNT2, GMIWR_PAGEMEM_SED_COUNT),
+ 0, 0
+ },
+ { "MAM_D0MEM", SOC15_REG_ENTRY(GC, 0, mmGCEA_EDC_CNT2),
+ SOC15_REG_FIELD(GCEA_EDC_CNT2, MAM_D0MEM_SED_COUNT),
+ 0, 0
+ },
+ { "MAM_D1MEM", SOC15_REG_ENTRY(GC, 0, mmGCEA_EDC_CNT2),
+ SOC15_REG_FIELD(GCEA_EDC_CNT2, MAM_D1MEM_SED_COUNT),
+ 0, 0
+ },
+ { "MAM_D2MEM", SOC15_REG_ENTRY(GC, 0, mmGCEA_EDC_CNT2),
+ SOC15_REG_FIELD(GCEA_EDC_CNT2, MAM_D2MEM_SED_COUNT),
+ 0, 0
+ },
+ { "MAM_D3MEM", SOC15_REG_ENTRY(GC, 0, mmGCEA_EDC_CNT2),
+ SOC15_REG_FIELD(GCEA_EDC_CNT2, MAM_D3MEM_SED_COUNT),
+ 0, 0
+ }
};
static int gfx_v9_0_ras_error_inject(struct amdgpu_device *adev,
@@ -5778,14 +5940,217 @@ static int gfx_v9_0_ras_error_inject(struct amdgpu_device *adev,
return ret;
}
+static const char *vml2_mems[] = {
+ "UTC_VML2_BANK_CACHE_0_BIGK_MEM0",
+ "UTC_VML2_BANK_CACHE_0_BIGK_MEM1",
+ "UTC_VML2_BANK_CACHE_0_4K_MEM0",
+ "UTC_VML2_BANK_CACHE_0_4K_MEM1",
+ "UTC_VML2_BANK_CACHE_1_BIGK_MEM0",
+ "UTC_VML2_BANK_CACHE_1_BIGK_MEM1",
+ "UTC_VML2_BANK_CACHE_1_4K_MEM0",
+ "UTC_VML2_BANK_CACHE_1_4K_MEM1",
+ "UTC_VML2_BANK_CACHE_2_BIGK_MEM0",
+ "UTC_VML2_BANK_CACHE_2_BIGK_MEM1",
+ "UTC_VML2_BANK_CACHE_2_4K_MEM0",
+ "UTC_VML2_BANK_CACHE_2_4K_MEM1",
+ "UTC_VML2_BANK_CACHE_3_BIGK_MEM0",
+ "UTC_VML2_BANK_CACHE_3_BIGK_MEM1",
+ "UTC_VML2_BANK_CACHE_3_4K_MEM0",
+ "UTC_VML2_BANK_CACHE_3_4K_MEM1",
+};
+
+static const char *vml2_walker_mems[] = {
+ "UTC_VML2_CACHE_PDE0_MEM0",
+ "UTC_VML2_CACHE_PDE0_MEM1",
+ "UTC_VML2_CACHE_PDE1_MEM0",
+ "UTC_VML2_CACHE_PDE1_MEM1",
+ "UTC_VML2_CACHE_PDE2_MEM0",
+ "UTC_VML2_CACHE_PDE2_MEM1",
+ "UTC_VML2_RDIF_LOG_FIFO",
+};
+
+static const char *atc_l2_cache_2m_mems[] = {
+ "UTC_ATCL2_CACHE_2M_BANK0_WAY0_MEM",
+ "UTC_ATCL2_CACHE_2M_BANK0_WAY1_MEM",
+ "UTC_ATCL2_CACHE_2M_BANK1_WAY0_MEM",
+ "UTC_ATCL2_CACHE_2M_BANK1_WAY1_MEM",
+};
+
+static const char *atc_l2_cache_4k_mems[] = {
+ "UTC_ATCL2_CACHE_4K_BANK0_WAY0_MEM0",
+ "UTC_ATCL2_CACHE_4K_BANK0_WAY0_MEM1",
+ "UTC_ATCL2_CACHE_4K_BANK0_WAY0_MEM2",
+ "UTC_ATCL2_CACHE_4K_BANK0_WAY0_MEM3",
+ "UTC_ATCL2_CACHE_4K_BANK0_WAY0_MEM4",
+ "UTC_ATCL2_CACHE_4K_BANK0_WAY0_MEM5",
+ "UTC_ATCL2_CACHE_4K_BANK0_WAY0_MEM6",
+ "UTC_ATCL2_CACHE_4K_BANK0_WAY0_MEM7",
+ "UTC_ATCL2_CACHE_4K_BANK0_WAY1_MEM0",
+ "UTC_ATCL2_CACHE_4K_BANK0_WAY1_MEM1",
+ "UTC_ATCL2_CACHE_4K_BANK0_WAY1_MEM2",
+ "UTC_ATCL2_CACHE_4K_BANK0_WAY1_MEM3",
+ "UTC_ATCL2_CACHE_4K_BANK0_WAY1_MEM4",
+ "UTC_ATCL2_CACHE_4K_BANK0_WAY1_MEM5",
+ "UTC_ATCL2_CACHE_4K_BANK0_WAY1_MEM6",
+ "UTC_ATCL2_CACHE_4K_BANK0_WAY1_MEM7",
+ "UTC_ATCL2_CACHE_4K_BANK1_WAY0_MEM0",
+ "UTC_ATCL2_CACHE_4K_BANK1_WAY0_MEM1",
+ "UTC_ATCL2_CACHE_4K_BANK1_WAY0_MEM2",
+ "UTC_ATCL2_CACHE_4K_BANK1_WAY0_MEM3",
+ "UTC_ATCL2_CACHE_4K_BANK1_WAY0_MEM4",
+ "UTC_ATCL2_CACHE_4K_BANK1_WAY0_MEM5",
+ "UTC_ATCL2_CACHE_4K_BANK1_WAY0_MEM6",
+ "UTC_ATCL2_CACHE_4K_BANK1_WAY0_MEM7",
+ "UTC_ATCL2_CACHE_4K_BANK1_WAY1_MEM0",
+ "UTC_ATCL2_CACHE_4K_BANK1_WAY1_MEM1",
+ "UTC_ATCL2_CACHE_4K_BANK1_WAY1_MEM2",
+ "UTC_ATCL2_CACHE_4K_BANK1_WAY1_MEM3",
+ "UTC_ATCL2_CACHE_4K_BANK1_WAY1_MEM4",
+ "UTC_ATCL2_CACHE_4K_BANK1_WAY1_MEM5",
+ "UTC_ATCL2_CACHE_4K_BANK1_WAY1_MEM6",
+ "UTC_ATCL2_CACHE_4K_BANK1_WAY1_MEM7",
+};
+
+static int gfx_v9_0_query_utc_edc_status(struct amdgpu_device *adev,
+ struct ras_err_data *err_data)
+{
+ uint32_t i, data;
+ uint32_t sec_count, ded_count;
+
+ WREG32_SOC15(GC, 0, mmVM_L2_MEM_ECC_INDEX, 255);
+ WREG32_SOC15(GC, 0, mmVM_L2_MEM_ECC_CNT, 0);
+ WREG32_SOC15(GC, 0, mmVM_L2_WALKER_MEM_ECC_INDEX, 255);
+ WREG32_SOC15(GC, 0, mmVM_L2_WALKER_MEM_ECC_CNT, 0);
+ WREG32_SOC15(GC, 0, mmATC_L2_CACHE_2M_EDC_INDEX, 255);
+ WREG32_SOC15(GC, 0, mmATC_L2_CACHE_2M_EDC_CNT, 0);
+ WREG32_SOC15(GC, 0, mmATC_L2_CACHE_4K_EDC_INDEX, 255);
+ WREG32_SOC15(GC, 0, mmATC_L2_CACHE_4K_EDC_CNT, 0);
+
+ for (i = 0; i < 16; i++) {
+ WREG32_SOC15(GC, 0, mmVM_L2_MEM_ECC_INDEX, i);
+ data = RREG32_SOC15(GC, 0, mmVM_L2_MEM_ECC_CNT);
+
+ sec_count = REG_GET_FIELD(data, VM_L2_MEM_ECC_CNT, SEC_COUNT);
+ if (sec_count) {
+ DRM_INFO("Instance[%d]: SubBlock %s, SEC %d\n", i,
+ vml2_mems[i], sec_count);
+ err_data->ce_count += sec_count;
+ }
+
+ ded_count = REG_GET_FIELD(data, VM_L2_MEM_ECC_CNT, DED_COUNT);
+ if (ded_count) {
+ DRM_INFO("Instance[%d]: SubBlock %s, DED %d\n", i,
+ vml2_mems[i], ded_count);
+ err_data->ue_count += ded_count;
+ }
+ }
+
+ for (i = 0; i < 7; i++) {
+ WREG32_SOC15(GC, 0, mmVM_L2_WALKER_MEM_ECC_INDEX, i);
+ data = RREG32_SOC15(GC, 0, mmVM_L2_WALKER_MEM_ECC_CNT);
+
+ sec_count = REG_GET_FIELD(data, VM_L2_WALKER_MEM_ECC_CNT,
+ SEC_COUNT);
+ if (sec_count) {
+ DRM_INFO("Instance[%d]: SubBlock %s, SEC %d\n", i,
+ vml2_walker_mems[i], sec_count);
+ err_data->ce_count += sec_count;
+ }
+
+ ded_count = REG_GET_FIELD(data, VM_L2_WALKER_MEM_ECC_CNT,
+ DED_COUNT);
+ if (ded_count) {
+ DRM_INFO("Instance[%d]: SubBlock %s, DED %d\n", i,
+ vml2_walker_mems[i], ded_count);
+ err_data->ue_count += ded_count;
+ }
+ }
+
+ for (i = 0; i < 4; i++) {
+ WREG32_SOC15(GC, 0, mmATC_L2_CACHE_2M_EDC_INDEX, i);
+ data = RREG32_SOC15(GC, 0, mmATC_L2_CACHE_2M_EDC_CNT);
+
+ sec_count = (data & 0x00006000L) >> 0xd;
+ if (sec_count) {
+ DRM_INFO("Instance[%d]: SubBlock %s, SEC %d\n", i,
+ atc_l2_cache_2m_mems[i], sec_count);
+ err_data->ce_count += sec_count;
+ }
+ }
+
+ for (i = 0; i < 32; i++) {
+ WREG32_SOC15(GC, 0, mmATC_L2_CACHE_4K_EDC_INDEX, i);
+ data = RREG32_SOC15(GC, 0, mmATC_L2_CACHE_4K_EDC_CNT);
+
+ sec_count = (data & 0x00006000L) >> 0xd;
+ if (sec_count) {
+ DRM_INFO("Instance[%d]: SubBlock %s, SEC %d\n", i,
+ atc_l2_cache_4k_mems[i], sec_count);
+ err_data->ce_count += sec_count;
+ }
+
+ ded_count = (data & 0x00018000L) >> 0xf;
+ if (ded_count) {
+ DRM_INFO("Instance[%d]: SubBlock %s, DED %d\n", i,
+ atc_l2_cache_4k_mems[i], ded_count);
+ err_data->ue_count += ded_count;
+ }
+ }
+
+ WREG32_SOC15(GC, 0, mmVM_L2_MEM_ECC_INDEX, 255);
+ WREG32_SOC15(GC, 0, mmVM_L2_WALKER_MEM_ECC_INDEX, 255);
+ WREG32_SOC15(GC, 0, mmATC_L2_CACHE_2M_EDC_INDEX, 255);
+ WREG32_SOC15(GC, 0, mmATC_L2_CACHE_4K_EDC_INDEX, 255);
+
+ return 0;
+}
+
+static int __get_ras_error_count(const struct soc15_reg_entry *reg,
+ uint32_t se_id, uint32_t inst_id, uint32_t value,
+ uint32_t *sec_count, uint32_t *ded_count)
+{
+ uint32_t i;
+ uint32_t sec_cnt, ded_cnt;
+
+ for (i = 0; i < ARRAY_SIZE(ras_subblock_regs); i++) {
+ if(ras_subblock_regs[i].reg_offset != reg->reg_offset ||
+ ras_subblock_regs[i].seg != reg->seg ||
+ ras_subblock_regs[i].inst != reg->inst)
+ continue;
+
+ sec_cnt = (value &
+ ras_subblock_regs[i].sec_count_mask) >>
+ ras_subblock_regs[i].sec_count_shift;
+ if (sec_cnt) {
+ DRM_INFO("GFX SubBlock %s, Instance[%d][%d], SEC %d\n",
+ ras_subblock_regs[i].name,
+ se_id, inst_id,
+ sec_cnt);
+ *sec_count += sec_cnt;
+ }
+
+ ded_cnt = (value &
+ ras_subblock_regs[i].ded_count_mask) >>
+ ras_subblock_regs[i].ded_count_shift;
+ if (ded_cnt) {
+ DRM_INFO("GFX SubBlock %s, Instance[%d][%d], DED %d\n",
+ ras_subblock_regs[i].name,
+ se_id, inst_id,
+ ded_cnt);
+ *ded_count += ded_cnt;
+ }
+ }
+
+ return 0;
+}
+
static int gfx_v9_0_query_ras_error_count(struct amdgpu_device *adev,
void *ras_error_status)
{
struct ras_err_data *err_data = (struct ras_err_data *)ras_error_status;
- uint32_t sec_count, ded_count;
- uint32_t i;
+ uint32_t sec_count = 0, ded_count = 0;
+ uint32_t i, j, k;
uint32_t reg_value;
- uint32_t se_id, instance_id;
if (adev->asic_type != CHIP_VEGA20)
return -EINVAL;
@@ -5794,52 +6159,29 @@ static int gfx_v9_0_query_ras_error_count(struct amdgpu_device *adev,
err_data->ce_count = 0;
mutex_lock(&adev->grbm_idx_mutex);
- for (se_id = 0; se_id < adev->gfx.config.max_shader_engines; se_id++) {
- for (instance_id = 0; instance_id < 256; instance_id++) {
- for (i = 0;
- i < sizeof(gfx_ras_edc_regs) / sizeof(gfx_ras_edc_regs[0]);
- i++) {
- if (se_id != 0 &&
- !gfx_ras_edc_regs[i].per_se_instance)
- continue;
- if (instance_id >= gfx_ras_edc_regs[i].num_instance)
- continue;
-
- gfx_v9_0_select_se_sh(adev, se_id, 0,
- instance_id);
-
- reg_value = RREG32(
- adev->reg_offset[gfx_ras_edc_regs[i].ip]
- [gfx_ras_edc_regs[i].inst]
- [gfx_ras_edc_regs[i].seg] +
- gfx_ras_edc_regs[i].reg_offset);
- sec_count = reg_value &
- gfx_ras_edc_regs[i].sec_count_mask;
- ded_count = reg_value &
- gfx_ras_edc_regs[i].ded_count_mask;
- if (sec_count) {
- DRM_INFO(
- "Instance[%d][%d]: SubBlock %s, SEC %d\n",
- se_id, instance_id,
- gfx_ras_edc_regs[i].name,
- sec_count);
- err_data->ce_count++;
- }
- if (ded_count) {
- DRM_INFO(
- "Instance[%d][%d]: SubBlock %s, DED %d\n",
- se_id, instance_id,
- gfx_ras_edc_regs[i].name,
- ded_count);
- err_data->ue_count++;
- }
+ for (i = 0; i < ARRAY_SIZE(sec_ded_counter_registers); i++) {
+ for (j = 0; j < sec_ded_counter_registers[i].se_num; j++) {
+ for (k = 0; k < sec_ded_counter_registers[i].instance; k++) {
+ gfx_v9_0_select_se_sh(adev, j, 0, k);
+ reg_value =
+ RREG32(SOC15_REG_ENTRY_OFFSET(sec_ded_counter_registers[i]));
+ if (reg_value)
+ __get_ras_error_count(&sec_ded_counter_registers[i],
+ j, k, reg_value,
+ &sec_count, &ded_count);
}
}
}
+
+ err_data->ce_count += sec_count;
+ err_data->ue_count += ded_count;
+
gfx_v9_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
mutex_unlock(&adev->grbm_idx_mutex);
+ gfx_v9_0_query_utc_edc_status(adev, err_data);
+
return 0;
}
diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c
index a9238735d361..b601c6740ef5 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c
@@ -46,21 +46,25 @@ u64 gfxhub_v2_0_get_mc_fb_offset(struct amdgpu_device *adev)
return (u64)RREG32_SOC15(GC, 0, mmGCMC_VM_FB_OFFSET) << 24;
}
-static void gfxhub_v2_0_init_gart_pt_regs(struct amdgpu_device *adev)
+void gfxhub_v2_0_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid,
+ uint64_t page_table_base)
{
- uint64_t value = amdgpu_gmc_pd_addr(adev->gart.bo);
+ /* two registers distance between mmGCVM_CONTEXT0_* to mmGCVM_CONTEXT1_* */
+ int offset = mmGCVM_CONTEXT1_PAGE_TABLE_BASE_ADDR_LO32
+ - mmGCVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32;
+ WREG32_SOC15_OFFSET(GC, 0, mmGCVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32,
+ offset * vmid, lower_32_bits(page_table_base));
- WREG32_SOC15(GC, 0, mmGCVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32,
- lower_32_bits(value));
-
- WREG32_SOC15(GC, 0, mmGCVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_HI32,
- upper_32_bits(value));
+ WREG32_SOC15_OFFSET(GC, 0, mmGCVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_HI32,
+ offset * vmid, upper_32_bits(page_table_base));
}
static void gfxhub_v2_0_init_gart_aperture_regs(struct amdgpu_device *adev)
{
- gfxhub_v2_0_init_gart_pt_regs(adev);
+ uint64_t pt_base = amdgpu_gmc_pd_addr(adev->gart.bo);
+
+ gfxhub_v2_0_setup_vm_pt_regs(adev, 0, pt_base);
WREG32_SOC15(GC, 0, mmGCVM_CONTEXT0_PAGE_TABLE_START_ADDR_LO32,
(u32)(adev->gmc.gart_start >> 12));
diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.h b/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.h
index 06807940748b..392b8cd94fc0 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.h
+++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.h
@@ -31,5 +31,7 @@ void gfxhub_v2_0_set_fault_enable_default(struct amdgpu_device *adev,
bool value);
void gfxhub_v2_0_init(struct amdgpu_device *adev);
u64 gfxhub_v2_0_get_mc_fb_offset(struct amdgpu_device *adev);
+void gfxhub_v2_0_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid,
+ uint64_t page_table_base);
#endif
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c
index 055a8cbf889b..6a73e8d95f0a 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c
@@ -625,7 +625,7 @@ static unsigned gmc_v10_0_get_vbios_fb_size(struct amdgpu_device *adev)
static int gmc_v10_0_sw_init(void *handle)
{
int dma_bits;
- int r, vram_width = 0, vram_type = 0;
+ int r, vram_width = 0, vram_type = 0, vram_vendor = 0;
struct amdgpu_device *adev = (struct amdgpu_device *)handle;
gfxhub_v2_0_init(adev);
@@ -633,13 +633,15 @@ static int gmc_v10_0_sw_init(void *handle)
spin_lock_init(&adev->gmc.invalidate_lock);
- r = amdgpu_atomfirmware_get_vram_info(adev, &vram_width, &vram_type);
+ r = amdgpu_atomfirmware_get_vram_info(adev,
+ &vram_width, &vram_type, &vram_vendor);
if (!amdgpu_emu_mode)
adev->gmc.vram_width = vram_width;
else
adev->gmc.vram_width = 1 * 128; /* numchan * chansize */
adev->gmc.vram_type = vram_type;
+ adev->gmc.vram_vendor = vram_vendor;
switch (adev->asic_type) {
case CHIP_NAVI10:
case CHIP_NAVI14:
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index 185f63ba3296..3784b62201b0 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -929,7 +929,7 @@ static unsigned gmc_v9_0_get_vbios_fb_size(struct amdgpu_device *adev)
static int gmc_v9_0_sw_init(void *handle)
{
int dma_bits;
- int r, vram_width = 0, vram_type = 0;
+ int r, vram_width = 0, vram_type = 0, vram_vendor = 0;
struct amdgpu_device *adev = (struct amdgpu_device *)handle;
gfxhub_v1_0_init(adev);
@@ -940,7 +940,8 @@ static int gmc_v9_0_sw_init(void *handle)
spin_lock_init(&adev->gmc.invalidate_lock);
- r = amdgpu_atomfirmware_get_vram_info(adev, &vram_width, &vram_type);
+ r = amdgpu_atomfirmware_get_vram_info(adev,
+ &vram_width, &vram_type, &vram_vendor);
if (amdgpu_sriov_vf(adev))
/* For Vega10 SR-IOV, vram_width can't be read from ATOM as RAVEN,
* and DF related registers is not readable, seems hardcord is the
@@ -964,6 +965,7 @@ static int gmc_v9_0_sw_init(void *handle)
}
adev->gmc.vram_type = vram_type;
+ adev->gmc.vram_vendor = vram_vendor;
switch (adev->asic_type) {
case CHIP_RAVEN:
adev->num_vmhubs = 2;
@@ -1147,13 +1149,7 @@ static void gmc_v9_0_init_golden_registers(struct amdgpu_device *adev)
*/
static int gmc_v9_0_gart_enable(struct amdgpu_device *adev)
{
- int r, i;
- bool value;
- u32 tmp;
-
- amdgpu_device_program_register_sequence(adev,
- golden_settings_vega10_hdp,
- ARRAY_SIZE(golden_settings_vega10_hdp));
+ int r;
if (adev->gart.bo == NULL) {
dev_err(adev->dev, "No VRAM object for PCIE GART.\n");
@@ -1163,15 +1159,6 @@ static int gmc_v9_0_gart_enable(struct amdgpu_device *adev)
if (r)
return r;
- switch (adev->asic_type) {
- case CHIP_RAVEN:
- /* TODO for renoir */
- mmhub_v1_0_update_power_gating(adev, true);
- break;
- default:
- break;
- }
-
r = gfxhub_v1_0_gart_enable(adev);
if (r)
return r;
@@ -1183,6 +1170,49 @@ static int gmc_v9_0_gart_enable(struct amdgpu_device *adev)
if (r)
return r;
+ DRM_INFO("PCIE GART of %uM enabled (table at 0x%016llX).\n",
+ (unsigned)(adev->gmc.gart_size >> 20),
+ (unsigned long long)amdgpu_bo_gpu_offset(adev->gart.bo));
+ adev->gart.ready = true;
+ return 0;
+}
+
+static int gmc_v9_0_hw_init(void *handle)
+{
+ struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+ bool value;
+ int r, i;
+ u32 tmp;
+
+ /* The sequence of these two function calls matters.*/
+ gmc_v9_0_init_golden_registers(adev);
+
+ if (adev->mode_info.num_crtc) {
+ if (adev->asic_type != CHIP_ARCTURUS) {
+ /* Lockout access through VGA aperture*/
+ WREG32_FIELD15(DCE, 0, VGA_HDP_CONTROL, VGA_MEMORY_DISABLE, 1);
+
+ /* disable VGA render */
+ WREG32_FIELD15(DCE, 0, VGA_RENDER_CONTROL, VGA_VSTATUS_CNTL, 0);
+ }
+ }
+
+ amdgpu_device_program_register_sequence(adev,
+ golden_settings_vega10_hdp,
+ ARRAY_SIZE(golden_settings_vega10_hdp));
+
+ switch (adev->asic_type) {
+ case CHIP_RAVEN:
+ /* TODO for renoir */
+ mmhub_v1_0_update_power_gating(adev, true);
+ break;
+ case CHIP_ARCTURUS:
+ WREG32_FIELD15(HDP, 0, HDP_MMHUB_CNTL, HDP_MMHUB_GCC, 1);
+ break;
+ default:
+ break;
+ }
+
WREG32_FIELD15(HDP, 0, HDP_MISC_CNTL, FLUSH_INVALIDATE_CACHE, 1);
tmp = RREG32_SOC15(HDP, 0, mmHDP_HOST_PATH_CNTL);
@@ -1211,29 +1241,6 @@ static int gmc_v9_0_gart_enable(struct amdgpu_device *adev)
if (adev->umc.funcs && adev->umc.funcs->init_registers)
adev->umc.funcs->init_registers(adev);
- DRM_INFO("PCIE GART of %uM enabled (table at 0x%016llX).\n",
- (unsigned)(adev->gmc.gart_size >> 20),
- (unsigned long long)amdgpu_bo_gpu_offset(adev->gart.bo));
- adev->gart.ready = true;
- return 0;
-}
-
-static int gmc_v9_0_hw_init(void *handle)
-{
- int r;
- struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-
- /* The sequence of these two function calls matters.*/
- gmc_v9_0_init_golden_registers(adev);
-
- if (adev->mode_info.num_crtc) {
- /* Lockout access through VGA aperture*/
- WREG32_FIELD15(DCE, 0, VGA_HDP_CONTROL, VGA_MEMORY_DISABLE, 1);
-
- /* disable VGA render */
- WREG32_FIELD15(DCE, 0, VGA_RENDER_CONTROL, VGA_VSTATUS_CNTL, 0);
- }
-
r = gmc_v9_0_gart_enable(adev);
return r;
diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c
index 4c7e8c64a94e..6965e1e6fa9e 100644
--- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c
@@ -31,7 +31,6 @@
#include "vega10_enum.h"
#include "soc15_common.h"
-#include "amdgpu_ras.h"
#define mmDAGB0_CNTL_MISC2_RV 0x008f
#define mmDAGB0_CNTL_MISC2_RV_BASE_IDX 0
diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c
index 86ed8cb915a8..2eea702de8ee 100644
--- a/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c
@@ -31,20 +31,25 @@
#include "soc15_common.h"
-static void mmhub_v2_0_init_gart_pt_regs(struct amdgpu_device *adev)
+void mmhub_v2_0_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid,
+ uint64_t page_table_base)
{
- uint64_t value = amdgpu_gmc_pd_addr(adev->gart.bo);
+ /* two registers distance between mmMMVM_CONTEXT0_* to mmMMVM_CONTEXT1_* */
+ int offset = mmMMVM_CONTEXT1_PAGE_TABLE_BASE_ADDR_LO32
+ - mmMMVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32;
- WREG32_SOC15(MMHUB, 0, mmMMVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32,
- lower_32_bits(value));
+ WREG32_SOC15_OFFSET(MMHUB, 0, mmMMVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32,
+ offset * vmid, lower_32_bits(page_table_base));
- WREG32_SOC15(MMHUB, 0, mmMMVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_HI32,
- upper_32_bits(value));
+ WREG32_SOC15_OFFSET(MMHUB, 0, mmMMVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_HI32,
+ offset * vmid, upper_32_bits(page_table_base));
}
static void mmhub_v2_0_init_gart_aperture_regs(struct amdgpu_device *adev)
{
- mmhub_v2_0_init_gart_pt_regs(adev);
+ uint64_t pt_base = amdgpu_gmc_pd_addr(adev->gart.bo);
+
+ mmhub_v2_0_setup_vm_pt_regs(adev, 0, pt_base);
WREG32_SOC15(MMHUB, 0, mmMMVM_CONTEXT0_PAGE_TABLE_START_ADDR_LO32,
(u32)(adev->gmc.gart_start >> 12));
diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.h b/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.h
index db16f3ece218..3ea4344f0315 100644
--- a/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.h
+++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.h
@@ -31,5 +31,7 @@ void mmhub_v2_0_init(struct amdgpu_device *adev);
int mmhub_v2_0_set_clockgating(struct amdgpu_device *adev,
enum amd_clockgating_state state);
void mmhub_v2_0_get_clockgating(struct amdgpu_device *adev, u32 *flags);
+void mmhub_v2_0_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid,
+ uint64_t page_table_base);
#endif
diff --git a/drivers/gpu/drm/amd/amdgpu/navi10_reg_init.c b/drivers/gpu/drm/amd/amdgpu/navi10_reg_init.c
index a56c93620e78..88efaecf9f70 100644
--- a/drivers/gpu/drm/amd/amdgpu/navi10_reg_init.c
+++ b/drivers/gpu/drm/amd/amdgpu/navi10_reg_init.c
@@ -24,7 +24,6 @@
#include "nv.h"
#include "soc15_common.h"
-#include "soc15_hw_ip.h"
#include "navi10_ip_offset.h"
int navi10_reg_base_init(struct amdgpu_device *adev)
diff --git a/drivers/gpu/drm/amd/amdgpu/navi12_reg_init.c b/drivers/gpu/drm/amd/amdgpu/navi12_reg_init.c
index cadc7603ca41..a786d159e5e9 100644
--- a/drivers/gpu/drm/amd/amdgpu/navi12_reg_init.c
+++ b/drivers/gpu/drm/amd/amdgpu/navi12_reg_init.c
@@ -24,7 +24,6 @@
#include "nv.h"
#include "soc15_common.h"
-#include "soc15_hw_ip.h"
#include "navi12_ip_offset.h"
int navi12_reg_base_init(struct amdgpu_device *adev)
diff --git a/drivers/gpu/drm/amd/amdgpu/navi14_reg_init.c b/drivers/gpu/drm/amd/amdgpu/navi14_reg_init.c
index 3b5f0f65e096..4ea1e8fbb601 100644
--- a/drivers/gpu/drm/amd/amdgpu/navi14_reg_init.c
+++ b/drivers/gpu/drm/amd/amdgpu/navi14_reg_init.c
@@ -24,7 +24,6 @@
#include "nv.h"
#include "soc15_common.h"
-#include "soc15_hw_ip.h"
#include "navi14_ip_offset.h"
int navi14_reg_base_init(struct amdgpu_device *adev)
diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v2_3.c b/drivers/gpu/drm/amd/amdgpu/nbio_v2_3.c
index a5fa741e4aff..f3a5276ce803 100644
--- a/drivers/gpu/drm/amd/amdgpu/nbio_v2_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/nbio_v2_3.c
@@ -27,11 +27,21 @@
#include "nbio/nbio_2_3_default.h"
#include "nbio/nbio_2_3_offset.h"
#include "nbio/nbio_2_3_sh_mask.h"
+#include <uapi/linux/kfd_ioctl.h>
#define smnPCIE_CONFIG_CNTL 0x11180044
#define smnCPM_CONTROL 0x11180460
#define smnPCIE_CNTL2 0x11180070
+
+static void nbio_v2_3_remap_hdp_registers(struct amdgpu_device *adev)
+{
+ WREG32_SOC15(NBIO, 0, mmREMAP_HDP_MEM_FLUSH_CNTL,
+ adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL);
+ WREG32_SOC15(NBIO, 0, mmREMAP_HDP_REG_FLUSH_CNTL,
+ adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_REG_FLUSH_CNTL);
+}
+
static u32 nbio_v2_3_get_rev_id(struct amdgpu_device *adev)
{
u32 tmp = RREG32_SOC15(NBIO, 0, mmRCC_DEV0_EPF0_STRAP0);
@@ -56,10 +66,9 @@ static void nbio_v2_3_hdp_flush(struct amdgpu_device *adev,
struct amdgpu_ring *ring)
{
if (!ring || !ring->funcs->emit_wreg)
- WREG32_SOC15_NO_KIQ(NBIO, 0, mmBIF_BX_PF_HDP_MEM_COHERENCY_FLUSH_CNTL, 0);
+ WREG32_NO_KIQ((adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2, 0);
else
- amdgpu_ring_emit_wreg(ring, SOC15_REG_OFFSET(
- NBIO, 0, mmBIF_BX_PF_HDP_MEM_COHERENCY_FLUSH_CNTL), 0);
+ amdgpu_ring_emit_wreg(ring, (adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2, 0);
}
static u32 nbio_v2_3_get_memsize(struct amdgpu_device *adev)
@@ -330,4 +339,5 @@ const struct amdgpu_nbio_funcs nbio_v2_3_funcs = {
.ih_control = nbio_v2_3_ih_control,
.init_registers = nbio_v2_3_init_registers,
.detect_hw_virt = nbio_v2_3_detect_hw_virt,
+ .remap_hdp_registers = nbio_v2_3_remap_hdp_registers,
};
diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c b/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c
index 238c2483496a..0db458f9fafc 100644
--- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c
+++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c
@@ -502,6 +502,13 @@ static void nbio_v7_4_query_ras_error_count(struct amdgpu_device *adev,
}
}
+static void nbio_v7_4_enable_doorbell_interrupt(struct amdgpu_device *adev,
+ bool enable)
+{
+ WREG32_FIELD15(NBIO, 0, BIF_DOORBELL_INT_CNTL,
+ DOORBELL_INTERRUPT_DISABLE, enable ? 0 : 1);
+}
+
const struct amdgpu_nbio_funcs nbio_v7_4_funcs = {
.get_hdp_flush_req_offset = nbio_v7_4_get_hdp_flush_req_offset,
.get_hdp_flush_done_offset = nbio_v7_4_get_hdp_flush_done_offset,
@@ -516,6 +523,7 @@ const struct amdgpu_nbio_funcs nbio_v7_4_funcs = {
.enable_doorbell_aperture = nbio_v7_4_enable_doorbell_aperture,
.enable_doorbell_selfring_aperture = nbio_v7_4_enable_doorbell_selfring_aperture,
.ih_doorbell_range = nbio_v7_4_ih_doorbell_range,
+ .enable_doorbell_interrupt = nbio_v7_4_enable_doorbell_interrupt,
.update_medium_grain_clock_gating = nbio_v7_4_update_medium_grain_clock_gating,
.update_medium_grain_light_sleep = nbio_v7_4_update_medium_grain_light_sleep,
.get_clockgating_state = nbio_v7_4_get_clockgating_state,
diff --git a/drivers/gpu/drm/amd/amdgpu/nv.c b/drivers/gpu/drm/amd/amdgpu/nv.c
index fb097aa089da..55a6ed09a953 100644
--- a/drivers/gpu/drm/amd/amdgpu/nv.c
+++ b/drivers/gpu/drm/amd/amdgpu/nv.c
@@ -318,10 +318,12 @@ static int nv_asic_reset(struct amdgpu_device *adev)
struct smu_context *smu = &adev->smu;
if (nv_asic_reset_method(adev) == AMD_RESET_METHOD_BACO) {
- amdgpu_inc_vram_lost(adev);
+ if (!adev->in_suspend)
+ amdgpu_inc_vram_lost(adev);
ret = smu_baco_reset(smu);
} else {
- amdgpu_inc_vram_lost(adev);
+ if (!adev->in_suspend)
+ amdgpu_inc_vram_lost(adev);
ret = nv_asic_mode1_reset(adev);
}
@@ -586,8 +588,11 @@ static const struct amdgpu_asic_funcs nv_asic_funcs =
static int nv_common_early_init(void *handle)
{
+#define MMIO_REG_HOLE_OFFSET (0x80000 - PAGE_SIZE)
struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+ adev->rmmio_remap.reg_offset = MMIO_REG_HOLE_OFFSET;
+ adev->rmmio_remap.bus_addr = adev->rmmio_base + MMIO_REG_HOLE_OFFSET;
adev->smc_rreg = NULL;
adev->smc_wreg = NULL;
adev->pcie_rreg = &nv_pcie_rreg;
@@ -713,6 +718,12 @@ static int nv_common_hw_init(void *handle)
nv_program_aspm(adev);
/* setup nbio registers */
adev->nbio.funcs->init_registers(adev);
+ /* remap HDP registers to a hole in mmio space,
+ * for the purpose of expose those registers
+ * to process space
+ */
+ if (adev->nbio.funcs->remap_hdp_registers)
+ adev->nbio.funcs->remap_hdp_registers(adev);
/* enable the doorbell aperture */
nv_enable_doorbell_aperture(adev, true);
diff --git a/drivers/gpu/drm/amd/amdgpu/nvd.h b/drivers/gpu/drm/amd/amdgpu/nvd.h
index 1de984647dbb..f3d8771ebed4 100644
--- a/drivers/gpu/drm/amd/amdgpu/nvd.h
+++ b/drivers/gpu/drm/amd/amdgpu/nvd.h
@@ -306,6 +306,7 @@
#define PACKET3_GET_LOD_STATS 0x8E
#define PACKET3_DRAW_MULTI_PREAMBLE 0x8F
#define PACKET3_FRAME_CONTROL 0x90
+# define FRAME_TMZ (1 << 0)
# define FRAME_CMD(x) ((x) << 28)
/*
* x=0: tmz_begin
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c
index 7e6e573dc1a9..1a3ba0bc1cda 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c
@@ -958,6 +958,7 @@ static uint32_t sdma_v4_0_rb_cntl(struct amdgpu_ring *ring, uint32_t rb_cntl)
/* Set ring buffer size in dwords */
uint32_t rb_bufsz = order_base_2(ring->ring_size / 4);
+ barrier(); /* work around https://bugs.llvm.org/show_bug.cgi?id=42576 */
rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, RB_SIZE, rb_bufsz);
#ifdef __BIG_ENDIAN
rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, RB_SWAP_ENABLE, 1);
@@ -1789,7 +1790,7 @@ static int sdma_v4_0_hw_init(void *handle)
if ((adev->asic_type == CHIP_RAVEN && adev->powerplay.pp_funcs &&
adev->powerplay.pp_funcs->set_powergating_by_smu) ||
- adev->asic_type == CHIP_RENOIR)
+ (adev->asic_type == CHIP_RENOIR && !adev->in_gpu_reset))
amdgpu_dpm_set_powergating_by_smu(adev, AMD_IP_BLOCK_TYPE_SDMA, false);
if (!amdgpu_sriov_vf(adev))
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c
index ad5c3566337c..3460c00f3eaa 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c
@@ -1126,7 +1126,7 @@ static void sdma_v5_0_ring_emit_pipeline_sync(struct amdgpu_ring *ring)
amdgpu_ring_write(ring, addr & 0xfffffffc);
amdgpu_ring_write(ring, upper_32_bits(addr) & 0xffffffff);
amdgpu_ring_write(ring, seq); /* reference */
- amdgpu_ring_write(ring, 0xfffffff); /* mask */
+ amdgpu_ring_write(ring, 0xffffffff); /* mask */
amdgpu_ring_write(ring, SDMA_PKT_POLL_REGMEM_DW5_RETRY_COUNT(0xfff) |
SDMA_PKT_POLL_REGMEM_DW5_INTERVAL(4)); /* retry count, poll interval */
}
diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c b/drivers/gpu/drm/amd/amdgpu/soc15.c
index 14c7fc141322..82b5bc4ddf9b 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc15.c
+++ b/drivers/gpu/drm/amd/amdgpu/soc15.c
@@ -492,10 +492,15 @@ static int soc15_asic_baco_reset(struct amdgpu_device *adev)
{
void *pp_handle = adev->powerplay.pp_handle;
const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs;
+ struct amdgpu_ras *ras = amdgpu_ras_get_context(adev);
if (!pp_funcs ||!pp_funcs->get_asic_baco_state ||!pp_funcs->set_asic_baco_state)
return -ENOENT;
+ /* avoid NBIF got stuck when do RAS recovery in BACO reset */
+ if (ras && ras->supported)
+ adev->nbio.funcs->enable_doorbell_interrupt(adev, false);
+
/* enter BACO state */
if (pp_funcs->set_asic_baco_state(pp_handle, 1))
return -EIO;
@@ -504,6 +509,10 @@ static int soc15_asic_baco_reset(struct amdgpu_device *adev)
if (pp_funcs->set_asic_baco_state(pp_handle, 0))
return -EIO;
+ /* re-enable doorbell interrupt after BACO exit */
+ if (ras && ras->supported)
+ adev->nbio.funcs->enable_doorbell_interrupt(adev, true);
+
dev_info(adev->dev, "GPU BACO reset\n");
adev->in_baco_reset = 1;
@@ -563,12 +572,14 @@ static int soc15_asic_reset(struct amdgpu_device *adev)
{
switch (soc15_asic_reset_method(adev)) {
case AMD_RESET_METHOD_BACO:
- amdgpu_inc_vram_lost(adev);
+ if (!adev->in_suspend)
+ amdgpu_inc_vram_lost(adev);
return soc15_asic_baco_reset(adev);
case AMD_RESET_METHOD_MODE2:
return soc15_mode2_reset(adev);
default:
- amdgpu_inc_vram_lost(adev);
+ if (!adev->in_suspend)
+ amdgpu_inc_vram_lost(adev);
return soc15_asic_mode1_reset(adev);
}
}
@@ -757,14 +768,24 @@ int soc15_set_ip_blocks(struct amdgpu_device *adev)
case CHIP_ARCTURUS:
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);
- if (likely(adev->firmware.load_type == AMDGPU_FW_LOAD_PSP))
- amdgpu_device_ip_block_add(adev, &psp_v11_0_ip_block);
+
+ if (amdgpu_sriov_vf(adev)) {
+ if (likely(adev->firmware.load_type == AMDGPU_FW_LOAD_PSP))
+ amdgpu_device_ip_block_add(adev, &psp_v11_0_ip_block);
+ amdgpu_device_ip_block_add(adev, &vega10_ih_ip_block);
+ } else {
+ amdgpu_device_ip_block_add(adev, &vega10_ih_ip_block);
+ if (likely(adev->firmware.load_type == AMDGPU_FW_LOAD_PSP))
+ amdgpu_device_ip_block_add(adev, &psp_v11_0_ip_block);
+ }
+
if (adev->enable_virtual_display || amdgpu_sriov_vf(adev))
amdgpu_device_ip_block_add(adev, &dce_virtual_ip_block);
amdgpu_device_ip_block_add(adev, &gfx_v9_0_ip_block);
amdgpu_device_ip_block_add(adev, &sdma_v4_0_ip_block);
- amdgpu_device_ip_block_add(adev, &smu_v11_0_ip_block);
+ if (!amdgpu_sriov_vf(adev))
+ amdgpu_device_ip_block_add(adev, &smu_v11_0_ip_block);
+
if (unlikely(adev->firmware.load_type == AMDGPU_FW_LOAD_DIRECT))
amdgpu_device_ip_block_add(adev, &vcn_v2_5_ip_block);
break;
@@ -783,8 +804,6 @@ int soc15_set_ip_blocks(struct amdgpu_device *adev)
#if defined(CONFIG_DRM_AMD_DC)
else if (amdgpu_device_has_dc_support(adev))
amdgpu_device_ip_block_add(adev, &dm_ip_block);
-#else
-# warning "Enable CONFIG_DRM_AMD_DC for display support on SOC15."
#endif
amdgpu_device_ip_block_add(adev, &vcn_v2_0_ip_block);
break;
@@ -1169,7 +1188,8 @@ static int soc15_common_early_init(void *handle)
AMD_CG_SUPPORT_SDMA_MGCG |
AMD_CG_SUPPORT_SDMA_LS |
AMD_CG_SUPPORT_MC_MGCG |
- AMD_CG_SUPPORT_MC_LS;
+ AMD_CG_SUPPORT_MC_LS |
+ AMD_CG_SUPPORT_IH_CG;
adev->pg_flags = 0;
adev->external_rev_id = adev->rev_id + 0x32;
break;
diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.h b/drivers/gpu/drm/amd/amdgpu/soc15.h
index a3dde0c31f57..9af6c6ffbfa2 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc15.h
+++ b/drivers/gpu/drm/amd/amdgpu/soc15.h
@@ -67,6 +67,8 @@ struct soc15_allowed_register_entry {
#define SOC15_REG_GOLDEN_VALUE(ip, inst, reg, and_mask, or_mask) \
{ ip##_HWIP, inst, reg##_BASE_IDX, reg, and_mask, or_mask }
+#define SOC15_REG_FIELD(reg, field) reg##__##field##_MASK, reg##__##field##__SHIFT
+
void soc15_grbm_select(struct amdgpu_device *adev,
u32 me, u32 pipe, u32 queue, u32 vmid);
int soc15_set_ip_blocks(struct amdgpu_device *adev);
diff --git a/drivers/gpu/drm/amd/amdgpu/soc15d.h b/drivers/gpu/drm/amd/amdgpu/soc15d.h
index edfe50821cd9..295d68c5811d 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc15d.h
+++ b/drivers/gpu/drm/amd/amdgpu/soc15d.h
@@ -286,6 +286,7 @@
#define PACKET3_WAIT_ON_DE_COUNTER_DIFF 0x88
#define PACKET3_SWITCH_BUFFER 0x8B
#define PACKET3_FRAME_CONTROL 0x90
+# define FRAME_TMZ (1 << 0)
# define FRAME_CMD(x) ((x) << 28)
/*
* x=0: tmz_begin
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c
index 07a2f36ea7ce..b23362102e51 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c
@@ -202,7 +202,6 @@ static int vcn_v1_0_hw_init(void *handle)
for (i = 0; i < adev->vcn.num_enc_rings; ++i) {
ring = &adev->vcn.inst->ring_enc[i];
- ring->sched.ready = true;
r = amdgpu_ring_test_helper(ring);
if (r)
goto done;
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c
index 5a590064bfff..16f192f6c967 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c
@@ -247,30 +247,21 @@ static int vcn_v2_0_hw_init(void *handle)
adev->nbio.funcs->vcn_doorbell_range(adev, ring->use_doorbell,
ring->doorbell_index, 0);
- ring->sched.ready = true;
- r = amdgpu_ring_test_ring(ring);
- if (r) {
- ring->sched.ready = false;
+ r = amdgpu_ring_test_helper(ring);
+ if (r)
goto done;
- }
for (i = 0; i < adev->vcn.num_enc_rings; ++i) {
ring = &adev->vcn.inst->ring_enc[i];
- ring->sched.ready = true;
- r = amdgpu_ring_test_ring(ring);
- if (r) {
- ring->sched.ready = false;
+ r = amdgpu_ring_test_helper(ring);
+ if (r)
goto done;
- }
}
ring = &adev->vcn.inst->ring_jpeg;
- ring->sched.ready = true;
- r = amdgpu_ring_test_ring(ring);
- if (r) {
- ring->sched.ready = false;
+ r = amdgpu_ring_test_helper(ring);
+ if (r)
goto done;
- }
done:
if (!r)
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c b/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c
index 22e4644521dc..d270df892223 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c
@@ -25,6 +25,7 @@
#include <drm/drmP.h>
#include "amdgpu.h"
#include "amdgpu_vcn.h"
+#include "amdgpu_pm.h"
#include "soc15.h"
#include "soc15d.h"
#include "vcn_v2_0.h"
@@ -258,29 +259,24 @@ static int vcn_v2_5_hw_init(void *handle)
adev->nbio.funcs->vcn_doorbell_range(adev, ring->use_doorbell,
ring->doorbell_index, j);
- r = amdgpu_ring_test_ring(ring);
- if (r) {
- ring->sched.ready = false;
+ r = amdgpu_ring_test_helper(ring);
+ if (r)
goto done;
- }
for (i = 0; i < adev->vcn.num_enc_rings; ++i) {
ring = &adev->vcn.inst[j].ring_enc[i];
+ /* disable encode rings till the robustness of the FW */
ring->sched.ready = false;
continue;
- r = amdgpu_ring_test_ring(ring);
- if (r) {
- ring->sched.ready = false;
+ r = amdgpu_ring_test_helper(ring);
+ if (r)
goto done;
- }
}
ring = &adev->vcn.inst[j].ring_jpeg;
- r = amdgpu_ring_test_ring(ring);
- if (r) {
- ring->sched.ready = false;
+ r = amdgpu_ring_test_helper(ring);
+ if (r)
goto done;
- }
}
done:
if (!r)
@@ -714,6 +710,9 @@ static int vcn_v2_5_start(struct amdgpu_device *adev)
uint32_t rb_bufsz, tmp;
int i, j, k, r;
+ if (adev->pm.dpm_enabled)
+ amdgpu_dpm_enable_uvd(adev, true);
+
for (i = 0; i < adev->vcn.num_vcn_inst; ++i) {
if (adev->vcn.harvest_config & (1 << i))
continue;
@@ -944,6 +943,9 @@ static int vcn_v2_5_stop(struct amdgpu_device *adev)
~UVD_POWER_STATUS__UVD_POWER_STATUS_MASK);
}
+ if (adev->pm.dpm_enabled)
+ amdgpu_dpm_enable_uvd(adev, false);
+
return 0;
}
diff --git a/drivers/gpu/drm/amd/amdgpu/vega10_ih.c b/drivers/gpu/drm/amd/amdgpu/vega10_ih.c
index d7135e5871d4..1aebf7dab4e5 100644
--- a/drivers/gpu/drm/amd/amdgpu/vega10_ih.c
+++ b/drivers/gpu/drm/amd/amdgpu/vega10_ih.c
@@ -673,10 +673,49 @@ static int vega10_ih_soft_reset(void *handle)
return 0;
}
+static void vega10_ih_update_clockgating_state(struct amdgpu_device *adev,
+ bool enable)
+{
+ uint32_t data, def, field_val;
+
+ if (adev->cg_flags & AMD_CG_SUPPORT_IH_CG) {
+ def = data = RREG32_SOC15(OSSSYS, 0, mmIH_CLK_CTRL);
+ field_val = enable ? 0 : 1;
+ /**
+ * Vega10 does not have IH_RETRY_INT_CAM_MEM_CLK_SOFT_OVERRIDE
+ * and IH_BUFFER_MEM_CLK_SOFT_OVERRIDE field.
+ */
+ if (adev->asic_type > CHIP_VEGA10) {
+ data = REG_SET_FIELD(data, IH_CLK_CTRL,
+ IH_RETRY_INT_CAM_MEM_CLK_SOFT_OVERRIDE, field_val);
+ data = REG_SET_FIELD(data, IH_CLK_CTRL,
+ IH_BUFFER_MEM_CLK_SOFT_OVERRIDE, field_val);
+ }
+
+ data = REG_SET_FIELD(data, IH_CLK_CTRL,
+ DBUS_MUX_CLK_SOFT_OVERRIDE, field_val);
+ data = REG_SET_FIELD(data, IH_CLK_CTRL,
+ OSSSYS_SHARE_CLK_SOFT_OVERRIDE, field_val);
+ data = REG_SET_FIELD(data, IH_CLK_CTRL,
+ LIMIT_SMN_CLK_SOFT_OVERRIDE, field_val);
+ data = REG_SET_FIELD(data, IH_CLK_CTRL,
+ DYN_CLK_SOFT_OVERRIDE, field_val);
+ data = REG_SET_FIELD(data, IH_CLK_CTRL,
+ REG_CLK_SOFT_OVERRIDE, field_val);
+ if (def != data)
+ WREG32_SOC15(OSSSYS, 0, mmIH_CLK_CTRL, data);
+ }
+}
+
static int vega10_ih_set_clockgating_state(void *handle,
enum amd_clockgating_state state)
{
+ struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+
+ vega10_ih_update_clockgating_state(adev,
+ state == AMD_CG_STATE_GATE ? true : false);
return 0;
+
}
static int vega10_ih_set_powergating_state(void *handle,
diff --git a/drivers/gpu/drm/amd/amdgpu/vega10_reg_init.c b/drivers/gpu/drm/amd/amdgpu/vega10_reg_init.c
index bd0580334f83..6b52a539d51b 100644
--- a/drivers/gpu/drm/amd/amdgpu/vega10_reg_init.c
+++ b/drivers/gpu/drm/amd/amdgpu/vega10_reg_init.c
@@ -24,7 +24,6 @@
#include "soc15.h"
#include "soc15_common.h"
-#include "soc15_hw_ip.h"
#include "vega10_ip_offset.h"
int vega10_reg_base_init(struct amdgpu_device *adev)
diff --git a/drivers/gpu/drm/amd/amdgpu/vega20_reg_init.c b/drivers/gpu/drm/amd/amdgpu/vega20_reg_init.c
index 587e33f5dcce..556f854e3551 100644
--- a/drivers/gpu/drm/amd/amdgpu/vega20_reg_init.c
+++ b/drivers/gpu/drm/amd/amdgpu/vega20_reg_init.c
@@ -24,7 +24,6 @@
#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)
diff --git a/drivers/gpu/drm/amd/amdkcl/kcl_device_cgroup.c b/drivers/gpu/drm/amd/amdkcl/kcl_device_cgroup.c
index 1cf144110568..30c14978d3dd 100644
--- a/drivers/gpu/drm/amd/amdkcl/kcl_device_cgroup.c
+++ b/drivers/gpu/drm/amd/amdkcl/kcl_device_cgroup.c
@@ -2,8 +2,8 @@
#include <kcl/kcl_device_cgroup.h>
#include "kcl_common.h"
-#if (defined(CONFIG_CGROUP_DEVICE) || defined(CONFIG_CGROUP_BPF)) && \
- !defined(HAVE___DEVCGROUP_CHECK_PERMISSION)
+#if defined(CONFIG_CGROUP_DEVICE) && \
+ !defined(HAVE_DEVCGROUP_CHECK_PERMISSION)
/*
__devcgroup_check_permission is introduced in v3.6-6796-gad676077a2ae
prototype is:
@@ -24,8 +24,8 @@ EXPORT_SYMBOL(__kcl_devcgroup_check_permission);
#endif
void amdkcl_dev_cgroup_init(void)
{
-#if (defined(CONFIG_CGROUP_DEVICE) || defined(CONFIG_CGROUP_BPF)) && \
- !defined(HAVE___DEVCGROUP_CHECK_PERMISSION)
+#if defined(CONFIG_CGROUP_DEVICE) && \
+ !defined(HAVE_DEVCGROUP_CHECK_PERMISSION)
__kcl_devcgroup_check_permission = amdkcl_fp_setup("__devcgroup_check_permission", NULL);
#endif
}
diff --git a/drivers/gpu/drm/amd/amdkcl/kcl_pci.c b/drivers/gpu/drm/amd/amdkcl/kcl_pci.c
index 310309c96d2c..373798d9999c 100644
--- a/drivers/gpu/drm/amd/amdkcl/kcl_pci.c
+++ b/drivers/gpu/drm/amd/amdkcl/kcl_pci.c
@@ -186,7 +186,7 @@ void amdkcl_pci_init(void)
_kcl_pcie_get_speed_cap = amdkcl_fp_setup("pcie_get_speed_cap", pcie_get_speed_cap);
_kcl_pcie_get_width_cap = amdkcl_fp_setup("pcie_get_width_cap", pcie_get_width_cap);
#endif
- _kcl_pcie_link_speed = (const unsigned char *) amdkcl_fp_setup("pcie_link_speed", _kcl_pcie_link_speed);
+ _kcl_pcie_link_speed = (const unsigned char *) amdkcl_fp_setup("pcie_link_speed", _kcl_pcie_link_speed_stub);
}
#if !defined(HAVE_PCIE_BANDWIDTH_AVAILABLE)
diff --git a/drivers/gpu/drm/amd/amdkfd/cik_event_interrupt.c b/drivers/gpu/drm/amd/amdkfd/cik_event_interrupt.c
index 177d1e5329a5..9f59ba93cfe0 100644
--- a/drivers/gpu/drm/amd/amdkfd/cik_event_interrupt.c
+++ b/drivers/gpu/drm/amd/amdkfd/cik_event_interrupt.c
@@ -33,7 +33,9 @@ static bool cik_event_interrupt_isr(struct kfd_dev *dev,
const struct cik_ih_ring_entry *ihre =
(const struct cik_ih_ring_entry *)ih_ring_entry;
const struct kfd2kgd_calls *f2g = dev->kfd2kgd;
- unsigned int vmid, pasid;
+ unsigned int vmid;
+ uint16_t pasid;
+ bool ret;
/* This workaround is due to HW/FW limitation on Hawaii that
* VMID and PASID are not written into ih_ring_entry
@@ -48,13 +50,13 @@ static bool cik_event_interrupt_isr(struct kfd_dev *dev,
*tmp_ihre = *ihre;
vmid = f2g->read_vmid_from_vmfault_reg(dev->kgd);
- pasid = f2g->get_atc_vmid_pasid_mapping_pasid(dev->kgd, vmid);
+ ret = f2g->get_atc_vmid_pasid_mapping_info(dev->kgd, vmid, &pasid);
tmp_ihre->ring_id &= 0x000000ff;
tmp_ihre->ring_id |= vmid << 8;
tmp_ihre->ring_id |= pasid << 16;
- return (pasid != 0) &&
+ return ret && (pasid != 0) &&
vmid >= dev->vm_info.first_vmid_kfd &&
vmid <= dev->vm_info.last_vmid_kfd;
}
diff --git a/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler.h b/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler.h
index 901fe3590165..d3400da6ab64 100644
--- a/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler.h
+++ b/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler.h
@@ -905,7 +905,7 @@ static const uint32_t cwsr_trap_gfx10_hex[] = {
0x7a5d0000, 0x807c817c,
0x807aff7a, 0x00000080,
0xbf0a717c, 0xbf85fff8,
- 0xbf820141, 0xbef4037e,
+ 0xbf820142, 0xbef4037e,
0x8775ff7f, 0x0000ffff,
0x8875ff75, 0x00040000,
0xbef60380, 0xbef703ff,
@@ -967,7 +967,7 @@ static const uint32_t cwsr_trap_gfx10_hex[] = {
0x725d0000, 0xe0304080,
0x725d0100, 0xe0304100,
0x725d0200, 0xe0304180,
- 0x725d0300, 0xbf820031,
+ 0x725d0300, 0xbf820032,
0xbef603ff, 0x01000000,
0xbef20378, 0x8078ff78,
0x00000400, 0xbefc0384,
@@ -992,83 +992,84 @@ static const uint32_t cwsr_trap_gfx10_hex[] = {
0x725d0000, 0xe0304100,
0x725d0100, 0xe0304200,
0x725d0200, 0xe0304300,
- 0x725d0300, 0xb9782a05,
- 0x80788178, 0x907c9973,
- 0x877c817c, 0xbf06817c,
- 0xbf850002, 0x8f788978,
- 0xbf820001, 0x8f788a78,
- 0xb9721e06, 0x8f728a72,
- 0x80787278, 0x8078ff78,
- 0x00000200, 0x80f8ff78,
- 0x00000050, 0xbef603ff,
- 0x01000000, 0xbefc03ff,
- 0x0000006c, 0x80f89078,
- 0xf429003a, 0xf0000000,
- 0xbf8cc07f, 0x80fc847c,
- 0xbf800000, 0xbe803100,
- 0xbe823102, 0x80f8a078,
- 0xf42d003a, 0xf0000000,
- 0xbf8cc07f, 0x80fc887c,
- 0xbf800000, 0xbe803100,
- 0xbe823102, 0xbe843104,
- 0xbe863106, 0x80f8c078,
- 0xf431003a, 0xf0000000,
- 0xbf8cc07f, 0x80fc907c,
- 0xbf800000, 0xbe803100,
- 0xbe823102, 0xbe843104,
- 0xbe863106, 0xbe883108,
- 0xbe8a310a, 0xbe8c310c,
- 0xbe8e310e, 0xbf06807c,
- 0xbf84fff0, 0xb9782a05,
- 0x80788178, 0x907c9973,
- 0x877c817c, 0xbf06817c,
- 0xbf850002, 0x8f788978,
- 0xbf820001, 0x8f788a78,
- 0xb9721e06, 0x8f728a72,
- 0x80787278, 0x8078ff78,
- 0x00000200, 0xbef603ff,
- 0x01000000, 0xf4211bfa,
+ 0x725d0300, 0xbf8c3f70,
+ 0xb9782a05, 0x80788178,
+ 0x907c9973, 0x877c817c,
+ 0xbf06817c, 0xbf850002,
+ 0x8f788978, 0xbf820001,
+ 0x8f788a78, 0xb9721e06,
+ 0x8f728a72, 0x80787278,
+ 0x8078ff78, 0x00000200,
+ 0x80f8ff78, 0x00000050,
+ 0xbef603ff, 0x01000000,
+ 0xbefc03ff, 0x0000006c,
+ 0x80f89078, 0xf429003a,
+ 0xf0000000, 0xbf8cc07f,
+ 0x80fc847c, 0xbf800000,
+ 0xbe803100, 0xbe823102,
+ 0x80f8a078, 0xf42d003a,
+ 0xf0000000, 0xbf8cc07f,
+ 0x80fc887c, 0xbf800000,
+ 0xbe803100, 0xbe823102,
+ 0xbe843104, 0xbe863106,
+ 0x80f8c078, 0xf431003a,
+ 0xf0000000, 0xbf8cc07f,
+ 0x80fc907c, 0xbf800000,
+ 0xbe803100, 0xbe823102,
+ 0xbe843104, 0xbe863106,
+ 0xbe883108, 0xbe8a310a,
+ 0xbe8c310c, 0xbe8e310e,
+ 0xbf06807c, 0xbf84fff0,
+ 0xb9782a05, 0x80788178,
+ 0x907c9973, 0x877c817c,
+ 0xbf06817c, 0xbf850002,
+ 0x8f788978, 0xbf820001,
+ 0x8f788a78, 0xb9721e06,
+ 0x8f728a72, 0x80787278,
+ 0x8078ff78, 0x00000200,
+ 0xbef603ff, 0x01000000,
+ 0xf4211bfa, 0xf0000000,
+ 0x80788478, 0xf4211b3a,
0xf0000000, 0x80788478,
- 0xf4211b3a, 0xf0000000,
- 0x80788478, 0xf4211b7a,
+ 0xf4211b7a, 0xf0000000,
+ 0x80788478, 0xf4211eba,
0xf0000000, 0x80788478,
- 0xf4211eba, 0xf0000000,
- 0x80788478, 0xf4211efa,
+ 0xf4211efa, 0xf0000000,
+ 0x80788478, 0xf4211c3a,
0xf0000000, 0x80788478,
- 0xf4211c3a, 0xf0000000,
- 0x80788478, 0xf4211c7a,
+ 0xf4211c7a, 0xf0000000,
+ 0x80788478, 0xf4211e7a,
0xf0000000, 0x80788478,
- 0xf4211e7a, 0xf0000000,
- 0x80788478, 0xf4211cfa,
+ 0xf4211cfa, 0xf0000000,
+ 0x80788478, 0xf4211bba,
0xf0000000, 0x80788478,
+ 0xbf8cc07f, 0xb9eef814,
0xf4211bba, 0xf0000000,
0x80788478, 0xbf8cc07f,
- 0xb9eef814, 0xf4211bba,
- 0xf0000000, 0x80788478,
- 0xbf8cc07f, 0xb9eef815,
- 0xbef2036d, 0x876dff72,
- 0x0000ffff, 0xbefc036f,
- 0xbefe037a, 0xbeff037b,
- 0x876f71ff, 0x000003ff,
- 0xb9ef4803, 0xb9f9f816,
- 0x876f71ff, 0xfffff800,
- 0x906f8b6f, 0xb9efa2c3,
- 0xb9f3f801, 0x876fff72,
- 0xfc000000, 0x906f9a6f,
- 0x8f6f906f, 0xbef30380,
+ 0xb9eef815, 0xbef2036d,
+ 0x876dff72, 0x0000ffff,
+ 0xbefc036f, 0xbefe037a,
+ 0xbeff037b, 0x876f71ff,
+ 0x000003ff, 0xb9ef4803,
+ 0xb9f9f816, 0x876f71ff,
+ 0xfffff800, 0x906f8b6f,
+ 0xb9efa2c3, 0xb9f3f801,
+ 0x876fff72, 0xfc000000,
+ 0x906f9a6f, 0x8f6f906f,
+ 0xbef30380, 0x88736f73,
+ 0x876fff72, 0x02000000,
+ 0x906f996f, 0x8f6f8f6f,
0x88736f73, 0x876fff72,
- 0x02000000, 0x906f996f,
- 0x8f6f8f6f, 0x88736f73,
- 0x876fff72, 0x01000000,
- 0x906f986f, 0x8f6f996f,
- 0x88736f73, 0x876fff70,
- 0x00800000, 0x906f976f,
- 0xb9f3f807, 0x87fe7e7e,
- 0x87ea6a6a, 0xb9f0f802,
- 0xbf8a0000, 0xbe80226c,
- 0xbf810000, 0xbf9f0000,
+ 0x01000000, 0x906f986f,
+ 0x8f6f996f, 0x88736f73,
+ 0x876fff70, 0x00800000,
+ 0x906f976f, 0xb9f3f807,
+ 0x87fe7e7e, 0x87ea6a6a,
+ 0xb9f0f802, 0xbf8a0000,
+ 0xbe80226c, 0xbf810000,
0xbf9f0000, 0xbf9f0000,
0xbf9f0000, 0xbf9f0000,
+ 0xbf9f0000, 0x00000000,
};
static const uint32_t cwsr_trap_arcturus_hex[] = {
0xbf820001, 0xbf8202c4,
diff --git a/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler_gfx10.asm b/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler_gfx10.asm
index cdaa523ce6be..4433bda2ce25 100644
--- a/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler_gfx10.asm
+++ b/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler_gfx10.asm
@@ -758,6 +758,7 @@ L_RESTORE_V0:
buffer_load_dword v1, v0, s_restore_buf_rsrc0, s_restore_mem_offset_save slc:1 glc:1 offset:256
buffer_load_dword v2, v0, s_restore_buf_rsrc0, s_restore_mem_offset_save slc:1 glc:1 offset:256*2
buffer_load_dword v3, v0, s_restore_buf_rsrc0, s_restore_mem_offset_save slc:1 glc:1 offset:256*3
+ s_waitcnt vmcnt(0)
/* restore SGPRs */
//will be 2+8+16*6
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
index bf199ca8aee1..f95741f8bcb0 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
@@ -310,7 +310,7 @@ static int kfd_ioctl_create_queue(struct file *filep, struct kfd_process *p,
goto err_bind_process;
}
- pr_debug("Creating queue for PASID %d on gpu 0x%x\n",
+ pr_debug("Creating queue for PASID 0x%x on gpu 0x%x\n",
p->pasid,
dev->id);
@@ -360,7 +360,7 @@ static int kfd_ioctl_destroy_queue(struct file *filp, struct kfd_process *p,
int retval;
struct kfd_ioctl_destroy_queue_args *args = data;
- pr_debug("Destroying queue id %d for pasid %d\n",
+ pr_debug("Destroying queue id %d for pasid 0x%x\n",
args->queue_id,
p->pasid);
@@ -406,7 +406,7 @@ static int kfd_ioctl_update_queue(struct file *filp, struct kfd_process *p,
properties.queue_percent = args->queue_percentage;
properties.priority = args->queue_priority;
- pr_debug("Updating queue id %d for pasid %d\n",
+ pr_debug("Updating queue id %d for pasid 0x%x\n",
args->queue_id, p->pasid);
mutex_lock(&p->mutex);
@@ -863,7 +863,7 @@ static int kfd_ioctl_get_process_apertures(struct file *filp,
struct kfd_process_device_apertures *pAperture;
struct kfd_process_device *pdd;
- dev_dbg(kfd_device, "get apertures for PASID %d", p->pasid);
+ dev_dbg(kfd_device, "get apertures for PASID 0x%x", p->pasid);
args->num_of_nodes = 0;
@@ -921,7 +921,7 @@ static int kfd_ioctl_get_process_apertures_new(struct file *filp,
uint32_t nodes = 0;
int ret;
- dev_dbg(kfd_device, "get apertures for PASID %d", p->pasid);
+ dev_dbg(kfd_device, "get apertures for PASID 0x%x", p->pasid);
if (args->num_of_nodes == 0) {
/* Return number of nodes, so that user space can alloacate
@@ -2998,7 +2998,7 @@ static long kfd_ioctl(struct file *filep, unsigned int cmd, unsigned long arg)
} else
goto err_i1;
- dev_dbg(kfd_device, "ioctl cmd 0x%x (#%d), arg 0x%lx\n", cmd, nr, arg);
+ dev_dbg(kfd_device, "ioctl cmd 0x%x (#0x%x), arg 0x%lx\n", cmd, nr, arg);
process = kfd_get_process(current);
if (IS_ERR(process)) {
@@ -3053,7 +3053,8 @@ err_i1:
kfree(kdata);
if (retcode)
- dev_dbg(kfd_device, "ret = %d\n", retcode);
+ dev_dbg(kfd_device, "ioctl cmd (#0x%x), arg 0x%lx, ret = %d\n",
+ nr, arg, retcode);
return retcode;
}
@@ -3074,7 +3075,7 @@ static int kfd_mmio_mmap(struct kfd_dev *dev, struct kfd_process *process,
vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot);
- pr_debug("Process %d mapping mmio page\n"
+ pr_debug("pasid 0x%x mapping mmio page\n"
" target user address == 0x%08llX\n"
" physical address == 0x%08llX\n"
" vm_flags == 0x%04lX\n"
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_crat.c b/drivers/gpu/drm/amd/amdkfd/kfd_crat.c
index 9577bb82e949..64eb065f8278 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_crat.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_crat.c
@@ -676,6 +676,7 @@ static int kfd_fill_gpu_cache_info(struct kfd_dev *kdev,
num_of_cache_types = ARRAY_SIZE(renoir_cache_info);
break;
case CHIP_NAVI10:
+ case CHIP_NAVI12:
case CHIP_NAVI14:
pcache_info = navi10_cache_info;
num_of_cache_types = ARRAY_SIZE(navi10_cache_info);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_dbgdev.c b/drivers/gpu/drm/amd/amdkfd/kfd_dbgdev.c
index 3635e0b4b3b7..1eb0c2bedcd9 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_dbgdev.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_dbgdev.c
@@ -775,6 +775,7 @@ int dbgdev_wave_reset_wavefronts(struct kfd_dev *dev, struct kfd_process *p)
{
int status = 0;
unsigned int vmid;
+ uint16_t queried_pasid;
union SQ_CMD_BITS reg_sq_cmd;
union GRBM_GFX_INDEX_BITS reg_gfx_index;
struct kfd_process_device *pdd;
@@ -796,19 +797,18 @@ int dbgdev_wave_reset_wavefronts(struct kfd_dev *dev, struct kfd_process *p)
*/
for (vmid = first_vmid_to_scan; vmid <= last_vmid_to_scan; vmid++) {
- if (dev->kfd2kgd->get_atc_vmid_pasid_mapping_valid
- (dev->kgd, vmid)) {
- if (dev->kfd2kgd->get_atc_vmid_pasid_mapping_pasid
- (dev->kgd, vmid) == p->pasid) {
- pr_debug("Killing wave fronts of vmid %d and pasid %d\n",
- vmid, p->pasid);
- break;
- }
+ status = dev->kfd2kgd->get_atc_vmid_pasid_mapping_info
+ (dev->kgd, vmid, &queried_pasid);
+
+ if (status && queried_pasid == p->pasid) {
+ pr_debug("Killing wave fronts of vmid %d and pasid 0x%x\n",
+ vmid, p->pasid);
+ break;
}
}
if (vmid > last_vmid_to_scan) {
- pr_err("Didn't find vmid for pasid %d\n", p->pasid);
+ pr_err("Didn't find vmid for pasid 0x%x\n", p->pasid);
return -EFAULT;
}
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_dbgmgr.c b/drivers/gpu/drm/amd/amdkfd/kfd_dbgmgr.c
index 9d4af961c5d1..9bfa50633654 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_dbgmgr.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_dbgmgr.c
@@ -96,7 +96,7 @@ bool kfd_dbgmgr_create(struct kfd_dbgmgr **ppmgr, struct kfd_dev *pdev)
long kfd_dbgmgr_register(struct kfd_dbgmgr *pmgr, struct kfd_process *p)
{
if (pmgr->pasid != 0) {
- pr_debug("H/W debugger is already active using pasid %d\n",
+ pr_debug("H/W debugger is already active using pasid 0x%x\n",
pmgr->pasid);
return -EBUSY;
}
@@ -117,7 +117,7 @@ long kfd_dbgmgr_unregister(struct kfd_dbgmgr *pmgr, struct kfd_process *p)
{
/* Is the requests coming from the already registered process? */
if (pmgr->pasid != p->pasid) {
- pr_debug("H/W debugger is not registered by calling pasid %d\n",
+ pr_debug("H/W debugger is not registered by calling pasid 0x%x\n",
p->pasid);
return -EINVAL;
}
@@ -134,7 +134,7 @@ long kfd_dbgmgr_wave_control(struct kfd_dbgmgr *pmgr,
{
/* Is the requests coming from the already registered process? */
if (pmgr->pasid != wac_info->process->pasid) {
- pr_debug("H/W debugger support was not registered for requester pasid %d\n",
+ pr_debug("H/W debugger support was not registered for requester pasid 0x%x\n",
wac_info->process->pasid);
return -EINVAL;
}
@@ -147,7 +147,7 @@ long kfd_dbgmgr_address_watch(struct kfd_dbgmgr *pmgr,
{
/* Is the requests coming from the already registered process? */
if (pmgr->pasid != adw_info->process->pasid) {
- pr_debug("H/W debugger support was not registered for requester pasid %d\n",
+ pr_debug("H/W debugger support was not registered for requester pasid 0x%x\n",
adw_info->process->pasid);
return -EINVAL;
}
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
index 82e1240bc16a..b5f86f3d605c 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
@@ -39,6 +39,41 @@
*/
static atomic_t kfd_locked = ATOMIC_INIT(0);
+#ifdef CONFIG_DRM_AMDGPU_CIK
+extern const struct kfd2kgd_calls gfx_v7_kfd2kgd;
+#endif
+extern const struct kfd2kgd_calls gfx_v8_kfd2kgd;
+extern const struct kfd2kgd_calls gfx_v9_kfd2kgd;
+extern const struct kfd2kgd_calls arcturus_kfd2kgd;
+extern const struct kfd2kgd_calls gfx_v10_kfd2kgd;
+
+static const struct kfd2kgd_calls *kfd2kgd_funcs[] = {
+#ifdef KFD_SUPPORT_IOMMU_V2
+#ifdef CONFIG_DRM_AMDGPU_CIK
+ [CHIP_KAVERI] = &gfx_v7_kfd2kgd,
+#endif
+ [CHIP_CARRIZO] = &gfx_v8_kfd2kgd,
+ [CHIP_RAVEN] = &gfx_v9_kfd2kgd,
+#endif
+#ifdef CONFIG_DRM_AMDGPU_CIK
+ [CHIP_HAWAII] = &gfx_v7_kfd2kgd,
+#endif
+ [CHIP_TONGA] = &gfx_v8_kfd2kgd,
+ [CHIP_FIJI] = &gfx_v8_kfd2kgd,
+ [CHIP_POLARIS10] = &gfx_v8_kfd2kgd,
+ [CHIP_POLARIS11] = &gfx_v8_kfd2kgd,
+ [CHIP_POLARIS12] = &gfx_v8_kfd2kgd,
+ [CHIP_VEGAM] = &gfx_v8_kfd2kgd,
+ [CHIP_VEGA10] = &gfx_v9_kfd2kgd,
+ [CHIP_VEGA12] = &gfx_v9_kfd2kgd,
+ [CHIP_VEGA20] = &gfx_v9_kfd2kgd,
+ [CHIP_RENOIR] = &gfx_v9_kfd2kgd,
+ [CHIP_ARCTURUS] = &arcturus_kfd2kgd,
+ [CHIP_NAVI10] = &gfx_v10_kfd2kgd,
+ [CHIP_NAVI12] = &gfx_v10_kfd2kgd,
+ [CHIP_NAVI14] = &gfx_v10_kfd2kgd,
+};
+
#ifdef KFD_SUPPORT_IOMMU_V2
static const struct kfd_device_info kaveri_device_info = {
.asic_family = CHIP_KAVERI,
@@ -387,6 +422,24 @@ static const struct kfd_device_info navi10_device_info = {
.num_sdma_queues_per_engine = 8,
};
+static const struct kfd_device_info navi12_device_info = {
+ .asic_family = CHIP_NAVI12,
+ .asic_name = "navi12",
+ .max_pasid_bits = 16,
+ .max_no_of_hqd = 24,
+ .doorbell_size = 8,
+ .ih_ring_entry_size = 8 * sizeof(uint32_t),
+ .event_interrupt_class = &event_interrupt_class_v9,
+ .num_of_watch_points = 4,
+ .mqd_size_aligned = MQD_SIZE_ALIGNED,
+ .needs_iommu_device = false,
+ .supports_cwsr = true,
+ .needs_pci_atomics = false,
+ .num_sdma_engines = 2,
+ .num_xgmi_sdma_engines = 0,
+ .num_sdma_queues_per_engine = 8,
+};
+
static const struct kfd_device_info navi14_device_info = {
.asic_family = CHIP_NAVI14,
.asic_name = "navi14",
@@ -425,6 +478,7 @@ static const struct kfd_device_info *kfd_supported_devices[][2] = {
[CHIP_RENOIR] = {&renoir_device_info, NULL},
[CHIP_ARCTURUS] = {&arcturus_device_info, &arcturus_device_info},
[CHIP_NAVI10] = {&navi10_device_info, NULL},
+ [CHIP_NAVI12] = {&navi12_device_info, &navi12_device_info},
[CHIP_NAVI14] = {&navi14_device_info, NULL},
};
@@ -440,20 +494,22 @@ void kfd_cleanup_processes_srcu(void);
#endif
struct kfd_dev *kgd2kfd_probe(struct kgd_dev *kgd,
- struct pci_dev *pdev, const struct kfd2kgd_calls *f2g,
- unsigned int asic_type, bool vf)
+ struct pci_dev *pdev, unsigned int asic_type, bool vf)
{
struct kfd_dev *kfd;
const struct kfd_device_info *device_info;
+ const struct kfd2kgd_calls *f2g;
- if (asic_type >= sizeof(kfd_supported_devices) / (sizeof(void *) * 2)) {
+ if (asic_type >= sizeof(kfd_supported_devices) / (sizeof(void *) * 2)
+ || asic_type >= sizeof(kfd2kgd_funcs) / sizeof(void *)) {
dev_err(kfd_device, "asic_type %d out of range\n", asic_type);
return NULL; /* asic_type out of range */
}
device_info = kfd_supported_devices[asic_type][vf];
+ f2g = kfd2kgd_funcs[asic_type];
- if (!device_info) {
+ if (!device_info || !f2g) {
dev_err(kfd_device, "%s %s not supported in kfd\n",
amdgpu_asic_name[asic_type], vf ? "VF" : "");
return NULL;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
index 2c3370051589..0b63740b4c63 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
@@ -224,20 +224,30 @@ static int allocate_vmid(struct device_queue_manager *dqm,
struct qcm_process_device *qpd,
struct queue *q)
{
- int bit, allocated_vmid;
+ int allocated_vmid = -1, i;
- if (dqm->vmid_bitmap == 0)
- return -ENOMEM;
+ for (i = dqm->dev->vm_info.first_vmid_kfd;
+ i <= dqm->dev->vm_info.last_vmid_kfd; i++) {
+ if (!dqm->vmid_pasid[i]) {
+ allocated_vmid = i;
+ break;
+ }
+ }
+
+ if (allocated_vmid < 0) {
+ pr_err("no more vmid to allocate\n");
+ return -ENOSPC;
+ }
+
+ pr_debug("vmid allocated: %d\n", allocated_vmid);
+
+ dqm->vmid_pasid[allocated_vmid] = q->process->pasid;
- bit = ffs(dqm->vmid_bitmap) - 1;
- dqm->vmid_bitmap &= ~(1 << bit);
+ set_pasid_vmid_mapping(dqm, q->process->pasid, allocated_vmid);
- allocated_vmid = bit + dqm->dev->vm_info.first_vmid_kfd;
- pr_debug("vmid allocation %d\n", allocated_vmid);
qpd->vmid = allocated_vmid;
q->properties.vmid = allocated_vmid;
- set_pasid_vmid_mapping(dqm, q->process->pasid, q->properties.vmid);
program_sh_mem_settings(dqm, qpd);
/* qpd->page_table_base is set earlier when register_process()
@@ -278,8 +288,6 @@ static void deallocate_vmid(struct device_queue_manager *dqm,
struct qcm_process_device *qpd,
struct queue *q)
{
- int bit = qpd->vmid - dqm->dev->vm_info.first_vmid_kfd;
-
/* On GFX v7, CP doesn't flush TC at dequeue */
if (q->device->device_info->asic_family == CHIP_HAWAII)
if (flush_texture_cache_nocpsch(q->device, qpd))
@@ -289,8 +297,8 @@ static void deallocate_vmid(struct device_queue_manager *dqm,
/* Release the vmid mapping */
set_pasid_vmid_mapping(dqm, 0, qpd->vmid);
+ dqm->vmid_pasid[qpd->vmid] = 0;
- dqm->vmid_bitmap |= (1 << bit);
qpd->vmid = 0;
q->properties.vmid = 0;
}
@@ -697,7 +705,7 @@ static int evict_process_queues_nocpsch(struct device_queue_manager *dqm,
goto out;
pdd = qpd_to_pdd(qpd);
- pr_info_ratelimited("Evicting PASID %u queues\n",
+ pr_info_ratelimited("Evicting PASID 0x%x queues\n",
pdd->process->pasid);
/* Mark all queues as evicted. Deactivate all active queues on
@@ -743,7 +751,7 @@ static int evict_process_queues_cpsch(struct device_queue_manager *dqm,
goto out;
pdd = qpd_to_pdd(qpd);
- pr_info_ratelimited("Evicting PASID %u queues\n",
+ pr_info_ratelimited("Evicting PASID 0x%x queues\n",
pdd->process->pasid);
/* Mark all queues as evicted. Deactivate all active queues on
@@ -794,7 +802,7 @@ static int restore_process_queues_nocpsch(struct device_queue_manager *dqm,
goto out;
}
- pr_info_ratelimited("Restoring PASID %u queues\n",
+ pr_info_ratelimited("Restoring PASID 0x%x queues\n",
pdd->process->pasid);
/* Update PD Base in QPD */
@@ -870,7 +878,7 @@ static int restore_process_queues_cpsch(struct device_queue_manager *dqm,
goto out;
}
- pr_info_ratelimited("Restoring PASID %u queues\n",
+ pr_info_ratelimited("Restoring PASID 0x%x queues\n",
pdd->process->pasid);
/* Update PD Base in QPD */
@@ -1007,7 +1015,8 @@ static int initialize_nocpsch(struct device_queue_manager *dqm)
dqm->allocated_queues[pipe] |= 1 << queue;
}
- dqm->vmid_bitmap = (1 << dqm->dev->vm_info.vmid_num_kfd) - 1;
+ memset(dqm->vmid_pasid, 0, sizeof(dqm->vmid_pasid));
+
dqm->sdma_bitmap = ~0ULL >> (64 - get_num_sdma_queues(dqm));
dqm->xgmi_sdma_bitmap = ~0ULL >> (64 - get_num_xgmi_sdma_queues(dqm));
@@ -1947,6 +1956,7 @@ struct device_queue_manager *device_queue_manager_init(struct kfd_dev *dev)
device_queue_manager_init_v9(&dqm->asic_ops);
break;
case CHIP_NAVI10:
+ case CHIP_NAVI12:
case CHIP_NAVI14:
device_queue_manager_init_v10_navi10(&dqm->asic_ops);
break;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h
index 52e54e0c0e82..2f5b336b8839 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h
@@ -35,6 +35,7 @@
#define USE_DEFAULT_GRACE_PERIOD 0xffffffff
+#define VMID_NUM 16
struct device_process_node {
struct qcm_process_device *qpd;
@@ -190,7 +191,8 @@ struct device_queue_manager {
unsigned int *allocated_queues;
uint64_t sdma_bitmap;
uint64_t xgmi_sdma_bitmap;
- unsigned int vmid_bitmap;
+ /* the pasid mapping for each kfd vmid */
+ uint16_t vmid_pasid[VMID_NUM];
uint64_t pipelines_addr;
struct kfd_mem_obj *pipeline_mem;
uint64_t fence_gpu_addr;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_events.c b/drivers/gpu/drm/amd/amdkfd/kfd_events.c
index a7ca6c682e3a..dca2d3658795 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_events.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_events.c
@@ -868,8 +868,8 @@ static void lookup_events_by_type_and_signal(struct kfd_process *p,
if (type == KFD_EVENT_TYPE_MEMORY) {
dev_warn(kfd_device,
- "Sending SIGSEGV to HSA Process with PID %d ",
- p->lead_thread->pid);
+ "Sending SIGSEGV to process %d (pasid 0x%x)",
+ p->lead_thread->pid, p->pasid);
send_sig(SIGSEGV, p->lead_thread, 0);
}
@@ -877,13 +877,13 @@ static void lookup_events_by_type_and_signal(struct kfd_process *p,
if (send_signal) {
if (send_sigterm) {
dev_warn(kfd_device,
- "Sending SIGTERM to HSA Process with PID %d ",
- p->lead_thread->pid);
+ "Sending SIGTERM to process %d (pasid 0x%x)",
+ p->lead_thread->pid, p->pasid);
send_sig(SIGTERM, p->lead_thread, 0);
} else {
dev_err(kfd_device,
- "HSA Process (PID %d) got unhandled exception",
- p->lead_thread->pid);
+ "Process %d (pasid 0x%x) got unhandled exception",
+ p->lead_thread->pid, p->pasid);
}
}
}
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c b/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c
index ae950633228c..bb77b8890e77 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c
@@ -413,6 +413,7 @@ int kfd_init_apertures(struct kfd_process *process)
case CHIP_RENOIR:
case CHIP_ARCTURUS:
case CHIP_NAVI10:
+ case CHIP_NAVI12:
case CHIP_NAVI14:
kfd_init_apertures_v9(pdd, id);
break;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c
index ab8a695c4a3c..ee82632cfed4 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c
@@ -58,8 +58,7 @@ static bool event_interrupt_isr_v9(struct kfd_dev *dev,
memcpy(patched_ihre, ih_ring_entry,
dev->device_info->ih_ring_entry_size);
- pasid = dev->kfd2kgd->get_atc_vmid_pasid_mapping_pasid(
- dev->kgd, vmid);
+ pasid = dev->dqm->vmid_pasid[vmid];
/* Patch the pasid field */
patched_ihre[3] = cpu_to_le32((le32_to_cpu(patched_ihre[3])
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_iommu.c b/drivers/gpu/drm/amd/amdkfd/kfd_iommu.c
index 1ec140a8f053..7bfc0cdcf5c2 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_iommu.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_iommu.c
@@ -160,7 +160,7 @@ static void iommu_pasid_shutdown_callback(struct pci_dev *pdev, int pasid)
if (!p)
return;
- pr_debug("Unbinding process %d from IOMMU\n", pasid);
+ pr_debug("Unbinding process 0x%x from IOMMU\n", pasid);
mutex_lock(kfd_get_dbgmgr_mutex());
@@ -194,7 +194,7 @@ static int iommu_invalid_ppr_cb(struct pci_dev *pdev, int pasid,
struct kfd_dev *dev;
dev_warn_ratelimited(kfd_device,
- "Invalid PPR device %x:%x.%x pasid %d address 0x%lX flags 0x%X",
+ "Invalid PPR device %x:%x.%x pasid 0x%x address 0x%lX flags 0x%X",
PCI_BUS_NUM(pdev->devfn),
PCI_SLOT(pdev->devfn),
PCI_FUNC(pdev->devfn),
@@ -235,7 +235,7 @@ static int kfd_bind_processes_to_device(struct kfd_dev *kfd)
err = amd_iommu_bind_pasid(kfd->pdev, p->pasid,
p->lead_thread);
if (err < 0) {
- pr_err("Unexpected pasid %d binding failure\n",
+ pr_err("Unexpected pasid 0x%x binding failure\n",
p->pasid);
mutex_unlock(&p->mutex);
break;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c b/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c
index ac031dc09d66..5e2d75ca2b62 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c
@@ -370,6 +370,7 @@ struct kernel_queue *kernel_queue_init(struct kfd_dev *dev,
kernel_queue_init_v9(&kq->ops_asic_specific);
break;
case CHIP_NAVI10:
+ case CHIP_NAVI12:
case CHIP_NAVI14:
kernel_queue_init_v10(&kq->ops_asic_specific);
break;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_module.c b/drivers/gpu/drm/amd/amdkfd/kfd_module.c
index d0a05024d38c..c7b857b542e8 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_module.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_module.c
@@ -90,7 +90,7 @@ static void kfd_exit(void)
kfd_chardev_exit();
}
-int kgd2kfd_init()
+int kgd2kfd_init(void)
{
return kfd_init();
}
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c
index ad70ebd15c6b..5462f9e5e690 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c
@@ -69,35 +69,13 @@ static void update_cu_mask(struct mqd_manager *mm, void *mqd,
static struct kfd_mem_obj *allocate_mqd(struct kfd_dev *kfd,
struct queue_properties *q)
{
- int retval;
- struct kfd_mem_obj *mqd_mem_obj = NULL;
+ struct kfd_mem_obj *mqd_mem_obj;
- /* From V9, for CWSR, the control stack is located on the next page
- * boundary after the mqd, we will use the gtt allocation function
- * instead of sub-allocation function.
- */
- if (kfd->cwsr_enabled && (q->type == KFD_QUEUE_TYPE_COMPUTE)) {
- mqd_mem_obj = kzalloc(sizeof(struct kfd_mem_obj), GFP_NOIO);
- if (!mqd_mem_obj)
- return NULL;
- retval = amdgpu_amdkfd_alloc_gtt_mem(kfd->kgd,
- ALIGN(q->ctl_stack_size, PAGE_SIZE) +
- ALIGN(sizeof(struct v10_compute_mqd), PAGE_SIZE),
- &(mqd_mem_obj->gtt_mem),
- &(mqd_mem_obj->gpu_addr),
- (void *)&(mqd_mem_obj->cpu_ptr), true);
- } else {
- retval = kfd_gtt_sa_allocate(kfd, sizeof(struct v10_compute_mqd),
- &mqd_mem_obj);
- }
-
- if (retval) {
- kfree(mqd_mem_obj);
+ if (kfd_gtt_sa_allocate(kfd, sizeof(struct v10_compute_mqd),
+ &mqd_mem_obj))
return NULL;
- }
return mqd_mem_obj;
-
}
static void init_mqd(struct mqd_manager *mm, void **mqd,
@@ -250,14 +228,7 @@ static int destroy_mqd(struct mqd_manager *mm, void *mqd,
static void free_mqd(struct mqd_manager *mm, void *mqd,
struct kfd_mem_obj *mqd_mem_obj)
{
- struct kfd_dev *kfd = mm->dev;
-
- if (mqd_mem_obj->gtt_mem) {
- amdgpu_amdkfd_free_gtt_mem(kfd->kgd, mqd_mem_obj->gtt_mem);
- kfree(mqd_mem_obj);
- } else {
- kfd_gtt_sa_free(mm->dev, mqd_mem_obj);
- }
+ kfd_gtt_sa_free(mm->dev, mqd_mem_obj);
}
static bool is_occupied(struct mqd_manager *mm, void *mqd,
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager.c
index 0c9c0c3fc51e..5069cf771949 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager.c
@@ -246,6 +246,7 @@ int pm_init(struct packet_manager *pm, struct device_queue_manager *dqm)
pm->pmf = &kfd_v9_pm_funcs;
break;
case CHIP_NAVI10:
+ case CHIP_NAVI12:
case CHIP_NAVI14:
pm->pmf = &kfd_v10_pm_funcs;
break;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
index 811291a7b7b7..6b54383e733c 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
@@ -44,7 +44,9 @@
#include <linux/interval_tree.h>
/* amdkcl: this header file is included in kcl_device_cgroup.h
#include <linux/device_cgroup.h>*/
-#include <drm/drmP.h>
+#include <drm/drm_file.h>
+#include <drm/drm_drv.h>
+#include <drm/drm_device.h>
#include <kgd_kfd_interface.h>
#include "amd_shared.h"
@@ -56,8 +58,6 @@
/* GPU ID hash width in bits */
#define KFD_GPU_ID_HASH_WIDTH 16
-struct drm_device;
-
/* Use upper bits of mmap offset to store KFD driver specific information.
* BITS[63:62] - Encode MMAP type
* BITS[61:46] - Encode gpu_id. To identify to which GPU the offset belongs to
@@ -818,7 +818,7 @@ struct kfd_process {
/* Use for delayed freeing of kfd_process structure */
struct rcu_head rcu;
- unsigned int pasid;
+ uint16_t pasid;
unsigned int doorbell_index;
/*
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
index 386ad6b7f99b..0d5a636a8d9a 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_process.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
@@ -471,7 +471,7 @@ static void kfd_process_destroy_pdds(struct kfd_process *p)
list_for_each_entry_safe(pdd, temp, &p->per_device_data,
per_device_list) {
- pr_debug("Releasing pdd (topology id %d) for process (pasid %d)\n",
+ pr_debug("Releasing pdd (topology id %d) for process (pasid 0x%x)\n",
pdd->dev->id, p->pasid);
if (pdd->drm_file) {
@@ -1314,7 +1314,7 @@ static void evict_process_worker(struct work_struct *work)
p->last_evict_timestamp = get_jiffies_64();
- pr_info("Started evicting pasid %d\n", p->pasid);
+ pr_debug("Started evicting pasid 0x%x\n", p->pasid);
ret = kfd_process_evict_queues(p);
if (!ret) {
dma_fence_signal(p->ef);
@@ -1327,9 +1327,9 @@ static void evict_process_worker(struct work_struct *work)
pr_debug("Process %d queues idle, doorbell unmapped\n",
p->pasid);
- pr_info("Finished evicting pasid %d\n", p->pasid);
+ pr_debug("Finished evicting pasid 0x%x\n", p->pasid);
} else
- pr_err("Failed to evict queues of pasid %d\n", p->pasid);
+ pr_err("Failed to evict queues of pasid 0x%x\n", p->pasid);
trace_kfd_evict_process_worker_end(p, ret ? "Failed" : "Success");
}
@@ -1348,7 +1348,7 @@ static void restore_process_worker(struct work_struct *work)
trace_kfd_restore_process_worker_start(p);
- pr_info("Started restoring pasid %d\n", p->pasid);
+ pr_debug("Started restoring pasid 0x%x\n", p->pasid);
/* Setting last_restore_timestamp before successful restoration.
* Otherwise this would have to be set by KGD (restore_process_bos)
@@ -1364,7 +1364,7 @@ static void restore_process_worker(struct work_struct *work)
ret = amdgpu_amdkfd_gpuvm_restore_process_bos(p->kgd_process_info,
&p->ef);
if (ret) {
- pr_info("Failed to restore BOs of pasid %d, retry after %d ms\n",
+ pr_debug("Failed to restore BOs of pasid 0x%x, retry after %d ms\n",
p->pasid, PROCESS_BACK_OFF_TIME_MS);
ret = queue_delayed_work(kfd_restore_wq, &p->restore_work,
@@ -1379,9 +1379,9 @@ static void restore_process_worker(struct work_struct *work)
ret = kfd_process_restore_queues(p);
trace_kfd_restore_process_worker_end(p, ret ? "Failed" : "Success");
if (!ret)
- pr_info("Finished restoring pasid %d\n", p->pasid);
+ pr_debug("Finished restoring pasid 0x%x\n", p->pasid);
else
- pr_err("Failed to restore queues of pasid %d\n", p->pasid);
+ pr_err("Failed to restore queues of pasid 0x%x\n", p->pasid);
}
void kfd_suspend_all_processes(void)
@@ -1401,7 +1401,7 @@ void kfd_suspend_all_processes(void)
cancel_delayed_work_sync(&p->restore_work);
if (kfd_process_evict_queues(p))
- pr_err("Failed to suspend process %d\n", p->pasid);
+ pr_err("Failed to suspend process 0x%x\n", p->pasid);
dma_fence_signal(p->ef);
dma_fence_put(p->ef);
p->ef = NULL;
@@ -1496,7 +1496,7 @@ int kfd_debugfs_mqds_by_process(struct seq_file *m, void *data)
#else
hash_for_each_rcu(kfd_processes_table, temp, p, kfd_processes) {
#endif
- seq_printf(m, "Process %d PASID %d:\n",
+ seq_printf(m, "Process %d PASID 0x%x:\n",
p->lead_thread->tgid, p->pasid);
mutex_lock(&p->mutex);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c
index a594945097a3..227fb0ec8115 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c
@@ -53,7 +53,7 @@ static int find_available_queue_slot(struct process_queue_manager *pqm,
pr_debug("The new slot id %lu\n", found);
if (found >= KFD_MAX_NUM_OF_QUEUES_PER_PROCESS) {
- pr_info("Cannot open more queues for process with pasid %d\n",
+ pr_info("Cannot open more queues for process with pasid 0x%x\n",
pqm->process->pasid);
return -ENOMEM;
}
@@ -298,7 +298,7 @@ int pqm_create_queue(struct process_queue_manager *pqm,
}
if (retval != 0) {
- pr_err("Pasid %d DQM create queue %d failed. ret %d\n",
+ pr_err("Pasid 0x%x DQM create queue %d failed. ret %d\n",
pqm->process->pasid, type, retval);
goto err_create_queue;
}
@@ -378,7 +378,7 @@ int pqm_destroy_queue(struct process_queue_manager *pqm, unsigned int qid)
dqm = pqn->q->device->dqm;
retval = dqm->ops.destroy_queue(dqm, &pdd->qpd, pqn->q);
if (retval) {
- pr_err("Pasid %d destroy queue %d failed, ret %d\n",
+ pr_err("Pasid 0x%x destroy queue %d failed, ret %d\n",
pqm->process->pasid,
pqn->q->properties.queue_id, retval);
if (retval != -ETIME)
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
index f1d5a4a8eb02..de2367018bc0 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
@@ -1419,6 +1419,7 @@ int kfd_topology_add_device(struct kfd_dev *gpu)
case CHIP_RENOIR:
case CHIP_ARCTURUS:
case CHIP_NAVI10:
+ case CHIP_NAVI12:
case CHIP_NAVI14:
dev->node_props.capability |= ((HSA_CAP_DOORBELL_TYPE_2_0 <<
HSA_CAP_DOORBELL_TYPE_TOTALBITS_SHIFT) &
diff --git a/drivers/gpu/drm/amd/display/Kconfig b/drivers/gpu/drm/amd/display/Kconfig
index b4504257873a..9eae7c67ceb5 100644
--- a/drivers/gpu/drm/amd/display/Kconfig
+++ b/drivers/gpu/drm/amd/display/Kconfig
@@ -22,16 +22,16 @@ config DRM_AMD_DC_DCN2_0
depends on DRM_AMD_DC && X86
depends on DRM_AMD_DC_DCN1_0
help
- Choose this option if you want to have
- Navi support for display engine
+ Choose this option if you want to have
+ Navi support for display engine
config DRM_AMD_DC_DCN2_1
- bool "DCN 2.1 family"
- depends on DRM_AMD_DC && X86
- depends on DRM_AMD_DC_DCN2_0
- help
- Choose this option if you want to have
- Renoir support for display engine
+ bool "DCN 2.1 family"
+ depends on DRM_AMD_DC && X86
+ depends on DRM_AMD_DC_DCN2_0
+ help
+ Choose this option if you want to have
+ Renoir support for display engine
config DRM_AMD_DC_DSC_SUPPORT
bool "DSC support"
@@ -40,16 +40,16 @@ config DRM_AMD_DC_DSC_SUPPORT
depends on DRM_AMD_DC_DCN1_0
depends on DRM_AMD_DC_DCN2_0
help
- Choose this option if you want to have
- Dynamic Stream Compression support
+ Choose this option if you want to have
+ Dynamic Stream Compression support
config DRM_AMD_DC_HDCP
- bool "Enable HDCP support in DC"
- depends on DRM_AMD_DC
- help
- Choose this option
- if you want to support
- HDCP authentication
+ bool "Enable HDCP support in DC"
+ depends on DRM_AMD_DC
+ help
+ Choose this option
+ if you want to support
+ HDCP authentication
config DEBUG_KERNEL_DC
bool "Enable kgdb break in DC"
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 ed0aa6b81775..10233187e165 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -983,27 +983,29 @@ static int detect_mst_link_for_all_connectors(struct drm_device *dev)
{
struct amdgpu_dm_connector *aconnector;
struct drm_connector *connector;
+ struct drm_connector_list_iter iter;
int ret = 0;
- drm_modeset_lock(&dev->mode_config.connection_mutex, NULL);
-
- list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
+ drm_connector_list_iter_begin(dev, &iter);
+ drm_for_each_connector_iter(connector, &iter) {
aconnector = to_amdgpu_dm_connector(connector);
if (aconnector->dc_link->type == dc_connection_mst_branch &&
aconnector->mst_mgr.aux) {
DRM_DEBUG_DRIVER("DM_MST: starting TM on aconnector: %p [id: %d]\n",
- aconnector, aconnector->base.base.id);
+ aconnector,
+ aconnector->base.base.id);
ret = drm_dp_mst_topology_mgr_set_mst(&aconnector->mst_mgr, true);
if (ret < 0) {
DRM_ERROR("DM_MST: Failed to start MST\n");
- ((struct dc_link *)aconnector->dc_link)->type = dc_connection_single;
- return ret;
- }
+ aconnector->dc_link->type =
+ dc_connection_single;
+ break;
}
+ }
}
+ drm_connector_list_iter_end(&iter);
- drm_modeset_unlock(&dev->mode_config.connection_mutex);
return ret;
}
@@ -1046,14 +1048,13 @@ static void s3_handle_mst(struct drm_device *dev, bool suspend)
{
struct amdgpu_dm_connector *aconnector;
struct drm_connector *connector;
+ struct drm_connector_list_iter iter;
struct drm_dp_mst_topology_mgr *mgr;
int ret;
bool need_hotplug = false;
- drm_modeset_lock(&dev->mode_config.connection_mutex, NULL);
-
- list_for_each_entry(connector, &dev->mode_config.connector_list,
- head) {
+ drm_connector_list_iter_begin(dev, &iter);
+ drm_for_each_connector_iter(connector, &iter) {
aconnector = to_amdgpu_dm_connector(connector);
if (aconnector->dc_link->type != dc_connection_mst_branch ||
aconnector->mst_port)
@@ -1071,8 +1072,7 @@ static void s3_handle_mst(struct drm_device *dev, bool suspend)
}
}
}
-
- drm_modeset_unlock(&dev->mode_config.connection_mutex);
+ drm_connector_list_iter_end(&iter);
if (need_hotplug)
drm_kms_helper_hotplug_event(dev);
@@ -1080,7 +1080,7 @@ static void s3_handle_mst(struct drm_device *dev, bool suspend)
/**
* dm_hw_init() - Initialize DC device
- * @handle: The base driver device containing the amdpgu_dm device.
+ * @handle: The base driver device containing the amdgpu_dm device.
*
* Initialize the &struct amdgpu_display_manager device. This involves calling
* the initializers of each DM component, then populating the struct with them.
@@ -1110,7 +1110,7 @@ static int dm_hw_init(void *handle)
/**
* dm_hw_fini() - Teardown DC device
- * @handle: The base driver device containing the amdpgu_dm device.
+ * @handle: The base driver device containing the amdgpu_dm device.
*
* Teardown components within &struct amdgpu_display_manager that require
* cleanup. This involves cleaning up the DRM device, DC, and any modules that
@@ -1265,6 +1265,7 @@ static int dm_resume(void *handle)
struct amdgpu_display_manager *dm = &adev->dm;
struct amdgpu_dm_connector *aconnector;
struct drm_connector *connector;
+ struct drm_connector_list_iter iter;
struct drm_crtc *crtc;
struct drm_crtc_state *new_crtc_state;
struct dm_crtc_state *dm_new_crtc_state;
@@ -1289,17 +1290,18 @@ static int dm_resume(void *handle)
/* program HPD filter */
dc_resume(dm->dc);
- /* On resume we need to rewrite the MSTM control bits to enamble MST*/
- s3_handle_mst(ddev, false);
-
/*
* early enable HPD Rx IRQ, should be done before set mode as short
* pulse interrupts are used for MST
*/
amdgpu_dm_irq_resume_early(adev);
+ /* On resume we need to rewrite the MSTM control bits to enable MST*/
+ s3_handle_mst(ddev, false);
+
/* Do detection*/
- list_for_each_entry(connector, &ddev->mode_config.connector_list, head) {
+ drm_connector_list_iter_begin(ddev, &iter);
+ drm_for_each_connector_iter(connector, &iter) {
aconnector = to_amdgpu_dm_connector(connector);
/*
@@ -1327,6 +1329,7 @@ static int dm_resume(void *handle)
amdgpu_dm_update_connector_after_detect(aconnector);
mutex_unlock(&aconnector->hpd_lock);
}
+ drm_connector_list_iter_end(&iter);
/* Force mode set in atomic commit */
#if !defined(for_each_new_crtc_in_state)
@@ -2625,8 +2628,6 @@ static int amdgpu_dm_initialize_drm_device(struct amdgpu_device *adev)
if (adev->asic_type != CHIP_CARRIZO && adev->asic_type != CHIP_STONEY)
dm->dc->debug.disable_stutter = amdgpu_pp_feature_mask & PP_STUTTER_MODE ? false : true;
- if (adev->asic_type == CHIP_RENOIR)
- dm->dc->debug.disable_stutter = true;
return 0;
fail:
@@ -3801,7 +3802,8 @@ static void fill_stream_properties_from_drm_display_mode(
struct dc_crtc_timing *timing_out = &stream->timing;
const struct drm_display_info *info = &connector->display_info;
struct amdgpu_dm_connector *aconnector = to_amdgpu_dm_connector(connector);
- memset(timing_out, 0, sizeof(struct dc_crtc_timing));
+ struct hdmi_vendor_infoframe hv_frame;
+ struct hdmi_avi_infoframe avi_frame;
timing_out->h_border_left = 0;
timing_out->h_border_right = 0;
@@ -3838,6 +3840,17 @@ static void fill_stream_properties_from_drm_display_mode(
timing_out->flags.VSYNC_POSITIVE_POLARITY = 1;
}
+ if (stream->signal == SIGNAL_TYPE_HDMI_TYPE_A) {
+#if defined(HAVE_2ARGS_DRM_HDMI_AVI_INFOFRAME_FROM_DISPLAY_MODE)
+ drm_hdmi_avi_infoframe_from_display_mode(&avi_frame, mode_in);
+#else
+ drm_hdmi_avi_infoframe_from_display_mode(&avi_frame, mode_in, false);
+#endif
+ timing_out->vic = avi_frame.video_code;
+ drm_hdmi_vendor_infoframe_from_display_mode(&hv_frame, (struct drm_connector *)connector, mode_in);
+ timing_out->hdmi_vic = hv_frame.vic;
+ }
+
timing_out->h_addressable = mode_in->crtc_hdisplay;
timing_out->h_total = mode_in->crtc_htotal;
timing_out->h_sync_width =
@@ -4063,6 +4076,9 @@ create_stream_for_sink(struct amdgpu_dm_connector *aconnector,
stream->dm_stream_context = aconnector;
+ stream->timing.flags.LTE_340MCSC_SCRAMBLE =
+ drm_connector->display_info.hdmi.scdc.scrambling.low_rates;
+
list_for_each_entry(preferred_mode, &aconnector->base.modes, head) {
/* Search for preferred mode */
if (preferred_mode->type & DRM_MODE_TYPE_PREFERRED) {
@@ -4137,6 +4153,9 @@ create_stream_for_sink(struct amdgpu_dm_connector *aconnector,
update_stream_signal(stream, sink);
+ if (stream->signal == SIGNAL_TYPE_HDMI_TYPE_A)
+ mod_build_hf_vsif_infopacket(stream, &stream->vsp_infopacket, false, false);
+
finish:
dc_sink_release(sink);
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.c
index ff91a4db4093..d76105283c5e 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.c
@@ -127,11 +127,16 @@ int amdgpu_dm_crtc_configure_crc_source(struct drm_crtc *crtc,
}
/* Configure dithering */
- if (!dm_need_crc_dither(source))
+ if (!dm_need_crc_dither(source)) {
dc_stream_set_dither_option(stream_state, DITHER_OPTION_TRUN8);
- else
+ dc_stream_set_dyn_expansion(stream_state->ctx->dc, stream_state,
+ DYN_EXPANSION_DISABLE);
+ } else {
dc_stream_set_dither_option(stream_state,
DITHER_OPTION_DEFAULT);
+ dc_stream_set_dyn_expansion(stream_state->ctx->dc, stream_state,
+ DYN_EXPANSION_AUTO);
+ }
unlock:
mutex_unlock(&adev->dm.dc_lock);
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c
index 2443c238c188..77181ddf6c8e 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c
@@ -29,7 +29,8 @@
#include "dm_helpers.h"
#include <drm/drm_hdcp.h>
-bool lp_write_i2c(void *handle, uint32_t address, const uint8_t *data, uint32_t size)
+static bool
+lp_write_i2c(void *handle, uint32_t address, const uint8_t *data, uint32_t size)
{
struct dc_link *link = handle;
@@ -39,7 +40,8 @@ bool lp_write_i2c(void *handle, uint32_t address, const uint8_t *data, uint32_t
return dm_helpers_submit_i2c(link->ctx, link, &cmd);
}
-bool lp_read_i2c(void *handle, uint32_t address, uint8_t offset, uint8_t *data, uint32_t size)
+static bool
+lp_read_i2c(void *handle, uint32_t address, uint8_t offset, uint8_t *data, uint32_t size)
{
struct dc_link *link = handle;
@@ -49,14 +51,16 @@ bool lp_read_i2c(void *handle, uint32_t address, uint8_t offset, uint8_t *data,
return dm_helpers_submit_i2c(link->ctx, link, &cmd);
}
-bool lp_write_dpcd(void *handle, uint32_t address, const uint8_t *data, uint32_t size)
+static bool
+lp_write_dpcd(void *handle, uint32_t address, const uint8_t *data, uint32_t size)
{
struct dc_link *link = handle;
return dm_helpers_dp_write_dpcd(link->ctx, link, address, data, size);
}
-bool lp_read_dpcd(void *handle, uint32_t address, uint8_t *data, uint32_t size)
+static bool
+lp_read_dpcd(void *handle, uint32_t address, uint8_t *data, uint32_t size)
{
struct dc_link *link = handle;
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.c
index 4e6da61d1a93..c68ab269523e 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.c
@@ -734,8 +734,10 @@ void amdgpu_dm_hpd_init(struct amdgpu_device *adev)
{
struct drm_device *dev = adev->ddev;
struct drm_connector *connector;
+ struct drm_connector_list_iter iter;
- list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
+ drm_connector_list_iter_begin(dev, &iter);
+ drm_for_each_connector_iter(connector, &iter) {
struct amdgpu_dm_connector *amdgpu_dm_connector =
to_amdgpu_dm_connector(connector);
@@ -753,6 +755,7 @@ void amdgpu_dm_hpd_init(struct amdgpu_device *adev)
true);
}
}
+ drm_connector_list_iter_end(&iter);
}
/**
@@ -767,8 +770,10 @@ void amdgpu_dm_hpd_fini(struct amdgpu_device *adev)
{
struct drm_device *dev = adev->ddev;
struct drm_connector *connector;
+ struct drm_connector_list_iter iter;
- list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
+ drm_connector_list_iter_begin(dev, &iter);
+ drm_for_each_connector_iter(connector, &iter) {
struct amdgpu_dm_connector *amdgpu_dm_connector =
to_amdgpu_dm_connector(connector);
const struct dc_link *dc_link = amdgpu_dm_connector->dc_link;
@@ -781,4 +786,5 @@ void amdgpu_dm_hpd_fini(struct amdgpu_device *adev)
false);
}
}
+ drm_connector_list_iter_end(&iter);
}
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
index 6b86eeb8f482..1292bca964ff 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
@@ -144,10 +144,8 @@ dm_dp_mst_connector_destroy(struct drm_connector *connector)
struct amdgpu_dm_connector *amdgpu_dm_connector = to_amdgpu_dm_connector(connector);
struct amdgpu_encoder *amdgpu_encoder = amdgpu_dm_connector->mst_encoder;
- if (amdgpu_dm_connector->edid) {
- kfree(amdgpu_dm_connector->edid);
- amdgpu_dm_connector->edid = NULL;
- }
+ kfree(amdgpu_dm_connector->edid);
+ amdgpu_dm_connector->edid = NULL;
drm_encoder_cleanup(&amdgpu_encoder->base);
kfree(amdgpu_encoder);
@@ -273,17 +271,17 @@ static int dm_dp_mst_get_modes(struct drm_connector *connector)
return ret;
}
-static struct drm_encoder *dm_mst_best_encoder(struct drm_connector *connector)
+static struct drm_encoder *
+dm_mst_atomic_best_encoder(struct drm_connector *connector,
+ struct drm_connector_state *connector_state)
{
- struct amdgpu_dm_connector *amdgpu_dm_connector = to_amdgpu_dm_connector(connector);
-
- return &amdgpu_dm_connector->mst_encoder->base;
+ return &to_amdgpu_dm_connector(connector)->mst_encoder->base;
}
static const struct drm_connector_helper_funcs dm_dp_mst_connector_helper_funcs = {
.get_modes = dm_dp_mst_get_modes,
.mode_valid = amdgpu_dm_connector_mode_valid,
- .best_encoder = dm_mst_best_encoder,
+ .atomic_best_encoder = dm_mst_atomic_best_encoder,
};
static void amdgpu_dm_encoder_destroy(struct drm_encoder *encoder)
@@ -495,6 +493,10 @@ void amdgpu_dm_initialize_dp_connector(struct amdgpu_display_manager *dm,
drm_dp_aux_register(&aconnector->dm_dp_aux.aux);
kcl_drm_dp_cec_register_connector(&aconnector->dm_dp_aux.aux,
aconnector->base.name, dm->adev->dev);
+
+ if (aconnector->base.connector_type == DRM_MODE_CONNECTOR_eDP)
+ return;
+
aconnector->mst_mgr.cbs = &dm_mst_cbs;
drm_dp_mst_topology_mgr_init(
&aconnector->mst_mgr,
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_pp_smu.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_pp_smu.c
index 1352019648c0..43a359c04a0f 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_pp_smu.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_pp_smu.c
@@ -590,10 +590,9 @@ void pp_rv_set_wm_ranges(struct pp_smu *pp,
if (pp_funcs && pp_funcs->set_watermarks_for_clocks_ranges)
pp_funcs->set_watermarks_for_clocks_ranges(pp_handle,
&wm_with_clock_ranges);
- else if (adev->smu.funcs &&
- adev->smu.funcs->set_watermarks_for_clock_ranges)
+ else
smu_set_watermarks_for_clock_ranges(&adev->smu,
- &wm_with_clock_ranges);
+ &wm_with_clock_ranges);
}
void pp_rv_set_pme_wa_enable(struct pp_smu *pp)
@@ -666,7 +665,6 @@ enum pp_smu_status pp_nv_set_wm_ranges(struct pp_smu *pp,
{
const struct dc_context *ctx = pp->dm;
struct amdgpu_device *adev = ctx->driver_context;
- struct smu_context *smu = &adev->smu;
struct dm_pp_wm_sets_with_clock_ranges_soc15 wm_with_clock_ranges;
struct dm_pp_clock_range_for_dmif_wm_set_soc15 *wm_dce_clocks =
wm_with_clock_ranges.wm_dmif_clocks_ranges;
@@ -709,15 +707,7 @@ enum pp_smu_status pp_nv_set_wm_ranges(struct pp_smu *pp,
ranges->writer_wm_sets[i].min_drain_clk_mhz * 1000;
}
- if (!smu->funcs)
- return PP_SMU_RESULT_UNSUPPORTED;
-
- /* 0: successful or smu.funcs->set_watermarks_for_clock_ranges = NULL;
- * 1: fail
- */
- if (smu_set_watermarks_for_clock_ranges(&adev->smu,
- &wm_with_clock_ranges))
- return PP_SMU_RESULT_UNSUPPORTED;
+ smu_set_watermarks_for_clock_ranges(&adev->smu, &wm_with_clock_ranges);
return PP_SMU_RESULT_OK;
}
@@ -902,6 +892,90 @@ enum pp_smu_status pp_nv_get_uclk_dpm_states(struct pp_smu *pp,
return PP_SMU_RESULT_FAIL;
}
+#ifdef CONFIG_DRM_AMD_DC_DCN2_1
+enum pp_smu_status pp_rn_get_dpm_clock_table(
+ struct pp_smu *pp, struct dpm_clocks *clock_table)
+{
+ const struct dc_context *ctx = pp->dm;
+ struct amdgpu_device *adev = ctx->driver_context;
+ struct smu_context *smu = &adev->smu;
+
+ if (!smu->ppt_funcs)
+ return PP_SMU_RESULT_UNSUPPORTED;
+
+ if (!smu->ppt_funcs->get_dpm_clock_table)
+ return PP_SMU_RESULT_UNSUPPORTED;
+
+ if (!smu->ppt_funcs->get_dpm_clock_table(smu, clock_table))
+ return PP_SMU_RESULT_OK;
+
+ return PP_SMU_RESULT_FAIL;
+}
+
+enum pp_smu_status pp_rn_set_wm_ranges(struct pp_smu *pp,
+ struct pp_smu_wm_range_sets *ranges)
+{
+ const struct dc_context *ctx = pp->dm;
+ struct amdgpu_device *adev = ctx->driver_context;
+ struct smu_context *smu = &adev->smu;
+ struct dm_pp_wm_sets_with_clock_ranges_soc15 wm_with_clock_ranges;
+ struct dm_pp_clock_range_for_dmif_wm_set_soc15 *wm_dce_clocks =
+ wm_with_clock_ranges.wm_dmif_clocks_ranges;
+ struct dm_pp_clock_range_for_mcif_wm_set_soc15 *wm_soc_clocks =
+ wm_with_clock_ranges.wm_mcif_clocks_ranges;
+ int32_t i;
+
+ if (!smu->funcs)
+ return PP_SMU_RESULT_UNSUPPORTED;
+
+ wm_with_clock_ranges.num_wm_dmif_sets = ranges->num_reader_wm_sets;
+ wm_with_clock_ranges.num_wm_mcif_sets = ranges->num_writer_wm_sets;
+
+ for (i = 0; i < wm_with_clock_ranges.num_wm_dmif_sets; i++) {
+ if (ranges->reader_wm_sets[i].wm_inst > 3)
+ wm_dce_clocks[i].wm_set_id = WM_SET_A;
+ else
+ wm_dce_clocks[i].wm_set_id =
+ ranges->reader_wm_sets[i].wm_inst;
+
+ wm_dce_clocks[i].wm_min_dcfclk_clk_in_khz =
+ ranges->reader_wm_sets[i].min_drain_clk_mhz;
+
+ wm_dce_clocks[i].wm_max_dcfclk_clk_in_khz =
+ ranges->reader_wm_sets[i].max_drain_clk_mhz;
+
+ wm_dce_clocks[i].wm_min_mem_clk_in_khz =
+ ranges->reader_wm_sets[i].min_fill_clk_mhz;
+
+ wm_dce_clocks[i].wm_max_mem_clk_in_khz =
+ ranges->reader_wm_sets[i].max_fill_clk_mhz;
+ }
+
+ for (i = 0; i < wm_with_clock_ranges.num_wm_mcif_sets; i++) {
+ if (ranges->writer_wm_sets[i].wm_inst > 3)
+ wm_soc_clocks[i].wm_set_id = WM_SET_A;
+ else
+ wm_soc_clocks[i].wm_set_id =
+ ranges->writer_wm_sets[i].wm_inst;
+ wm_soc_clocks[i].wm_min_socclk_clk_in_khz =
+ ranges->writer_wm_sets[i].min_fill_clk_mhz;
+
+ wm_soc_clocks[i].wm_max_socclk_clk_in_khz =
+ ranges->writer_wm_sets[i].max_fill_clk_mhz;
+
+ wm_soc_clocks[i].wm_min_mem_clk_in_khz =
+ ranges->writer_wm_sets[i].min_drain_clk_mhz;
+
+ wm_soc_clocks[i].wm_max_mem_clk_in_khz =
+ ranges->writer_wm_sets[i].max_drain_clk_mhz;
+ }
+
+ smu_set_watermarks_for_clock_ranges(&adev->smu, &wm_with_clock_ranges);
+
+ return PP_SMU_RESULT_OK;
+}
+#endif
+
void dm_pp_get_funcs(
struct dc_context *ctx,
struct pp_smu_funcs *funcs)
@@ -946,6 +1020,15 @@ void dm_pp_get_funcs(
funcs->nv_funcs.set_pstate_handshake_support = pp_nv_set_pstate_handshake_support;
break;
#endif
+
+#ifdef CONFIG_DRM_AMD_DC_DCN2_1
+ case DCN_VERSION_2_1:
+ funcs->ctx.ver = PP_SMU_VER_RN;
+ funcs->rn_funcs.pp_smu.dm = ctx;
+ funcs->rn_funcs.set_wm_ranges = pp_rn_set_wm_ranges;
+ funcs->rn_funcs.get_dpm_clock_table = pp_rn_get_dpm_clock_table;
+ break;
+#endif
default:
DRM_ERROR("smu version is not supported !\n");
break;
diff --git a/drivers/gpu/drm/amd/display/dc/bios/bios_parser.c b/drivers/gpu/drm/amd/display/dc/bios/bios_parser.c
index 207f6084525c..7466e6332299 100644
--- a/drivers/gpu/drm/amd/display/dc/bios/bios_parser.c
+++ b/drivers/gpu/drm/amd/display/dc/bios/bios_parser.c
@@ -2541,7 +2541,6 @@ static enum bp_result construct_integrated_info(
/* Sort voltage table from low to high*/
if (result == BP_RESULT_OK) {
- struct clock_voltage_caps temp = {0, 0};
uint32_t i;
uint32_t j;
@@ -2551,10 +2550,8 @@ static enum bp_result construct_integrated_info(
info->disp_clk_voltage[j].max_supported_clk <
info->disp_clk_voltage[j-1].max_supported_clk) {
/* swap j and j - 1*/
- temp = info->disp_clk_voltage[j-1];
- info->disp_clk_voltage[j-1] =
- info->disp_clk_voltage[j];
- info->disp_clk_voltage[j] = temp;
+ swap(info->disp_clk_voltage[j - 1],
+ info->disp_clk_voltage[j]);
}
}
}
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 c9f65c4df530..b4bbfb7bde12 100644
--- a/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c
+++ b/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c
@@ -1611,8 +1611,6 @@ static enum bp_result construct_integrated_info(
struct atom_common_table_header *header;
struct atom_data_revision revision;
-
- struct clock_voltage_caps temp = {0, 0};
uint32_t i;
uint32_t j;
@@ -1642,10 +1640,8 @@ static enum bp_result construct_integrated_info(
info->disp_clk_voltage[j-1].max_supported_clk
) {
/* swap j and j - 1*/
- temp = info->disp_clk_voltage[j-1];
- info->disp_clk_voltage[j-1] =
- info->disp_clk_voltage[j];
- info->disp_clk_voltage[j] = temp;
+ swap(info->disp_clk_voltage[j - 1],
+ info->disp_clk_voltage[j]);
}
}
}
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/clk_mgr.c
index 8da1256bc144..8af8fab14bcb 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/clk_mgr.c
@@ -63,6 +63,31 @@ int clk_mgr_helper_get_active_display_cnt(
return display_count;
}
+void clk_mgr_exit_optimized_pwr_state(const struct dc *dc, struct clk_mgr *clk_mgr)
+{
+ struct dc_link *edp_link = get_edp_link(dc);
+
+ if (dc->hwss.exit_optimized_pwr_state)
+ dc->hwss.exit_optimized_pwr_state(dc, dc->current_state);
+
+ if (edp_link) {
+ clk_mgr->psr_allow_active_cache = edp_link->psr_allow_active;
+ dc_link_set_psr_allow_active(edp_link, false, false);
+ }
+
+}
+
+void clk_mgr_optimize_pwr_state(const struct dc *dc, struct clk_mgr *clk_mgr)
+{
+ struct dc_link *edp_link = get_edp_link(dc);
+
+ if (edp_link)
+ dc_link_set_psr_allow_active(edp_link, clk_mgr->psr_allow_active_cache, false);
+
+ if (dc->hwss.optimize_pwr_state)
+ dc->hwss.optimize_pwr_state(dc, dc->current_state);
+
+}
struct clk_mgr *dc_clk_mgr_create(struct dc_context *ctx, struct pp_smu_funcs *pp_smu, struct dccg *dccg)
{
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn20/dcn20_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn20/dcn20_clk_mgr.c
index 559e16983f91..ecd2cb4840e3 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn20/dcn20_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn20/dcn20_clk_mgr.c
@@ -349,12 +349,36 @@ void dcn2_get_clock(struct clk_mgr *clk_mgr,
}
}
+static bool dcn2_are_clock_states_equal(struct dc_clocks *a,
+ struct dc_clocks *b)
+{
+ if (a->dispclk_khz != b->dispclk_khz)
+ return false;
+ else if (a->dppclk_khz != b->dppclk_khz)
+ return false;
+ else if (a->dcfclk_khz != b->dcfclk_khz)
+ return false;
+ else if (a->socclk_khz != b->socclk_khz)
+ return false;
+ else if (a->dcfclk_deep_sleep_khz != b->dcfclk_deep_sleep_khz)
+ return false;
+ else if (a->phyclk_khz != b->phyclk_khz)
+ return false;
+ else if (a->dramclk_khz != b->dramclk_khz)
+ return false;
+ else if (a->p_state_change_support != b->p_state_change_support)
+ return false;
+
+ return true;
+}
+
static struct clk_mgr_funcs dcn2_funcs = {
.get_dp_ref_clk_frequency = dce12_get_dp_ref_freq_khz,
.update_clocks = dcn2_update_clocks,
.init_clocks = dcn2_init_clocks,
.enable_pme_wa = dcn2_enable_pme_wa,
.get_clock = dcn2_get_clock,
+ .are_clock_states_equal = dcn2_are_clock_states_equal,
};
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn20/dcn20_clk_mgr.h b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn20/dcn20_clk_mgr.h
index ac31a9787305..c9fd824f3c23 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn20/dcn20_clk_mgr.h
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn20/dcn20_clk_mgr.h
@@ -50,4 +50,5 @@ void dcn2_get_clock(struct clk_mgr *clk_mgr,
enum dc_clock_type clock_type,
struct dc_clock_config *clock_cfg);
+void dcn20_update_clocks_update_dentist(struct clk_mgr_internal *clk_mgr);
#endif //__DCN20_CLK_MGR_H__
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c
index 787f94d815f4..93e46e376bb1 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c
@@ -52,6 +52,45 @@
#define REG(reg_name) \
(CLK_BASE.instance[0].segment[mm ## reg_name ## _BASE_IDX] + mm ## reg_name)
+
+/* TODO: evaluate how to lower or disable all dcn clocks in screen off case */
+int rn_get_active_display_cnt_wa(
+ struct dc *dc,
+ struct dc_state *context)
+{
+ int i, display_count;
+ bool hdmi_present = false;
+
+ display_count = 0;
+ for (i = 0; i < context->stream_count; i++) {
+ const struct dc_stream_state *stream = context->streams[i];
+
+ if (stream->signal == SIGNAL_TYPE_HDMI_TYPE_A)
+ hdmi_present = true;
+ }
+
+ for (i = 0; i < dc->link_count; i++) {
+ const struct dc_link *link = dc->links[i];
+
+ /*
+ * Only notify active stream or virtual stream.
+ * Need to notify virtual stream to work around
+ * headless case. HPD does not fire when system is in
+ * S0i2.
+ */
+ /* abusing the fact that the dig and phy are coupled to see if the phy is enabled */
+ if (link->connector_signal == SIGNAL_TYPE_VIRTUAL ||
+ link->link_enc->funcs->is_dig_enabled(link->link_enc))
+ display_count++;
+ }
+
+ /* WA for hang on HDMI after display off back back on*/
+ if (display_count == 0 && hdmi_present)
+ display_count = 1;
+
+ return display_count;
+}
+
void rn_update_clocks(struct clk_mgr *clk_mgr_base,
struct dc_state *context,
bool safe_to_lower)
@@ -62,17 +101,36 @@ void rn_update_clocks(struct clk_mgr *clk_mgr_base,
int display_count;
bool update_dppclk = false;
bool update_dispclk = false;
- bool enter_display_off = false;
bool dpp_clock_lowered = false;
- struct dmcu *dmcu = clk_mgr_base->ctx->dc->res_pool->dmcu;
- display_count = clk_mgr_helper_get_active_display_cnt(dc, context);
+ struct dmcu *dmcu = clk_mgr_base->ctx->dc->res_pool->dmcu;
- if (display_count == 0)
- enter_display_off = true;
+ if (dc->work_arounds.skip_clock_update)
+ return;
- if (enter_display_off == safe_to_lower) {
- rn_vbios_smu_set_display_count(clk_mgr, display_count);
+ /*
+ * if it is safe to lower, but we are already in the lower state, we don't have to do anything
+ * also if safe to lower is false, we just go in the higher state
+ */
+ if (safe_to_lower) {
+ /* check that we're not already in lower */
+ if (clk_mgr_base->clks.pwr_state != DCN_PWR_STATE_OPTIMIZED) {
+
+ display_count = rn_get_active_display_cnt_wa(dc, context);
+ /* if we can go lower, go lower */
+ if (display_count == 0) {
+ rn_vbios_smu_set_dcn_low_power_state(clk_mgr, DCN_PWR_STATE_OPTIMIZED);
+ /* update power state */
+ clk_mgr_base->clks.pwr_state = DCN_PWR_STATE_OPTIMIZED;
+ }
+ }
+ } else {
+ /* check that we're not already in the normal state */
+ if (clk_mgr_base->clks.pwr_state != DCN_PWR_STATE_NORMAL) {
+ rn_vbios_smu_set_dcn_low_power_state(clk_mgr, DCN_PWR_STATE_NORMAL);
+ /* update power state */
+ clk_mgr_base->clks.pwr_state = DCN_PWR_STATE_NORMAL;
+ }
}
if (should_set_clock(safe_to_lower, new_clocks->phyclk_khz, clk_mgr_base->clks.phyclk_khz)) {
@@ -329,10 +387,19 @@ void rn_enable_pme_wa(struct clk_mgr *clk_mgr_base)
rn_vbios_smu_enable_pme_wa(clk_mgr);
}
+void rn_init_clocks(struct clk_mgr *clk_mgr)
+{
+ memset(&(clk_mgr->clks), 0, sizeof(struct dc_clocks));
+ // Assumption is that boot state always supports pstate
+ clk_mgr->clks.p_state_change_support = true;
+ clk_mgr->clks.prev_p_state_change_support = true;
+ clk_mgr->clks.pwr_state = DCN_PWR_STATE_NORMAL;
+}
+
static struct clk_mgr_funcs dcn21_funcs = {
.get_dp_ref_clk_frequency = dce12_get_dp_ref_freq_khz,
.update_clocks = rn_update_clocks,
- .init_clocks = dcn2_init_clocks,
+ .init_clocks = rn_init_clocks,
.enable_pme_wa = rn_enable_pme_wa,
/* .dump_clk_registers = rn_dump_clk_registers */
};
@@ -405,7 +472,7 @@ struct clk_bw_params rn_bw_params = {
}
};
-void build_watermark_ranges(struct clk_bw_params *bw_params, struct pp_smu_wm_range_sets *ranges)
+void rn_build_watermark_ranges(struct clk_bw_params *bw_params, struct pp_smu_wm_range_sets *ranges)
{
int i, num_valid_sets;
@@ -462,23 +529,42 @@ void build_watermark_ranges(struct clk_bw_params *bw_params, struct pp_smu_wm_ra
}
-void clk_mgr_helper_populate_bw_params(struct clk_bw_params *bw_params, struct dpm_clocks *clock_table, struct hw_asic_id *asic_id)
+static unsigned int find_dcfclk_for_voltage(struct dpm_clocks *clock_table, unsigned int voltage)
{
int i;
+ for (i = 0; i < PP_SMU_NUM_DCFCLK_DPM_LEVELS; i++) {
+ if (clock_table->DcfClocks[i].Vol == voltage)
+ return clock_table->DcfClocks[i].Freq;
+ }
+
+ ASSERT(0);
+ return 0;
+}
+
+void rn_clk_mgr_helper_populate_bw_params(struct clk_bw_params *bw_params, struct dpm_clocks *clock_table, struct hw_asic_id *asic_id)
+{
+ int i, j = 0;
+
ASSERT(PP_SMU_NUM_FCLK_DPM_LEVELS <= MAX_NUM_DPM_LVL);
- for (i = 0; i < PP_SMU_NUM_FCLK_DPM_LEVELS; i++) {
- if (clock_table->FClocks[i].Freq == 0)
+ /* Find lowest DPM, FCLK is filled in reverse order*/
+
+ for (i = PP_SMU_NUM_FCLK_DPM_LEVELS - 1; i >= 0; i--) {
+ if (clock_table->FClocks[i].Freq != 0) {
+ j = i;
break;
+ }
+ }
- bw_params->clk_table.entries[i].dcfclk_mhz = clock_table->DcfClocks[i].Freq;
- bw_params->clk_table.entries[i].fclk_mhz = clock_table->FClocks[i].Freq;
- bw_params->clk_table.entries[i].memclk_mhz = clock_table->MemClocks[i].Freq;
- bw_params->clk_table.entries[i].socclk_mhz = clock_table->SocClocks[i].Freq;
- bw_params->clk_table.entries[i].voltage = clock_table->FClocks[i].Vol;
+ bw_params->clk_table.num_entries = j + 1;
+
+ for (i = 0; i < bw_params->clk_table.num_entries; i++, j--) {
+ bw_params->clk_table.entries[i].fclk_mhz = clock_table->FClocks[j].Freq;
+ bw_params->clk_table.entries[i].memclk_mhz = clock_table->MemClocks[j].Freq;
+ bw_params->clk_table.entries[i].voltage = clock_table->FClocks[j].Vol;
+ bw_params->clk_table.entries[i].dcfclk_mhz = find_dcfclk_for_voltage(clock_table, clock_table->FClocks[j].Vol);
}
- bw_params->clk_table.num_entries = i;
bw_params->vram_type = asic_id->vram_type;
bw_params->num_channels = asic_id->vram_width / DDR4_DRAM_WIDTH;
@@ -486,7 +572,7 @@ void clk_mgr_helper_populate_bw_params(struct clk_bw_params *bw_params, struct d
for (i = 0; i < WM_SET_COUNT; i++) {
bw_params->wm_table.entries[i].wm_inst = i;
- if (clock_table->FClocks[i].Freq == 0) {
+ if (i >= bw_params->clk_table.num_entries) {
bw_params->wm_table.entries[i].valid = false;
continue;
}
@@ -565,7 +651,7 @@ void rn_clk_mgr_construct(
if (pp_smu) {
pp_smu->rn_funcs.get_dpm_clock_table(&pp_smu->rn_funcs.pp_smu, &clock_table);
- clk_mgr_helper_populate_bw_params(clk_mgr->base.bw_params, &clock_table, &ctx->asic_id);
+ rn_clk_mgr_helper_populate_bw_params(clk_mgr->base.bw_params, &clock_table, &ctx->asic_id);
}
/*
@@ -576,15 +662,16 @@ void rn_clk_mgr_construct(
if (!debug->disable_pplib_wm_range) {
struct pp_smu_wm_range_sets ranges = {0};
- build_watermark_ranges(clk_mgr->base.bw_params, &ranges);
+ rn_build_watermark_ranges(clk_mgr->base.bw_params, &ranges);
/* Notify PP Lib/SMU which Watermarks to use for which clock ranges */
if (pp_smu && pp_smu->rn_funcs.set_wm_ranges)
pp_smu->rn_funcs.set_wm_ranges(&pp_smu->rn_funcs.pp_smu, &ranges);
}
- /* enable powerfeatures when displaycount goes to 0 */
- if (!debug->disable_48mhz_pwrdwn)
- rn_vbios_smu_enable_48mhz_tmdp_refclk_pwrdwn(clk_mgr);
+ if (!IS_FPGA_MAXIMUS_DC(ctx->dce_environment) && clk_mgr->smu_ver >= 0x00371500) {
+ /* enable powerfeatures when displaycount goes to 0 */
+ rn_vbios_smu_enable_48mhz_tmdp_refclk_pwrdwn(clk_mgr, !debug->disable_48mhz_pwrdwn);
+ }
}
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.h b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.h
index aadec06fde10..761bfda970a5 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.h
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.h
@@ -26,11 +26,20 @@
#ifndef __RN_CLK_MGR_H__
#define __RN_CLK_MGR_H__
+#include "clk_mgr.h"
+#include "dm_pp_smu.h"
+
struct rn_clk_registers {
uint32_t CLK1_CLK0_CURRENT_CNT; /* DPREFCLK */
};
-
+void rn_build_watermark_ranges(
+ struct clk_bw_params *bw_params,
+ struct pp_smu_wm_range_sets *ranges);
+void rn_clk_mgr_helper_populate_bw_params(
+ struct clk_bw_params *bw_params,
+ struct dpm_clocks *clock_table,
+ struct hw_asic_id *asic_id);
void rn_clk_mgr_construct(struct dc_context *ctx,
struct clk_mgr_internal *clk_mgr,
struct pp_smu_funcs *pp_smu,
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr_vbios_smu.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr_vbios_smu.c
index 50984c1811bb..8e860f567d5c 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr_vbios_smu.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr_vbios_smu.c
@@ -175,20 +175,27 @@ int rn_vbios_smu_set_dppclk(struct clk_mgr_internal *clk_mgr, int requested_dpp_
return actual_dppclk_set_mhz * 1000;
}
-void rn_vbios_smu_set_display_count(struct clk_mgr_internal *clk_mgr, int display_count)
+void rn_vbios_smu_set_dcn_low_power_state(struct clk_mgr_internal *clk_mgr, enum dcn_pwr_state state)
{
+ int disp_count;
+
+ if (state == DCN_PWR_STATE_OPTIMIZED)
+ disp_count = 0;
+ else
+ disp_count = 1;
+
rn_vbios_smu_send_msg_with_param(
- clk_mgr,
- VBIOSSMC_MSG_SetDisplayCount,
- display_count);
+ clk_mgr,
+ VBIOSSMC_MSG_SetDisplayCount,
+ disp_count);
}
-void rn_vbios_smu_enable_48mhz_tmdp_refclk_pwrdwn(struct clk_mgr_internal *clk_mgr)
+void rn_vbios_smu_enable_48mhz_tmdp_refclk_pwrdwn(struct clk_mgr_internal *clk_mgr, bool enable)
{
rn_vbios_smu_send_msg_with_param(
clk_mgr,
VBIOSSMC_MSG_EnableTmdp48MHzRefclkPwrDown,
- 0);
+ enable);
}
void rn_vbios_smu_enable_pme_wa(struct clk_mgr_internal *clk_mgr)
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr_vbios_smu.h b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr_vbios_smu.h
index da3a49487c6d..ccc01879c9d4 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr_vbios_smu.h
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr_vbios_smu.h
@@ -33,8 +33,8 @@ int rn_vbios_smu_set_hard_min_dcfclk(struct clk_mgr_internal *clk_mgr, int reque
int rn_vbios_smu_set_min_deep_sleep_dcfclk(struct clk_mgr_internal *clk_mgr, int requested_min_ds_dcfclk_khz);
void rn_vbios_smu_set_phyclk(struct clk_mgr_internal *clk_mgr, int requested_phyclk_khz);
int rn_vbios_smu_set_dppclk(struct clk_mgr_internal *clk_mgr, int requested_dpp_khz);
-void rn_vbios_smu_set_display_count(struct clk_mgr_internal *clk_mgr, int display_count);
-void rn_vbios_smu_enable_48mhz_tmdp_refclk_pwrdwn(struct clk_mgr_internal *clk_mgr);
+void rn_vbios_smu_set_dcn_low_power_state(struct clk_mgr_internal *clk_mgr, int display_count);
+void rn_vbios_smu_enable_48mhz_tmdp_refclk_pwrdwn(struct clk_mgr_internal *clk_mgr, bool enable);
void rn_vbios_smu_enable_pme_wa(struct clk_mgr_internal *clk_mgr);
#endif /* DAL_DC_DCN10_RV1_CLK_MGR_VBIOS_SMU_H_ */
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c
index 79ffe35d4eae..e05018b422e0 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc.c
@@ -408,6 +408,27 @@ bool dc_stream_get_crc(struct dc *dc, struct dc_stream_state *stream,
return false;
}
+void dc_stream_set_dyn_expansion(struct dc *dc, struct dc_stream_state *stream,
+ enum dc_dynamic_expansion option)
+{
+ /* OPP FMT dyn expansion updates*/
+ int i = 0;
+ struct pipe_ctx *pipe_ctx;
+
+ for (i = 0; i < MAX_PIPES; i++) {
+ if (dc->current_state->res_ctx.pipe_ctx[i].stream
+ == stream) {
+ pipe_ctx = &dc->current_state->res_ctx.pipe_ctx[i];
+ pipe_ctx->stream_res.opp->dyn_expansion = option;
+ pipe_ctx->stream_res.opp->funcs->opp_set_dyn_expansion(
+ pipe_ctx->stream_res.opp,
+ COLOR_SPACE_YCBCR601,
+ stream->timing.display_color_depth,
+ stream->signal);
+ }
+ }
+}
+
void dc_stream_set_dither_option(struct dc_stream_state *stream,
enum dc_dither_option option)
{
@@ -915,15 +936,11 @@ static void program_timing_sync(
/* set first pipe with plane as master */
for (j = 0; j < group_size; j++) {
- struct pipe_ctx *temp;
-
if (pipe_set[j]->plane_state) {
if (j == 0)
break;
- temp = pipe_set[0];
- pipe_set[0] = pipe_set[j];
- pipe_set[j] = temp;
+ swap(pipe_set[0], pipe_set[j]);
break;
}
}
@@ -980,6 +997,10 @@ bool dc_validate_seamless_boot_timing(const struct dc *dc,
struct dc_crtc_timing *crtc_timing)
{
struct timing_generator *tg;
+ struct stream_encoder *se = NULL;
+
+ struct dc_crtc_timing hw_crtc_timing = {0};
+
struct dc_link *link = sink->link;
unsigned int i, enc_inst, tg_inst = 0;
@@ -999,6 +1020,9 @@ bool dc_validate_seamless_boot_timing(const struct dc *dc,
for (i = 0; i < dc->res_pool->stream_enc_count; i++) {
if (dc->res_pool->stream_enc[i]->id == enc_inst) {
+
+ se = dc->res_pool->stream_enc[i];
+
tg_inst = dc->res_pool->stream_enc[i]->funcs->dig_source_otg(
dc->res_pool->stream_enc[i]);
break;
@@ -1014,10 +1038,46 @@ bool dc_validate_seamless_boot_timing(const struct dc *dc,
tg = dc->res_pool->timing_generators[tg_inst];
- if (!tg->funcs->is_matching_timing)
+ if (!tg->funcs->get_hw_timing)
+ return false;
+
+ if (!tg->funcs->get_hw_timing(tg, &hw_crtc_timing))
+ return false;
+
+ if (crtc_timing->h_total != hw_crtc_timing.h_total)
+ return false;
+
+ if (crtc_timing->h_border_left != hw_crtc_timing.h_border_left)
+ return false;
+
+ if (crtc_timing->h_addressable != hw_crtc_timing.h_addressable)
+ return false;
+
+ if (crtc_timing->h_border_right != hw_crtc_timing.h_border_right)
return false;
- if (!tg->funcs->is_matching_timing(tg, crtc_timing))
+ if (crtc_timing->h_front_porch != hw_crtc_timing.h_front_porch)
+ return false;
+
+ if (crtc_timing->h_sync_width != hw_crtc_timing.h_sync_width)
+ return false;
+
+ if (crtc_timing->v_total != hw_crtc_timing.v_total)
+ return false;
+
+ if (crtc_timing->v_border_top != hw_crtc_timing.v_border_top)
+ return false;
+
+ if (crtc_timing->v_addressable != hw_crtc_timing.v_addressable)
+ return false;
+
+ if (crtc_timing->v_border_bottom != hw_crtc_timing.v_border_bottom)
+ return false;
+
+ if (crtc_timing->v_front_porch != hw_crtc_timing.v_front_porch)
+ return false;
+
+ if (crtc_timing->v_sync_width != hw_crtc_timing.v_sync_width)
return false;
if (dc_is_dp_signal(link->connector_signal)) {
@@ -1030,6 +1090,20 @@ bool dc_validate_seamless_boot_timing(const struct dc *dc,
if (crtc_timing->pix_clk_100hz != pix_clk_100hz)
return false;
+ if (!se->funcs->dp_get_pixel_format)
+ return false;
+
+ if (!se->funcs->dp_get_pixel_format(
+ se,
+ &hw_crtc_timing.pixel_encoding,
+ &hw_crtc_timing.display_color_depth))
+ return false;
+
+ if (hw_crtc_timing.display_color_depth != crtc_timing->display_color_depth)
+ return false;
+
+ if (hw_crtc_timing.pixel_encoding != crtc_timing->pixel_encoding)
+ return false;
}
return true;
@@ -1657,8 +1731,16 @@ enum surface_update_type dc_check_update_surfaces_for_stream(
updates[i].surface->update_flags.raw = 0xFFFFFFFF;
}
- if (type == UPDATE_TYPE_FAST && memcmp(&dc->current_state->bw_ctx.bw.dcn.clk, &dc->clk_mgr->clks, offsetof(struct dc_clocks, prev_p_state_change_support)) != 0)
- dc->optimized_required = true;
+ if (type == UPDATE_TYPE_FAST) {
+ // If there's an available clock comparator, we use that.
+ if (dc->clk_mgr->funcs->are_clock_states_equal) {
+ if (!dc->clk_mgr->funcs->are_clock_states_equal(&dc->clk_mgr->clks, &dc->current_state->bw_ctx.bw.dcn.clk))
+ dc->optimized_required = true;
+ // Else we fallback to mem compare.
+ } else if (memcmp(&dc->current_state->bw_ctx.bw.dcn.clk, &dc->clk_mgr->clks, offsetof(struct dc_clocks, prev_p_state_change_support)) != 0) {
+ dc->optimized_required = true;
+ }
+ }
return type;
}
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 321ca57e5b6c..2f017e90c541 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_link.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_link.c
@@ -683,7 +683,7 @@ static bool is_same_edid(struct dc_edid *old_edid, struct dc_edid *new_edid)
return (memcmp(old_edid->raw_edid, new_edid->raw_edid, new_edid->length) == 0);
}
-bool wait_for_alt_mode(struct dc_link *link)
+static bool wait_for_alt_mode(struct dc_link *link)
{
/**
@@ -741,7 +741,7 @@ bool wait_for_alt_mode(struct dc_link *link)
* This does not create remote sinks but will trigger DM
* to start MST detection if a branch is detected.
*/
-bool dc_link_detect(struct dc_link *link, enum dc_detect_reason reason)
+bool dc_link_detect_helper(struct dc_link *link, enum dc_detect_reason reason)
{
struct dc_sink_init_data sink_init_data = { 0 };
struct display_sink_capability sink_caps = { 0 };
@@ -757,6 +757,7 @@ bool dc_link_detect(struct dc_link *link, enum dc_detect_reason reason)
bool same_dpcd = true;
enum dc_connection_type new_connection_type = dc_connection_none;
bool perform_dp_seamless_boot = false;
+
DC_LOGGER_INIT(link->ctx->logger);
if (dc_is_virtual_signal(link->connector_signal))
@@ -869,7 +870,8 @@ bool dc_link_detect(struct dc_link *link, enum dc_detect_reason reason)
* empty which leads to allocate_mst_payload() has "0"
* pbn_per_slot value leading to exception on dc_fixpt_div()
*/
- link->verified_link_cap = link->reported_link_cap;
+ dp_verify_mst_link_cap(link);
+
if (prev_sink != NULL)
dc_sink_release(prev_sink);
return false;
@@ -1063,6 +1065,23 @@ bool dc_link_detect(struct dc_link *link, enum dc_detect_reason reason)
dc_sink_release(prev_sink);
return true;
+
+}
+
+bool dc_link_detect(struct dc_link *link, enum dc_detect_reason reason)
+{
+ const struct dc *dc = link->dc;
+ bool ret;
+
+ /* get out of low power state */
+ clk_mgr_exit_optimized_pwr_state(dc, dc->clk_mgr);
+
+ ret = dc_link_detect_helper(link, reason);
+
+ /* Go back to power optimized state */
+ clk_mgr_optimize_pwr_state(dc, dc->clk_mgr);
+
+ return ret;
}
bool dc_link_get_hpd_state(struct dc_link *dc_link)
@@ -2235,7 +2254,7 @@ static bool dp_active_dongle_validate_timing(
break;
}
- if (dongle_caps->dongle_type != DISPLAY_DONGLE_DP_HDMI_CONVERTER ||
+ if (dpcd_caps->dongle_type != DISPLAY_DONGLE_DP_HDMI_CONVERTER ||
dongle_caps->extendedCapValid == false)
return true;
@@ -2399,13 +2418,17 @@ bool dc_link_set_abm_disable(const struct dc_link *link)
return true;
}
-bool dc_link_set_psr_enable(const struct dc_link *link, bool enable, bool wait)
+bool dc_link_set_psr_allow_active(struct dc_link *link, bool allow_active, bool wait)
{
struct dc *core_dc = link->ctx->dc;
struct dmcu *dmcu = core_dc->res_pool->dmcu;
- if ((dmcu != NULL && dmcu->funcs->is_dmcu_initialized(dmcu)) && link->psr_enabled)
- dmcu->funcs->set_psr_enable(dmcu, enable, wait);
+
+
+ if ((dmcu != NULL && dmcu->funcs->is_dmcu_initialized(dmcu)) && link->psr_feature_enabled)
+ dmcu->funcs->set_psr_enable(dmcu, allow_active, wait);
+
+ link->psr_allow_active = allow_active;
return true;
}
@@ -2716,6 +2739,10 @@ void core_link_enable_stream(
enum dc_status status;
DC_LOGGER_INIT(pipe_ctx->stream->ctx->logger);
+ if (!IS_FPGA_MAXIMUS_DC(core_dc->ctx->dce_environment) &&
+ dc_is_virtual_signal(pipe_ctx->stream->signal))
+ return;
+
if (!dc_is_virtual_signal(pipe_ctx->stream->signal)) {
stream->link->link_enc->funcs->setup(
stream->link->link_enc,
@@ -2858,6 +2885,10 @@ void core_link_disable_stream(struct pipe_ctx *pipe_ctx)
struct dc_stream_state *stream = pipe_ctx->stream;
struct dc_link *link = stream->sink->link;
+ if (!IS_FPGA_MAXIMUS_DC(core_dc->ctx->dce_environment) &&
+ dc_is_virtual_signal(pipe_ctx->stream->signal))
+ return;
+
#if defined(CONFIG_DRM_AMD_DC_HDCP)
update_psp_stream_config(pipe_ctx, true);
#endif
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c b/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c
index 7c78caf7a602..0f59b68aa4c2 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c
@@ -1409,6 +1409,9 @@ static struct dc_link_settings get_max_link_cap(struct dc_link *link)
if (link->link_enc->features.flags.bits.IS_HBR3_CAPABLE)
max_link_cap.link_rate = LINK_RATE_HIGH3;
+ if (link->link_enc->funcs->get_max_link_cap)
+ link->link_enc->funcs->get_max_link_cap(link->link_enc, &max_link_cap);
+
/* Lower link settings based on sink's link cap */
if (link->reported_link_cap.lane_count < max_link_cap.lane_count)
max_link_cap.lane_count =
@@ -1653,11 +1656,14 @@ bool dp_verify_link_cap_with_retries(
for (i = 0; i < attempts; i++) {
int fail_count = 0;
- enum dc_connection_type type;
+ enum dc_connection_type type = dc_connection_none;
memset(&link->verified_link_cap, 0,
sizeof(struct dc_link_settings));
- if (!dc_link_detect_sink(link, &type)) {
+ if (!dc_link_detect_sink(link, &type) || type == dc_connection_none) {
+ link->verified_link_cap.lane_count = LANE_COUNT_ONE;
+ link->verified_link_cap.link_rate = LINK_RATE_LOW;
+ link->verified_link_cap.link_spread = LINK_SPREAD_DISABLED;
break;
} else if (dp_verify_link_cap(link,
&link->reported_link_cap,
@@ -1670,6 +1676,19 @@ bool dp_verify_link_cap_with_retries(
return success;
}
+bool dp_verify_mst_link_cap(
+ struct dc_link *link)
+{
+ struct dc_link_settings max_link_cap = {0};
+
+ max_link_cap = get_max_link_cap(link);
+ link->verified_link_cap = get_common_supported_link_settings(
+ link->reported_link_cap,
+ max_link_cap);
+
+ return true;
+}
+
static struct dc_link_settings get_common_supported_link_settings(
struct dc_link_settings link_setting_a,
struct dc_link_settings link_setting_b)
@@ -2057,11 +2076,11 @@ static bool allow_hpd_rx_irq(const struct dc_link *link)
return false;
}
-static bool handle_hpd_irq_psr_sink(const struct dc_link *link)
+static bool handle_hpd_irq_psr_sink(struct dc_link *link)
{
union dpcd_psr_configuration psr_configuration;
- if (!link->psr_enabled)
+ if (!link->psr_feature_enabled)
return false;
dm_helpers_dp_read_dpcd(
@@ -2100,8 +2119,8 @@ static bool handle_hpd_irq_psr_sink(const struct dc_link *link)
sizeof(psr_error_status.raw));
/* PSR error, disable and re-enable PSR */
- dc_link_set_psr_enable(link, false, true);
- dc_link_set_psr_enable(link, true, true);
+ dc_link_set_psr_allow_active(link, false, true);
+ dc_link_set_psr_allow_active(link, true, true);
return true;
} else if (psr_sink_psr_status.bits.SINK_SELF_REFRESH_STATUS ==
@@ -2556,6 +2575,7 @@ static void get_active_converter_info(
uint8_t data, struct dc_link *link)
{
union dp_downstream_port_present ds_port = { .byte = data };
+ memset(&link->dpcd_caps.dongle_caps, 0, sizeof(link->dpcd_caps.dongle_caps));
/* decode converter info*/
if (!ds_port.fields.PORT_PRESENT) {
@@ -2702,6 +2722,7 @@ static void dp_wa_power_up_0010FA(struct dc_link *link, uint8_t *dpcd_data,
* keep receiver powered all the time.*/
case DP_BRANCH_DEVICE_ID_0010FA:
case DP_BRANCH_DEVICE_ID_0080E1:
+ case DP_BRANCH_DEVICE_ID_00E04C:
link->wa_flags.dp_keep_receiver_powered = true;
break;
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_stream.c b/drivers/gpu/drm/amd/display/dc/core/dc_stream.c
index 41032c4c5bdf..4431cc6000a1 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_stream.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_stream.c
@@ -420,10 +420,10 @@ bool dc_stream_add_writeback(struct dc *dc,
if (dwb->funcs->is_enabled(dwb)) {
/* writeback pipe already enabled, only need to update */
- dc->hwss.update_writeback(dc, stream_status, wb_info);
+ dc->hwss.update_writeback(dc, stream_status, wb_info, dc->current_state);
} else {
/* Enable writeback pipe from scratch*/
- dc->hwss.enable_writeback(dc, stream_status, wb_info);
+ dc->hwss.enable_writeback(dc, stream_status, wb_info, dc->current_state);
}
}
diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index 5c4428aa951e..83a63190533f 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -39,7 +39,7 @@
#include "inc/hw/dmcu.h"
#include "dml/display_mode_lib.h"
-#define DC_VER "3.2.51.1"
+#define DC_VER "3.2.54"
#define MAX_SURFACES 3
#define MAX_PLANES 6
@@ -245,6 +245,18 @@ enum wm_report_mode {
WM_REPORT_DEFAULT = 0,
WM_REPORT_OVERRIDE = 1,
};
+enum dtm_pstate{
+ dtm_level_p0 = 0,/*highest voltage*/
+ dtm_level_p1,
+ dtm_level_p2,
+ dtm_level_p3,
+ dtm_level_p4,/*when active_display_count = 0*/
+};
+
+enum dcn_pwr_state {
+ DCN_PWR_STATE_OPTIMIZED = 0,
+ DCN_PWR_STATE_NORMAL = 1
+};
/*
* For any clocks that may differ per pipe
@@ -260,12 +272,13 @@ struct dc_clocks {
int phyclk_khz;
int dramclk_khz;
bool p_state_change_support;
-
+ enum dcn_pwr_state pwr_state;
/*
* Elements below are not compared for the purposes of
* optimization required
*/
bool prev_p_state_change_support;
+ enum dtm_pstate dtm_level;
int max_supported_dppclk_khz;
int max_supported_dispclk_khz;
int bw_dppclk_khz; /*a copy of dppclk_khz*/
diff --git a/drivers/gpu/drm/amd/display/dc/dc_hw_types.h b/drivers/gpu/drm/amd/display/dc/dc_hw_types.h
index 2869b26d966a..e0856bb8511f 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_hw_types.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_hw_types.h
@@ -578,6 +578,11 @@ enum dc_quantization_range {
QUANTIZATION_RANGE_LIMITED
};
+enum dc_dynamic_expansion {
+ DYN_EXPANSION_AUTO,
+ DYN_EXPANSION_DISABLE
+};
+
/* XFM */
/* used in struct dc_plane_state */
diff --git a/drivers/gpu/drm/amd/display/dc/dc_link.h b/drivers/gpu/drm/amd/display/dc/dc_link.h
index 45e6195c5395..f24fd19ed93d 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_link.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_link.h
@@ -126,7 +126,8 @@ struct dc_link {
unsigned short chip_caps;
unsigned int dpcd_sink_count;
enum edp_revision edp_revision;
- bool psr_enabled;
+ bool psr_feature_enabled;
+ bool psr_allow_active;
/* MST record stream using this link */
struct link_flags {
@@ -158,6 +159,18 @@ static inline struct dc_link *dc_get_link_at_index(struct dc *dc, uint32_t link_
return dc->links[link_index];
}
+static inline struct dc_link *get_edp_link(const struct dc *dc)
+{
+ int i;
+
+ // report any eDP links, even unconnected DDI's
+ for (i = 0; i < dc->link_count; i++) {
+ if (dc->links[i]->connector_signal == SIGNAL_TYPE_EDP)
+ return dc->links[i];
+ }
+ return NULL;
+}
+
/* Set backlight level of an embedded panel (eDP, LVDS).
* backlight_pwm_u16_16 is unsigned 32 bit with 16 bit integer
* and 16 bit fractional, where 1.0 is max backlight value.
@@ -170,7 +183,7 @@ int dc_link_get_backlight_level(const struct dc_link *dc_link);
bool dc_link_set_abm_disable(const struct dc_link *dc_link);
-bool dc_link_set_psr_enable(const struct dc_link *dc_link, bool enable, bool wait);
+bool dc_link_set_psr_allow_active(struct dc_link *dc_link, bool enable, bool wait);
bool dc_link_get_psr_state(const struct dc_link *dc_link, uint32_t *psr_state);
diff --git a/drivers/gpu/drm/amd/display/dc/dc_stream.h b/drivers/gpu/drm/amd/display/dc/dc_stream.h
index 3c061d4f214f..fdb6adc37857 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_stream.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_stream.h
@@ -451,6 +451,9 @@ void dc_stream_set_static_screen_events(struct dc *dc,
int num_streams,
const struct dc_static_screen_events *events);
+void dc_stream_set_dyn_expansion(struct dc *dc, struct dc_stream_state *stream,
+ enum dc_dynamic_expansion option);
+
void dc_stream_set_dither_option(struct dc_stream_state *stream,
enum dc_dither_option option);
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 b5c97b313c54..4a22b50bd38a 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_abm.c
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_abm.c
@@ -75,6 +75,9 @@ static bool dce_abm_set_pipe(struct abm *abm, uint32_t controller_id)
/* notifyDMCUMsg */
REG_UPDATE(MASTER_COMM_CNTL_REG, MASTER_COMM_INTERRUPT, 1);
+ REG_WAIT(MASTER_COMM_CNTL_REG, MASTER_COMM_INTERRUPT, 0,
+ 1, 80000);
+
return true;
}
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_aux.c b/drivers/gpu/drm/amd/display/dc/dce/dce_aux.c
index 79a16942ce98..16960ef29132 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_aux.c
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_aux.c
@@ -472,7 +472,7 @@ int dce_aux_transfer_raw(struct ddc_service *ddc,
aux_req.action = i2caux_action_from_payload(payload);
aux_req.address = payload->address;
- aux_req.delay = payload->defer_delay * 10;
+ aux_req.delay = 0;
aux_req.length = payload->length;
aux_req.data = payload->data;
@@ -541,8 +541,15 @@ bool dce_aux_transfer_with_retries(struct ddc_service *ddc,
case AUX_TRANSACTION_REPLY_AUX_DEFER:
case AUX_TRANSACTION_REPLY_I2C_OVER_AUX_NACK:
case AUX_TRANSACTION_REPLY_I2C_OVER_AUX_DEFER:
- if (++aux_defer_retries >= AUX_MAX_DEFER_RETRIES)
+ if (++aux_defer_retries >= AUX_MAX_DEFER_RETRIES) {
goto fail;
+ } else {
+ if ((*payload->reply == AUX_TRANSACTION_REPLY_AUX_DEFER) ||
+ (*payload->reply == AUX_TRANSACTION_REPLY_I2C_OVER_AUX_DEFER)) {
+ if (payload->defer_delay > 0)
+ msleep(payload->defer_delay);
+ }
+ }
break;
case AUX_TRANSACTION_REPLY_I2C_DEFER:
diff --git a/drivers/gpu/drm/amd/display/dc/dce100/dce100_resource.c b/drivers/gpu/drm/amd/display/dc/dce100/dce100_resource.c
index 99c9c5e7476d..f9adbd571e96 100644
--- a/drivers/gpu/drm/amd/display/dc/dce100/dce100_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dce100/dce100_resource.c
@@ -665,6 +665,7 @@ struct clock_source *dce100_clock_source_create(
return &clk_src->base;
}
+ kfree(clk_src);
BREAK_TO_DEBUGGER();
return NULL;
}
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 c273490ddcab..27542c22fa55 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
@@ -941,7 +941,6 @@ void dce110_enable_audio_stream(struct pipe_ctx *pipe_ctx)
{
/* notify audio driver for audio modes of monitor */
struct dc *core_dc;
- struct pp_smu_funcs *pp_smu = NULL;
struct clk_mgr *clk_mgr;
unsigned int i, num_audio = 1;
@@ -954,9 +953,6 @@ void dce110_enable_audio_stream(struct pipe_ctx *pipe_ctx)
if (pipe_ctx->stream_res.audio && pipe_ctx->stream_res.audio->enabled == true)
return;
- if (core_dc->res_pool->pp_smu)
- pp_smu = core_dc->res_pool->pp_smu;
-
if (pipe_ctx->stream_res.audio) {
for (i = 0; i < MAX_PIPES; i++) {
/*current_state not updated yet*/
@@ -981,7 +977,6 @@ void dce110_enable_audio_stream(struct pipe_ctx *pipe_ctx)
void dce110_disable_audio_stream(struct pipe_ctx *pipe_ctx)
{
struct dc *dc;
- struct pp_smu_funcs *pp_smu = NULL;
struct clk_mgr *clk_mgr;
if (!pipe_ctx || !pipe_ctx->stream)
@@ -998,9 +993,6 @@ void dce110_disable_audio_stream(struct pipe_ctx *pipe_ctx)
if (pipe_ctx->stream_res.audio) {
pipe_ctx->stream_res.audio->enabled = false;
- if (dc->res_pool->pp_smu)
- pp_smu = dc->res_pool->pp_smu;
-
if (dc_is_dp_signal(pipe_ctx->stream->signal))
pipe_ctx->stream_res.stream_enc->funcs->dp_audio_disable(
pipe_ctx->stream_res.stream_enc);
@@ -1415,7 +1407,7 @@ static enum dc_status apply_single_controller_ctx_to_hw(
pipe_ctx->plane_res.scl_data.lb_params.alpha_en = pipe_ctx->bottom_pipe != 0;
- pipe_ctx->stream->link->psr_enabled = false;
+ pipe_ctx->stream->link->psr_feature_enabled = false;
return DC_OK;
}
@@ -1526,18 +1518,6 @@ static struct dc_stream_state *get_edp_stream(struct dc_state *context)
return NULL;
}
-static struct dc_link *get_edp_link(struct dc *dc)
-{
- int i;
-
- // report any eDP links, even unconnected DDI's
- for (i = 0; i < dc->link_count; i++) {
- if (dc->links[i]->connector_signal == SIGNAL_TYPE_EDP)
- return dc->links[i];
- }
- return NULL;
-}
-
static struct dc_link *get_edp_link_with_sink(
struct dc *dc,
struct dc_state *context)
@@ -1831,7 +1811,7 @@ static bool should_enable_fbc(struct dc *dc,
return false;
/* PSR should not be enabled */
- if (pipe_ctx->stream->link->psr_enabled)
+ if (pipe_ctx->stream->link->psr_feature_enabled)
return false;
/* Nothing to compress */
@@ -2461,7 +2441,6 @@ static void dce110_program_front_end_for_pipe(
struct dc *dc, struct pipe_ctx *pipe_ctx)
{
struct mem_input *mi = pipe_ctx->plane_res.mi;
- struct pipe_ctx *old_pipe = NULL;
struct dc_plane_state *plane_state = pipe_ctx->plane_state;
struct xfm_grph_csc_adjustment adjust;
struct out_csc_color_matrix tbl_entry;
@@ -2469,9 +2448,6 @@ static void dce110_program_front_end_for_pipe(
DC_LOGGER_INIT();
memset(&tbl_entry, 0, sizeof(tbl_entry));
- if (dc->current_state)
- old_pipe = &dc->current_state->res_ctx.pipe_ctx[pipe_ctx->pipe_idx];
-
memset(&adjust, 0, sizeof(adjust));
adjust.gamut_adjust_type = GRAPHICS_GAMUT_ADJUST_TYPE_BYPASS;
diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_resource.c b/drivers/gpu/drm/amd/display/dc/dce110/dce110_resource.c
index b35caf9fad49..baea8e8be72b 100644
--- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_resource.c
@@ -712,6 +712,7 @@ struct clock_source *dce110_clock_source_create(
return &clk_src->base;
}
+ kfree(clk_src);
BREAK_TO_DEBUGGER();
return NULL;
}
diff --git a/drivers/gpu/drm/amd/display/dc/dce112/dce112_resource.c b/drivers/gpu/drm/amd/display/dc/dce112/dce112_resource.c
index 254a6452b2b3..0dd3134b06bc 100644
--- a/drivers/gpu/drm/amd/display/dc/dce112/dce112_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dce112/dce112_resource.c
@@ -685,6 +685,7 @@ struct clock_source *dce112_clock_source_create(
return &clk_src->base;
}
+ kfree(clk_src);
BREAK_TO_DEBUGGER();
return NULL;
}
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 fde60d41f865..d8488385d681 100644
--- a/drivers/gpu/drm/amd/display/dc/dce120/dce120_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dce120/dce120_resource.c
@@ -498,6 +498,7 @@ static struct clock_source *dce120_clock_source_create(
return &clk_src->base;
}
+ kfree(clk_src);
BREAK_TO_DEBUGGER();
return NULL;
}
diff --git a/drivers/gpu/drm/amd/display/dc/dce80/dce80_resource.c b/drivers/gpu/drm/amd/display/dc/dce80/dce80_resource.c
index 86be482b6816..0fb8d2ba3564 100644
--- a/drivers/gpu/drm/amd/display/dc/dce80/dce80_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dce80/dce80_resource.c
@@ -699,6 +699,7 @@ struct clock_source *dce80_clock_source_create(
return &clk_src->base;
}
+ kfree(clk_src);
BREAK_TO_DEBUGGER();
return NULL;
}
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 d8b2da18db39..997e9582edc7 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.c
@@ -129,7 +129,7 @@ void dpp_set_gamut_remap_bypass(struct dcn10_dpp *dpp)
#define IDENTITY_RATIO(ratio) (dc_fixpt_u2d19(ratio) == (1 << 19))
-static bool dpp_get_optimal_number_of_taps(
+bool dpp1_get_optimal_number_of_taps(
struct dpp *dpp,
struct scaler_data *scl_data,
const struct scaling_taps *in_taps)
@@ -521,7 +521,7 @@ static const struct dpp_funcs dcn10_dpp_funcs = {
.dpp_read_state = dpp_read_state,
.dpp_reset = dpp_reset,
.dpp_set_scaler = dpp1_dscl_set_scaler_manual_scale,
- .dpp_get_optimal_number_of_taps = dpp_get_optimal_number_of_taps,
+ .dpp_get_optimal_number_of_taps = dpp1_get_optimal_number_of_taps,
.dpp_set_gamut_remap = dpp1_cm_set_gamut_remap,
.dpp_set_csc_adjustment = dpp1_cm_set_output_csc_adjustment,
.dpp_set_csc_default = dpp1_cm_set_output_csc_default,
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.h
index e2c613611ac9..1d4a7d640334 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.h
@@ -1504,6 +1504,11 @@ void dpp1_set_hdr_multiplier(
struct dpp *dpp_base,
uint32_t multiplier);
+bool dpp1_get_optimal_number_of_taps(
+ struct dpp *dpp,
+ struct scaler_data *scl_data,
+ const struct scaling_taps *in_taps);
+
void dpp1_construct(struct dcn10_dpp *dpp1,
struct dc_context *ctx,
uint32_t inst,
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 5d3b4858c6eb..acccf8c11306 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
@@ -1451,15 +1451,15 @@ static void log_tf(struct dc_context *ctx,
DC_LOG_ALL_TF_CHANNELS("Logging all channels...");
for (i = 0; i < hw_points_num; i++) {
- DC_LOG_GAMMA("R\t%d\t%llu\n", i, tf->tf_pts.red[i].value);
- DC_LOG_ALL_TF_CHANNELS("G\t%d\t%llu\n", i, tf->tf_pts.green[i].value);
- DC_LOG_ALL_TF_CHANNELS("B\t%d\t%llu\n", i, tf->tf_pts.blue[i].value);
+ DC_LOG_GAMMA("R\t%d\t%llu", i, tf->tf_pts.red[i].value);
+ DC_LOG_ALL_TF_CHANNELS("G\t%d\t%llu", i, tf->tf_pts.green[i].value);
+ DC_LOG_ALL_TF_CHANNELS("B\t%d\t%llu", i, tf->tf_pts.blue[i].value);
}
for (i = hw_points_num; i < MAX_NUM_HW_POINTS; i++) {
- DC_LOG_ALL_GAMMA("R\t%d\t%llu\n", i, tf->tf_pts.red[i].value);
- DC_LOG_ALL_TF_CHANNELS("G\t%d\t%llu\n", i, tf->tf_pts.green[i].value);
- DC_LOG_ALL_TF_CHANNELS("B\t%d\t%llu\n", i, tf->tf_pts.blue[i].value);
+ DC_LOG_ALL_GAMMA("R\t%d\t%llu", i, tf->tf_pts.red[i].value);
+ DC_LOG_ALL_TF_CHANNELS("G\t%d\t%llu", i, tf->tf_pts.green[i].value);
+ DC_LOG_ALL_TF_CHANNELS("B\t%d\t%llu", i, tf->tf_pts.blue[i].value);
}
}
@@ -2510,8 +2510,10 @@ static void program_all_pipe_in_tree(
pipe_ctx->stream_res.tg->funcs->set_vtg_params(
pipe_ctx->stream_res.tg, &pipe_ctx->stream->timing);
- dc->hwss.blank_pixel_data(dc, pipe_ctx, blank);
+ if (dc->hwss.setup_vupdate_interrupt)
+ dc->hwss.setup_vupdate_interrupt(pipe_ctx);
+ dc->hwss.blank_pixel_data(dc, pipe_ctx, blank);
}
if (pipe_ctx->plane_state != NULL)
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
index 8bf5f0f2301d..0c12395cfa36 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.h
@@ -337,6 +337,7 @@ struct dcn10_link_enc_registers {
type RDPCS_TX_FIFO_ERROR_MASK;\
type RDPCS_DPALT_DISABLE_TOGGLE_MASK;\
type RDPCS_DPALT_4LANE_TOGGLE_MASK;\
+ type RDPCS_PHY_DPALT_DP4;\
type RDPCS_PHY_DPALT_DISABLE;\
type RDPCS_PHY_DPALT_DISABLE_ACK;\
type RDPCS_PHY_DP_MPLLB_V2I;\
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_opp.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_opp.c
index 7045c00edab9..5be042acf9fa 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_opp.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_opp.c
@@ -238,6 +238,9 @@ void opp1_set_dyn_expansion(
FMT_DYNAMIC_EXP_EN, 0,
FMT_DYNAMIC_EXP_MODE, 0);
+ if (opp->dyn_expansion == DYN_EXPANSION_DISABLE)
+ return;
+
/*00 - 10-bit -> 12-bit dynamic expansion*/
/*01 - 8-bit -> 12-bit dynamic expansion*/
if (signal == SIGNAL_TYPE_HDMI_TYPE_A ||
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 f3cade20e45c..8710f3ac2abf 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c
@@ -1238,59 +1238,25 @@ bool optc1_is_stereo_left_eye(struct timing_generator *optc)
return ret;
}
-bool optc1_is_matching_timing(struct timing_generator *tg,
- const struct dc_crtc_timing *otg_timing)
+bool optc1_get_hw_timing(struct timing_generator *tg,
+ struct dc_crtc_timing *hw_crtc_timing)
{
- struct dc_crtc_timing hw_crtc_timing = {0};
struct dcn_otg_state s = {0};
- if (tg == NULL || otg_timing == NULL)
+ if (tg == NULL || hw_crtc_timing == NULL)
return false;
optc1_read_otg_state(DCN10TG_FROM_TG(tg), &s);
- hw_crtc_timing.h_total = s.h_total + 1;
- hw_crtc_timing.h_addressable = s.h_total - ((s.h_total - s.h_blank_start) + s.h_blank_end);
- hw_crtc_timing.h_front_porch = s.h_total + 1 - s.h_blank_start;
- hw_crtc_timing.h_sync_width = s.h_sync_a_end - s.h_sync_a_start;
+ hw_crtc_timing->h_total = s.h_total + 1;
+ hw_crtc_timing->h_addressable = s.h_total - ((s.h_total - s.h_blank_start) + s.h_blank_end);
+ hw_crtc_timing->h_front_porch = s.h_total + 1 - s.h_blank_start;
+ hw_crtc_timing->h_sync_width = s.h_sync_a_end - s.h_sync_a_start;
- hw_crtc_timing.v_total = s.v_total + 1;
- hw_crtc_timing.v_addressable = s.v_total - ((s.v_total - s.v_blank_start) + s.v_blank_end);
- hw_crtc_timing.v_front_porch = s.v_total + 1 - s.v_blank_start;
- hw_crtc_timing.v_sync_width = s.v_sync_a_end - s.v_sync_a_start;
-
- if (otg_timing->h_total != hw_crtc_timing.h_total)
- return false;
-
- if (otg_timing->h_border_left != hw_crtc_timing.h_border_left)
- return false;
-
- if (otg_timing->h_addressable != hw_crtc_timing.h_addressable)
- return false;
-
- if (otg_timing->h_border_right != hw_crtc_timing.h_border_right)
- return false;
-
- if (otg_timing->h_front_porch != hw_crtc_timing.h_front_porch)
- return false;
-
- if (otg_timing->h_sync_width != hw_crtc_timing.h_sync_width)
- return false;
-
- if (otg_timing->v_total != hw_crtc_timing.v_total)
- return false;
-
- if (otg_timing->v_border_top != hw_crtc_timing.v_border_top)
- return false;
-
- if (otg_timing->v_addressable != hw_crtc_timing.v_addressable)
- return false;
-
- if (otg_timing->v_border_bottom != hw_crtc_timing.v_border_bottom)
- return false;
-
- if (otg_timing->v_sync_width != hw_crtc_timing.v_sync_width)
- return false;
+ hw_crtc_timing->v_total = s.v_total + 1;
+ hw_crtc_timing->v_addressable = s.v_total - ((s.v_total - s.v_blank_start) + s.v_blank_end);
+ hw_crtc_timing->v_front_porch = s.v_total + 1 - s.v_blank_start;
+ hw_crtc_timing->v_sync_width = s.v_sync_a_end - s.v_sync_a_start;
return true;
}
@@ -1494,7 +1460,6 @@ static const struct timing_generator_funcs dcn10_tg_funcs = {
.get_frame_count = optc1_get_vblank_counter,
.get_scanoutpos = optc1_get_crtc_scanoutpos,
.get_otg_active_size = optc1_get_otg_active_size,
- .is_matching_timing = optc1_is_matching_timing,
.set_early_control = optc1_set_early_control,
/* used by enable_timing_synchronization. Not need for FPGA */
.wait_for_state = optc1_wait_for_state,
@@ -1522,7 +1487,8 @@ static const struct timing_generator_funcs dcn10_tg_funcs = {
.configure_crc = optc1_configure_crc,
.set_vtg_params = optc1_set_vtg_params,
.program_manual_trigger = optc1_program_manual_trigger,
- .setup_manual_trigger = optc1_setup_manual_trigger
+ .setup_manual_trigger = optc1_setup_manual_trigger,
+ .get_hw_timing = optc1_get_hw_timing,
};
void dcn10_timing_generator_init(struct optc *optc1)
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 83575599672e..c8d795b335ba 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.h
@@ -547,9 +547,8 @@ struct dcn_otg_state {
void optc1_read_otg_state(struct optc *optc1,
struct dcn_otg_state *s);
-bool optc1_is_matching_timing(
- struct timing_generator *tg,
- const struct dc_crtc_timing *otg_timing);
+bool optc1_get_hw_timing(struct timing_generator *tg,
+ struct dc_crtc_timing *hw_crtc_timing);
bool optc1_validate_timing(
struct timing_generator *optc,
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 4e633f3a99d4..01a3b782eee1 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c
@@ -784,6 +784,7 @@ struct clock_source *dcn10_clock_source_create(
return &clk_src->base;
}
+ kfree(clk_src);
BREAK_TO_DEBUGGER();
return NULL;
}
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 6800b906a86e..f10c1554ec01 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
@@ -1552,6 +1552,66 @@ unsigned int enc1_dig_source_otg(
return tg_inst;
}
+bool enc1_stream_encoder_dp_get_pixel_format(
+ struct stream_encoder *enc,
+ enum dc_pixel_encoding *encoding,
+ enum dc_color_depth *depth)
+{
+ uint32_t hw_encoding = 0;
+ uint32_t hw_depth = 0;
+ struct dcn10_stream_encoder *enc1 = DCN10STRENC_FROM_STRENC(enc);
+
+ if (enc == NULL ||
+ encoding == NULL ||
+ depth == NULL)
+ return false;
+
+ REG_GET_2(DP_PIXEL_FORMAT,
+ DP_PIXEL_ENCODING, &hw_encoding,
+ DP_COMPONENT_DEPTH, &hw_depth);
+
+ switch (hw_depth) {
+ case DP_COMPONENT_PIXEL_DEPTH_6BPC:
+ *depth = COLOR_DEPTH_666;
+ break;
+ case DP_COMPONENT_PIXEL_DEPTH_8BPC:
+ *depth = COLOR_DEPTH_888;
+ break;
+ case DP_COMPONENT_PIXEL_DEPTH_10BPC:
+ *depth = COLOR_DEPTH_101010;
+ break;
+ case DP_COMPONENT_PIXEL_DEPTH_12BPC:
+ *depth = COLOR_DEPTH_121212;
+ break;
+ case DP_COMPONENT_PIXEL_DEPTH_16BPC:
+ *depth = COLOR_DEPTH_161616;
+ break;
+ default:
+ *depth = COLOR_DEPTH_UNDEFINED;
+ break;
+ }
+
+ switch (hw_encoding) {
+ case DP_PIXEL_ENCODING_TYPE_RGB444:
+ *encoding = PIXEL_ENCODING_RGB;
+ break;
+ case DP_PIXEL_ENCODING_TYPE_YCBCR422:
+ *encoding = PIXEL_ENCODING_YCBCR422;
+ break;
+ case DP_PIXEL_ENCODING_TYPE_YCBCR444:
+ case DP_PIXEL_ENCODING_TYPE_Y_ONLY:
+ *encoding = PIXEL_ENCODING_YCBCR444;
+ break;
+ case DP_PIXEL_ENCODING_TYPE_YCBCR420:
+ *encoding = PIXEL_ENCODING_YCBCR420;
+ break;
+ default:
+ *encoding = PIXEL_ENCODING_UNDEFINED;
+ break;
+ }
+ return true;
+}
+
static const struct stream_encoder_funcs dcn10_str_enc_funcs = {
.dp_set_stream_attribute =
enc1_stream_encoder_dp_set_stream_attribute,
@@ -1588,6 +1648,8 @@ static const struct stream_encoder_funcs dcn10_str_enc_funcs = {
.dig_connect_to_otg = enc1_dig_connect_to_otg,
.hdmi_reset_stream_attribute = enc1_reset_hdmi_stream_attribute,
.dig_source_otg = enc1_dig_source_otg,
+
+ .dp_get_pixel_format = enc1_stream_encoder_dp_get_pixel_format,
};
void dcn10_stream_encoder_construct(
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.h
index a512cbea00d1..c9cbc21d121e 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.h
@@ -621,4 +621,9 @@ void get_audio_clock_info(
void enc1_reset_hdmi_stream_attribute(
struct stream_encoder *enc);
+bool enc1_stream_encoder_dp_get_pixel_format(
+ struct stream_encoder *enc,
+ enum dc_pixel_encoding *encoding,
+ enum dc_color_depth *depth);
+
#endif /* __DC_STREAM_ENCODER_DCN10_H__ */
diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp.c
index 2f5aade1e882..4d7e45892f08 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp.c
@@ -376,13 +376,6 @@ bool dpp2_get_optimal_number_of_taps(
struct scaler_data *scl_data,
const struct scaling_taps *in_taps)
{
- uint32_t pixel_width;
-
- if (scl_data->viewport.width > scl_data->recout.width)
- pixel_width = scl_data->recout.width;
- else
- pixel_width = scl_data->viewport.width;
-
/* Some ASICs does not support FP16 scaling, so we reject modes require this*/
if (scl_data->viewport.width != scl_data->h_active &&
scl_data->viewport.height != scl_data->v_active &&
@@ -464,7 +457,7 @@ static struct dpp_funcs dcn20_dpp_funcs = {
.dpp_read_state = dpp20_read_state,
.dpp_reset = dpp_reset,
.dpp_set_scaler = dpp1_dscl_set_scaler_manual_scale,
- .dpp_get_optimal_number_of_taps = dpp2_get_optimal_number_of_taps,
+ .dpp_get_optimal_number_of_taps = dpp1_get_optimal_number_of_taps,
.dpp_set_gamut_remap = dpp1_cm_set_gamut_remap,
.dpp_set_csc_adjustment = NULL,
.dpp_set_csc_default = NULL,
diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp.h b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp.h
index f8db8ad593f8..5b03b737b1d6 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp.h
@@ -705,11 +705,6 @@ void dpp2_set_hdr_multiplier(
struct dpp *dpp_base,
uint32_t multiplier);
-bool dpp2_get_optimal_number_of_taps(
- struct dpp *dpp,
- struct scaler_data *scl_data,
- const struct scaling_taps *in_taps);
-
bool dpp2_construct(struct dcn20_dpp *dpp2,
struct dc_context *ctx,
uint32_t inst,
diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dwb_scl.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dwb_scl.c
index cd8bc92ce3ba..880954ac0b02 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dwb_scl.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dwb_scl.c
@@ -722,7 +722,6 @@ bool dwb_program_horz_scalar(struct dcn20_dwbc *dwbc20,
struct scaling_taps num_taps)
{
uint32_t h_ratio_luma = 1;
- uint32_t h_ratio_chroma = 1;
uint32_t h_taps_luma = num_taps.h_taps;
uint32_t h_taps_chroma = num_taps.h_taps_c;
int32_t h_init_phase_luma = 0;
@@ -747,7 +746,6 @@ bool dwb_program_horz_scalar(struct dcn20_dwbc *dwbc20,
h_ratio_luma = -1;
else
h_ratio_luma = dc_fixpt_u3d19(tmp_h_ratio_luma) << 5;
- h_ratio_chroma = h_ratio_luma * 2;
/*Program ratio*/
REG_UPDATE(WBSCL_HORZ_FILTER_SCALE_RATIO, WBSCL_H_SCALE_RATIO, h_ratio_luma);
@@ -803,7 +801,6 @@ bool dwb_program_vert_scalar(struct dcn20_dwbc *dwbc20,
enum dwb_subsample_position subsample_position)
{
uint32_t v_ratio_luma = 1;
- uint32_t v_ratio_chroma = 1;
uint32_t v_taps_luma = num_taps.v_taps;
uint32_t v_taps_chroma = num_taps.v_taps_c;
int32_t v_init_phase_luma = 0;
@@ -827,7 +824,6 @@ bool dwb_program_vert_scalar(struct dcn20_dwbc *dwbc20,
v_ratio_luma = -1;
else
v_ratio_luma = dc_fixpt_u3d19(tmp_v_ratio_luma) << 5;
- v_ratio_chroma = v_ratio_luma * 2;
/*Program ratio*/
REG_UPDATE(WBSCL_VERT_FILTER_SCALE_RATIO, WBSCL_V_SCALE_RATIO, v_ratio_luma);
diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c
index 84aae9c05781..ce2530509e12 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c
@@ -1370,6 +1370,9 @@ static void dcn20_program_pipe(
pipe_ctx->stream_res.tg->funcs->set_vtg_params(
pipe_ctx->stream_res.tg, &pipe_ctx->stream->timing);
+
+ if (dc->hwss.setup_vupdate_interrupt)
+ dc->hwss.setup_vupdate_interrupt(pipe_ctx);
}
if (pipe_ctx->update_flags.bits.odm)
@@ -1396,6 +1399,26 @@ static void dcn20_program_pipe(
*/
if (pipe_ctx->update_flags.bits.enable || pipe_ctx->stream->update_flags.bits.out_tf)
dc->hwss.set_output_transfer_func(pipe_ctx, pipe_ctx->stream);
+
+ /* If the pipe has been enabled or has a different opp, we
+ * should reprogram the fmt. This deals with cases where
+ * interation between mpc and odm combine on different streams
+ * causes a different pipe to be chosen to odm combine with.
+ */
+ if (pipe_ctx->update_flags.bits.enable
+ || pipe_ctx->update_flags.bits.opp_changed) {
+
+ pipe_ctx->stream_res.opp->funcs->opp_set_dyn_expansion(
+ pipe_ctx->stream_res.opp,
+ COLOR_SPACE_YCBCR601,
+ pipe_ctx->stream->timing.display_color_depth,
+ pipe_ctx->stream->signal);
+
+ pipe_ctx->stream_res.opp->funcs->opp_program_fmt(
+ pipe_ctx->stream_res.opp,
+ &pipe_ctx->stream->bit_depth_params,
+ &pipe_ctx->stream->clamping);
+ }
}
static bool does_pipe_need_lock(struct pipe_ctx *pipe)
@@ -1581,8 +1604,12 @@ bool dcn20_update_bandwidth(
pipe_ctx->stream_res.tg->funcs->set_vtg_params(
pipe_ctx->stream_res.tg, &pipe_ctx->stream->timing);
+
if (pipe_ctx->prev_odm_pipe == NULL)
dc->hwss.blank_pixel_data(dc, pipe_ctx, blank);
+
+ if (dc->hwss.setup_vupdate_interrupt)
+ dc->hwss.setup_vupdate_interrupt(pipe_ctx);
}
pipe_ctx->plane_res.hubp->funcs->hubp_setup(
@@ -1599,7 +1626,8 @@ bool dcn20_update_bandwidth(
static void dcn20_enable_writeback(
struct dc *dc,
const struct dc_stream_status *stream_status,
- struct dc_writeback_info *wb_info)
+ struct dc_writeback_info *wb_info,
+ struct dc_state *context)
{
struct dwbc *dwb;
struct mcif_wb *mcif_wb;
@@ -1616,7 +1644,7 @@ static void dcn20_enable_writeback(
optc->funcs->set_dwb_source(optc, wb_info->dwb_pipe_inst);
/* set MCIF_WB buffer and arbitration configuration */
mcif_wb->funcs->config_mcif_buf(mcif_wb, &wb_info->mcif_buf_params, wb_info->dwb_params.dest_height);
- mcif_wb->funcs->config_mcif_arb(mcif_wb, &dc->current_state->bw_ctx.bw.dcn.bw_writeback.mcif_wb_arb[wb_info->dwb_pipe_inst]);
+ mcif_wb->funcs->config_mcif_arb(mcif_wb, &context->bw_ctx.bw.dcn.bw_writeback.mcif_wb_arb[wb_info->dwb_pipe_inst]);
/* Enable MCIF_WB */
mcif_wb->funcs->enable_mcif(mcif_wb);
/* Enable DWB */
@@ -2181,8 +2209,10 @@ static void dcn20_enable_stream(struct pipe_ctx *pipe_ctx)
link->link_enc->funcs->connect_dig_be_to_fe(link->link_enc,
pipe_ctx->stream_res.stream_enc->id, true);
- if (link->dc->hwss.program_dmdata_engine)
- link->dc->hwss.program_dmdata_engine(pipe_ctx);
+ if (pipe_ctx->plane_state && pipe_ctx->plane_state->flip_immediate != 1) {
+ if (link->dc->hwss.program_dmdata_engine)
+ link->dc->hwss.program_dmdata_engine(pipe_ctx);
+ }
link->dc->hwss.update_info_frame(pipe_ctx);
diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.c
index dda90995ba93..3b613fb93ef8 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.c
@@ -460,7 +460,7 @@ static struct timing_generator_funcs dcn20_tg_funcs = {
.set_vtg_params = optc1_set_vtg_params,
.program_manual_trigger = optc2_program_manual_trigger,
.setup_manual_trigger = optc2_setup_manual_trigger,
- .is_matching_timing = optc1_is_matching_timing
+ .get_hw_timing = optc1_get_hw_timing,
};
void dcn20_timing_generator_init(struct optc *optc1)
diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c
index 8d81c65157d4..d8f9527523e6 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c
@@ -579,11 +579,13 @@ static const struct dcn2_dpp_registers tf_regs[] = {
};
static const struct dcn2_dpp_shift tf_shift = {
- TF_REG_LIST_SH_MASK_DCN20(__SHIFT)
+ TF_REG_LIST_SH_MASK_DCN20(__SHIFT),
+ TF_DEBUG_REG_LIST_SH_DCN10
};
static const struct dcn2_dpp_mask tf_mask = {
- TF_REG_LIST_SH_MASK_DCN20(_MASK)
+ TF_REG_LIST_SH_MASK_DCN20(_MASK),
+ TF_DEBUG_REG_LIST_MASK_DCN10
};
#define dwbc_regs_dcn2(id)\
@@ -1075,6 +1077,7 @@ struct clock_source *dcn20_clock_source_create(
return &clk_src->base;
}
+ kfree(clk_src);
BREAK_TO_DEBUGGER();
return NULL;
}
@@ -1151,6 +1154,8 @@ static const struct resource_create_funcs res_create_maximus_funcs = {
.create_hwseq = dcn20_hwseq_create,
};
+static void dcn20_pp_smu_destroy(struct pp_smu_funcs **pp_smu);
+
void dcn20_clock_source_destroy(struct clock_source **clk_src)
{
kfree(TO_DCE110_CLK_SRC(*clk_src));
@@ -1889,7 +1894,7 @@ int dcn20_populate_dml_pipes_from_context(
break;
case PIXEL_ENCODING_YCBCR420:
pipes[pipe_cnt].dout.output_format = dm_420;
- pipes[pipe_cnt].dout.output_bpp = (output_bpc * 3) / 2;
+ pipes[pipe_cnt].dout.output_bpp = (output_bpc * 3.0) / 2;
break;
case PIXEL_ENCODING_YCBCR422:
if (true) /* todo */
@@ -1903,6 +1908,11 @@ int dcn20_populate_dml_pipes_from_context(
pipes[pipe_cnt].dout.output_bpp = output_bpc * 3;
}
+#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT
+ if (res_ctx->pipe_ctx[i].stream->timing.flags.DSC)
+ pipes[pipe_cnt].dout.output_bpp = res_ctx->pipe_ctx[i].stream->timing.dsc_cfg.bits_per_pixel / 16.0;
+#endif
+
/* todo: default max for now, until there is logic reflecting this in dc*/
pipes[pipe_cnt].dout.output_bpc = 12;
/*
@@ -2199,7 +2209,8 @@ static struct pipe_ctx *dcn20_find_secondary_pipe(struct dc *dc,
*/
if (secondary_pipe == NULL) {
for (j = dc->res_pool->pipe_count - 1; j >= 0; j--) {
- if (dc->current_state->res_ctx.pipe_ctx[j].top_pipe == NULL) {
+ if (dc->current_state->res_ctx.pipe_ctx[j].top_pipe == NULL
+ && dc->current_state->res_ctx.pipe_ctx[j].prev_odm_pipe == NULL) {
preferred_pipe_idx = j;
if (res_ctx->pipe_ctx[preferred_pipe_idx].stream == NULL) {
@@ -2919,7 +2930,7 @@ bool dcn20_mmhubbub_create(struct dc_context *ctx, struct resource_pool *pool)
return true;
}
-struct pp_smu_funcs *dcn20_pp_smu_create(struct dc_context *ctx)
+static struct pp_smu_funcs *dcn20_pp_smu_create(struct dc_context *ctx)
{
struct pp_smu_funcs *pp_smu = kzalloc(sizeof(*pp_smu), GFP_KERNEL);
@@ -2934,7 +2945,7 @@ struct pp_smu_funcs *dcn20_pp_smu_create(struct dc_context *ctx)
return pp_smu;
}
-void dcn20_pp_smu_destroy(struct pp_smu_funcs **pp_smu)
+static void dcn20_pp_smu_destroy(struct pp_smu_funcs **pp_smu)
{
if (pp_smu && *pp_smu) {
kfree(*pp_smu);
@@ -3012,7 +3023,7 @@ static void cap_soc_clocks(
static void update_bounding_box(struct dc *dc, struct _vcs_dpi_soc_bounding_box_st *bb,
struct pp_smu_nv_clock_table *max_clocks, unsigned int *uclk_states, unsigned int num_states)
{
- struct _vcs_dpi_voltage_scaling_st calculated_states[MAX_CLOCK_LIMIT_STATES] = {0};
+ struct _vcs_dpi_voltage_scaling_st calculated_states[MAX_CLOCK_LIMIT_STATES];
int i;
int num_calculated_states = 0;
int min_dcfclk = 0;
@@ -3020,6 +3031,8 @@ static void update_bounding_box(struct dc *dc, struct _vcs_dpi_soc_bounding_box_
if (num_states == 0)
return;
+ memset(calculated_states, 0, sizeof(calculated_states));
+
if (dc->bb_overrides.min_dcfclk_mhz > 0)
min_dcfclk = dc->bb_overrides.min_dcfclk_mhz;
else
diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.h b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.h
index 44f95aa0d61e..55006462f481 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.h
@@ -95,9 +95,6 @@ struct display_stream_compressor *dcn20_dsc_create(
struct dc_context *ctx, uint32_t inst);
void dcn20_dsc_destroy(struct display_stream_compressor **dsc);
-struct pp_smu_funcs *dcn20_pp_smu_create(struct dc_context *ctx);
-void dcn20_pp_smu_destroy(struct pp_smu_funcs **pp_smu);
-
struct hubp *dcn20_hubp_create(
struct dc_context *ctx,
uint32_t inst);
diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_stream_encoder.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_stream_encoder.c
index b7d977b4b0d5..412d3032e4ef 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_stream_encoder.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_stream_encoder.c
@@ -576,6 +576,10 @@ static const struct stream_encoder_funcs dcn20_str_enc_funcs = {
.set_avmute = enc1_stream_encoder_set_avmute,
.dig_connect_to_otg = enc1_dig_connect_to_otg,
.dig_source_otg = enc1_dig_source_otg,
+
+ .dp_get_pixel_format =
+ enc1_stream_encoder_dp_get_pixel_format,
+
#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT
.enc_read_state = enc2_read_state,
.dp_set_dsc_config = enc2_dp_set_dsc_config,
diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c
index de182185fe1f..f0c95d82feb2 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c
@@ -636,6 +636,8 @@ static const struct dcn10_stream_encoder_mask se_mask = {
SE_COMMON_MASK_SH_LIST_DCN20(_MASK)
};
+static void dcn21_pp_smu_destroy(struct pp_smu_funcs **pp_smu);
+
static struct input_pixel_processor *dcn21_ipp_create(
struct dc_context *ctx, uint32_t inst)
{
@@ -726,7 +728,7 @@ static const struct resource_caps res_cap_rn = {
.num_timing_generator = 4,
.num_opp = 4,
.num_video_plane = 4,
- .num_audio = 6, // 6 audio endpoints. 4 audio streams
+ .num_audio = 4, // 4 audio endpoints. 4 audio streams
.num_stream_encoder = 5,
.num_pll = 5, // maybe 3 because the last two used for USB-c
.num_dwb = 1,
@@ -804,7 +806,7 @@ static const struct dc_debug_options debug_defaults_drv = {
.disable_pplib_wm_range = false,
.scl_reset_length10 = true,
.sanity_checks = true,
- .disable_48mhz_pwrdwn = true,
+ .disable_48mhz_pwrdwn = false,
};
static const struct dc_debug_options debug_defaults_diags = {
@@ -939,7 +941,7 @@ static void destruct(struct dcn21_resource_pool *pool)
dcn_dccg_destroy(&pool->base.dccg);
if (pool->base.pp_smu != NULL)
- dcn20_pp_smu_destroy(&pool->base.pp_smu);
+ dcn21_pp_smu_destroy(&pool->base.pp_smu);
}
@@ -1278,7 +1280,6 @@ static void update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_param
dcn2_1_ip.max_num_otg = pool->base.res_cap->num_timing_generator;
dcn2_1_ip.max_num_dpp = pool->base.pipe_count;
dcn2_1_soc.num_chans = bw_params->num_channels;
- dcn2_1_soc.num_states = 0;
for (i = 0; i < clk_table->num_entries; i++) {
@@ -1288,8 +1289,9 @@ static void update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_param
dcn2_1_soc.clock_limits[i].socclk_mhz = clk_table->entries[i].socclk_mhz;
/* This is probably wrong, TODO: find correct calculation */
dcn2_1_soc.clock_limits[i].dram_speed_mts = clk_table->entries[i].memclk_mhz * 16 / 1000;
- dcn2_1_soc.num_states++;
}
+ dcn2_1_soc.clock_limits[i] = dcn2_1_soc.clock_limits[i - i];
+ dcn2_1_soc.num_states = i;
}
/* Temporary Place holder until we can get them from fuse */
@@ -1317,32 +1319,42 @@ static struct dpm_clocks dummy_clocks = {
};
-enum pp_smu_status dummy_set_wm_ranges(struct pp_smu *pp,
+static enum pp_smu_status dummy_set_wm_ranges(struct pp_smu *pp,
struct pp_smu_wm_range_sets *ranges)
{
return PP_SMU_RESULT_OK;
}
-enum pp_smu_status dummy_get_dpm_clock_table(struct pp_smu *pp,
+static enum pp_smu_status dummy_get_dpm_clock_table(struct pp_smu *pp,
struct dpm_clocks *clock_table)
{
*clock_table = dummy_clocks;
return PP_SMU_RESULT_OK;
}
-struct pp_smu_funcs *dcn21_pp_smu_create(struct dc_context *ctx)
+static struct pp_smu_funcs *dcn21_pp_smu_create(struct dc_context *ctx)
{
struct pp_smu_funcs *pp_smu = kzalloc(sizeof(*pp_smu), GFP_KERNEL);
- pp_smu->ctx.ver = PP_SMU_VER_RN;
+ if (!pp_smu)
+ return pp_smu;
+
+ if (IS_FPGA_MAXIMUS_DC(ctx->dce_environment)) {
+ pp_smu->ctx.ver = PP_SMU_VER_RN;
+ pp_smu->rn_funcs.get_dpm_clock_table = dummy_get_dpm_clock_table;
+ pp_smu->rn_funcs.set_wm_ranges = dummy_set_wm_ranges;
+ } else {
- pp_smu->rn_funcs.get_dpm_clock_table = dummy_get_dpm_clock_table;
- pp_smu->rn_funcs.set_wm_ranges = dummy_set_wm_ranges;
+ dm_pp_get_funcs(ctx, pp_smu);
+
+ if (pp_smu->ctx.ver != PP_SMU_VER_RN)
+ pp_smu = memset(pp_smu, 0, sizeof(struct pp_smu_funcs));
+ }
return pp_smu;
}
-void dcn21_pp_smu_destroy(struct pp_smu_funcs **pp_smu)
+static void dcn21_pp_smu_destroy(struct pp_smu_funcs **pp_smu)
{
if (pp_smu && *pp_smu) {
kfree(*pp_smu);
diff --git a/drivers/gpu/drm/amd/display/dc/dm_pp_smu.h b/drivers/gpu/drm/amd/display/dc/dm_pp_smu.h
index ab5caa231872..65763e558673 100644
--- a/drivers/gpu/drm/amd/display/dc/dm_pp_smu.h
+++ b/drivers/gpu/drm/amd/display/dc/dm_pp_smu.h
@@ -250,7 +250,7 @@ struct pp_smu_funcs_nv {
#if defined(CONFIG_DRM_AMD_DC_DCN2_1)
#define PP_SMU_NUM_SOCCLK_DPM_LEVELS 8
-#define PP_SMU_NUM_DCFCLK_DPM_LEVELS 4
+#define PP_SMU_NUM_DCFCLK_DPM_LEVELS 8
#define PP_SMU_NUM_FCLK_DPM_LEVELS 4
#define PP_SMU_NUM_MEMCLK_DPM_LEVELS 4
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_mode_vba_21.c b/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_mode_vba_21.c
index 456cd0e3289c..3b6ed60dcd35 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_mode_vba_21.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_mode_vba_21.c
@@ -39,9 +39,6 @@
* ways. Unless there is something clearly wrong with it the code should
* remain as-is as it provides us with a guarantee from HW that it is correct.
*/
-
-typedef unsigned int uint;
-
typedef struct {
double DPPCLK;
double DISPCLK;
@@ -4774,7 +4771,7 @@ void dml21_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l
mode_lib->vba.MaximumReadBandwidthWithoutPrefetch = 0.0;
mode_lib->vba.MaximumReadBandwidthWithPrefetch = 0.0;
for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) {
- uint m;
+ unsigned int m;
locals->cursor_bw[k] = 0;
locals->cursor_bw_pre[k] = 0;
@@ -5285,7 +5282,7 @@ static void CalculateWatermarksAndDRAMSpeedChangeSupport(
double SecondMinActiveDRAMClockChangeMarginOneDisplayInVBLank;
double FullDETBufferingTimeYStutterCriticalPlane = 0;
double TimeToFinishSwathTransferStutterCriticalPlane = 0;
- uint k, j;
+ unsigned int k, j;
mode_lib->vba.TotalActiveDPP = 0;
mode_lib->vba.TotalDCCActiveDPP = 0;
@@ -5507,7 +5504,7 @@ static void CalculateDCFCLKDeepSleep(
double DPPCLK[],
double *DCFCLKDeepSleep)
{
- uint k;
+ unsigned int k;
double DisplayPipeLineDeliveryTimeLuma;
double DisplayPipeLineDeliveryTimeChroma;
//double DCFCLKDeepSleepPerPlane[DC__NUM_DPP__MAX];
@@ -5727,7 +5724,7 @@ static void CalculatePixelDeliveryTimes(
double DisplayPipeRequestDeliveryTimeChromaPrefetch[])
{
double req_per_swath_ub;
- uint k;
+ unsigned int k;
for (k = 0; k < NumberOfActivePlanes; ++k) {
if (VRatio[k] <= 1) {
@@ -5869,7 +5866,7 @@ static void CalculateMetaAndPTETimes(
unsigned int dpte_groups_per_row_chroma_ub;
unsigned int num_group_per_lower_vm_stage;
unsigned int num_req_per_lower_vm_stage;
- uint k;
+ unsigned int k;
for (k = 0; k < NumberOfActivePlanes; ++k) {
if (GPUVMEnable == true) {
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 f4c1ef9046bf..83f84cdd4055 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
@@ -269,7 +269,7 @@ struct writeback_st {
struct _vcs_dpi_display_output_params_st {
int dp_lanes;
- int output_bpp;
+ double output_bpp;
int dsc_enable;
int wb_enable;
int num_active_wb;
diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.c b/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.c
index 65cf4edddaff..362dc6ea98ae 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.c
@@ -434,6 +434,8 @@ static void fetch_pipe_params(struct display_mode_lib *mode_lib)
dst->odm_combine;
mode_lib->vba.OutputFormat[mode_lib->vba.NumberOfActivePlanes] =
(enum output_format_class) (dout->output_format);
+ mode_lib->vba.OutputBpp[mode_lib->vba.NumberOfActivePlanes] =
+ dout->output_bpp;
mode_lib->vba.Output[mode_lib->vba.NumberOfActivePlanes] =
(enum output_encoder_class) (dout->output_type);
diff --git a/drivers/gpu/drm/amd/display/dc/dsc/rc_calc.c b/drivers/gpu/drm/amd/display/dc/dsc/rc_calc.c
index ca51e83f8764..76c4b12d6824 100644
--- a/drivers/gpu/drm/amd/display/dc/dsc/rc_calc.c
+++ b/drivers/gpu/drm/amd/display/dc/dsc/rc_calc.c
@@ -177,7 +177,6 @@ void calc_rc_params(struct rc_params *rc, enum colour_mode cm, enum bits_per_com
{
float bpp_group;
float initial_xmit_delay_factor;
- int source_bpp;
int padding_pixels;
int i;
@@ -217,8 +216,6 @@ void calc_rc_params(struct rc_params *rc, enum colour_mode cm, enum bits_per_com
rc->initial_xmit_delay++;
}
- source_bpp = MODE_SELECT(bpc * 3, bpc * 2, bpc * 1.5);
-
rc->flatness_min_qp = ((bpc == BPC_8) ? (3) : ((bpc == BPC_10) ? (7) : (11))) - ((minor_version == 1 && cm == CM_444) ? 1 : 0);
rc->flatness_max_qp = ((bpc == BPC_8) ? (12) : ((bpc == BPC_10) ? (16) : (20))) - ((minor_version == 1 && cm == CM_444) ? 1 : 0);
rc->flatness_det_thresh = 2 << (bpc - 8);
diff --git a/drivers/gpu/drm/amd/display/dc/inc/dc_link_dp.h b/drivers/gpu/drm/amd/display/dc/inc/dc_link_dp.h
index 08a4df2c61a8..967706e7898e 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/dc_link_dp.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/dc_link_dp.h
@@ -43,6 +43,9 @@ bool dp_verify_link_cap_with_retries(
struct dc_link_settings *known_limit_link_setting,
int attempts);
+bool dp_verify_mst_link_cap(
+ struct dc_link *link);
+
bool dp_validate_mode_timing(
struct dc_link *link,
const struct dc_crtc_timing *timing);
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h b/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h
index 76f9ad1b23df..f2e21cb9fbd5 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h
@@ -47,7 +47,7 @@
#ifdef CONFIG_DRM_AMD_DC_DCN2_1
/* Will these bw structures be ASIC specific? */
-#define MAX_NUM_DPM_LVL 4
+#define MAX_NUM_DPM_LVL 8
#define WM_SET_COUNT 4
@@ -149,6 +149,7 @@ struct wm_table {
struct clk_bw_params {
unsigned int vram_type;
unsigned int num_channels;
+ unsigned int dispclk_vco_khz;
struct clk_limit_table clk_table;
struct wm_table wm_table;
};
@@ -180,12 +181,16 @@ struct clk_mgr_funcs {
struct dc_state *context,
enum dc_clock_type clock_type,
struct dc_clock_config *clock_cfg);
+
+ bool (*are_clock_states_equal) (struct dc_clocks *a,
+ struct dc_clocks *b);
};
struct clk_mgr {
struct dc_context *ctx;
struct clk_mgr_funcs *funcs;
struct dc_clocks clks;
+ bool psr_allow_active_cache;
int dprefclk_khz; // Used by program pixel clock in clock source funcs, need to figureout where this goes
#ifdef CONFIG_DRM_AMD_DC_DCN2_1
struct clk_bw_params *bw_params;
@@ -199,4 +204,8 @@ struct clk_mgr *dc_clk_mgr_create(struct dc_context *ctx, struct pp_smu_funcs *p
void dc_destroy_clk_mgr(struct clk_mgr *clk_mgr);
+void clk_mgr_exit_optimized_pwr_state(const struct dc *dc, struct clk_mgr *clk_mgr);
+
+void clk_mgr_optimize_pwr_state(const struct dc *dc, struct clk_mgr *clk_mgr);
+
#endif /* __DAL_CLK_MGR_H__ */
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr_internal.h b/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr_internal.h
index 213046de1675..2e8cd7956a17 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr_internal.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr_internal.h
@@ -184,6 +184,21 @@ struct clk_mgr_registers {
uint32_t MP1_SMN_C2PMSG_91;
};
+enum clock_type {
+ clock_type_dispclk = 1,
+ clock_type_dcfclk,
+ clock_type_socclk,
+ clock_type_pixelclk,
+ clock_type_phyclk,
+ clock_type_dppclk,
+ clock_type_fclk,
+ clock_type_dcfdsclk,
+ clock_type_dscclk,
+ clock_type_uclk,
+ clock_type_dramclk,
+};
+
+
struct state_dependent_clocks {
int display_clk_khz;
int pixel_clk_khz;
@@ -281,8 +296,14 @@ static inline bool should_set_clock(bool safe_to_lower, int calc_clk, int cur_cl
static inline bool should_update_pstate_support(bool safe_to_lower, bool calc_support, bool cur_support)
{
- // Whenever we are transitioning pstate support, we always want to notify prior to committing state
- return (calc_support != cur_support) ? !safe_to_lower : false;
+ if (cur_support != calc_support) {
+ if (calc_support == true && safe_to_lower)
+ return true;
+ else if (calc_support == false && !safe_to_lower)
+ return true;
+ }
+
+ return false;
}
int clk_mgr_helper_get_active_display_cnt(
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/link_encoder.h b/drivers/gpu/drm/amd/display/dc/inc/hw/link_encoder.h
index abb4e4237fb6..b21909216fb6 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/link_encoder.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/link_encoder.h
@@ -184,6 +184,10 @@ struct link_encoder_funcs {
bool (*fec_is_active)(struct link_encoder *enc);
#endif
bool (*is_in_alt_mode) (struct link_encoder *enc);
+
+ void (*get_max_link_cap)(struct link_encoder *enc,
+ struct dc_link_settings *link_settings);
+
enum signal_type (*get_dig_mode)(
struct link_encoder *enc);
};
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/opp.h b/drivers/gpu/drm/amd/display/dc/inc/hw/opp.h
index 957e9047381a..18def2b6fafe 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/opp.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/opp.h
@@ -208,6 +208,7 @@ struct output_pixel_processor {
struct mpc_tree mpc_tree_params;
bool mpcc_disconnect_pending[MAX_PIPES];
const struct opp_funcs *funcs;
+ uint32_t dyn_expansion;
};
enum fmt_stereo_action {
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/stream_encoder.h b/drivers/gpu/drm/amd/display/dc/inc/hw/stream_encoder.h
index fe9b7a10a1c3..6305e388612a 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/stream_encoder.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/stream_encoder.h
@@ -214,6 +214,11 @@ struct stream_encoder_funcs {
unsigned int (*dig_source_otg)(
struct stream_encoder *enc);
+ bool (*dp_get_pixel_format)(
+ struct stream_encoder *enc,
+ enum dc_pixel_encoding *encoding,
+ enum dc_color_depth *depth);
+
#if defined(CONFIG_DRM_AMD_DC_DCN2_0)
#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT
void (*enc_read_state)(struct stream_encoder *enc, struct enc_state *s);
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/timing_generator.h b/drivers/gpu/drm/amd/display/dc/inc/hw/timing_generator.h
index 6196cc32356e..27c73caf74ee 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/timing_generator.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/timing_generator.h
@@ -261,6 +261,8 @@ struct timing_generator_funcs {
void (*program_manual_trigger)(struct timing_generator *optc);
void (*setup_manual_trigger)(struct timing_generator *optc);
+ bool (*get_hw_timing)(struct timing_generator *optc,
+ struct dc_crtc_timing *hw_crtc_timing);
void (*set_vtg_params)(struct timing_generator *optc,
const struct dc_crtc_timing *dc_crtc_timing);
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 cbac3b61da94..e775d7aa062f 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h
@@ -232,6 +232,13 @@ struct hw_sequencer_funcs {
struct dc *dc,
struct dc_state *context);
+ void (*exit_optimized_pwr_state)(
+ const struct dc *dc,
+ struct dc_state *context);
+ void (*optimize_pwr_state)(
+ const struct dc *dc,
+ struct dc_state *context);
+
#if defined(CONFIG_DRM_AMD_DC_DCN2_0)
bool (*update_bandwidth)(
struct dc *dc,
@@ -324,10 +331,12 @@ struct hw_sequencer_funcs {
struct dc_state *context);
void (*update_writeback)(struct dc *dc,
const struct dc_stream_status *stream_status,
- struct dc_writeback_info *wb_info);
+ struct dc_writeback_info *wb_info,
+ struct dc_state *context);
void (*enable_writeback)(struct dc *dc,
const struct dc_stream_status *stream_status,
- struct dc_writeback_info *wb_info);
+ struct dc_writeback_info *wb_info,
+ struct dc_state *context);
void (*disable_writeback)(struct dc *dc,
unsigned int dwb_pipe_inst);
#endif
diff --git a/drivers/gpu/drm/amd/display/include/ddc_service_types.h b/drivers/gpu/drm/amd/display/include/ddc_service_types.h
index 18961707db23..9ad49da50a17 100644
--- a/drivers/gpu/drm/amd/display/include/ddc_service_types.h
+++ b/drivers/gpu/drm/amd/display/include/ddc_service_types.h
@@ -31,6 +31,8 @@
#define DP_BRANCH_DEVICE_ID_0022B9 0x0022B9
#define DP_BRANCH_DEVICE_ID_00001A 0x00001A
#define DP_BRANCH_DEVICE_ID_0080E1 0x0080e1
+#define DP_BRANCH_DEVICE_ID_90CC24 0x90CC24
+#define DP_BRANCH_DEVICE_ID_00E04C 0x00E04C
enum ddc_result {
DDC_RESULT_UNKNOWN = 0,
diff --git a/drivers/gpu/drm/amd/display/modules/freesync/freesync.c b/drivers/gpu/drm/amd/display/modules/freesync/freesync.c
index 5e5ce9e5eab7..9d68cfecd472 100644
--- a/drivers/gpu/drm/amd/display/modules/freesync/freesync.c
+++ b/drivers/gpu/drm/amd/display/modules/freesync/freesync.c
@@ -232,6 +232,10 @@ static void update_v_total_for_static_ramp(
current_duration_in_us) * (stream->timing.pix_clk_100hz / 10)),
stream->timing.h_total), 1000);
+ /* v_total cannot be less than nominal */
+ if (v_total < stream->timing.v_total)
+ v_total = stream->timing.v_total;
+
in_out_vrr->adjust.v_total_min = v_total;
in_out_vrr->adjust.v_total_max = v_total;
}
@@ -741,6 +745,10 @@ void mod_freesync_build_vrr_params(struct mod_freesync *mod_freesync,
nominal_field_rate_in_uhz =
mod_freesync_calc_nominal_field_rate(stream);
+ /* Rounded to the nearest Hz */
+ nominal_field_rate_in_uhz = 1000000ULL *
+ div_u64(nominal_field_rate_in_uhz + 500000, 1000000);
+
min_refresh_in_uhz = in_config->min_refresh_in_uhz;
max_refresh_in_uhz = in_config->max_refresh_in_uhz;
@@ -979,13 +987,9 @@ void mod_freesync_get_settings(struct mod_freesync *mod_freesync,
unsigned int *inserted_frames,
unsigned int *inserted_duration_in_us)
{
- struct core_freesync *core_freesync = NULL;
-
if (mod_freesync == NULL)
return;
- core_freesync = MOD_FREESYNC_TO_CORE(mod_freesync);
-
if (vrr->supported) {
*v_total_min = vrr->adjust.v_total_min;
*v_total_max = vrr->adjust.v_total_max;
@@ -1000,14 +1004,13 @@ unsigned long long mod_freesync_calc_nominal_field_rate(
const struct dc_stream_state *stream)
{
unsigned long long nominal_field_rate_in_uhz = 0;
+ unsigned int total = stream->timing.h_total * stream->timing.v_total;
- /* Calculate nominal field rate for stream */
+ /* Calculate nominal field rate for stream, rounded up to nearest integer */
nominal_field_rate_in_uhz = stream->timing.pix_clk_100hz / 10;
nominal_field_rate_in_uhz *= 1000ULL * 1000ULL * 1000ULL;
- nominal_field_rate_in_uhz = div_u64(nominal_field_rate_in_uhz,
- stream->timing.h_total);
- nominal_field_rate_in_uhz = div_u64(nominal_field_rate_in_uhz,
- stream->timing.v_total);
+
+ nominal_field_rate_in_uhz = div_u64(nominal_field_rate_in_uhz, total);
return nominal_field_rate_in_uhz;
}
diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.h b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.h
index 402bb7999093..5664bc0b5bd0 100644
--- a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.h
+++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.h
@@ -176,7 +176,7 @@ enum mod_hdcp_hdcp1_state_id {
H1_A0_WAIT_FOR_ACTIVE_RX,
H1_A1_EXCHANGE_KSVS,
H1_A2_COMPUTATIONS_A3_VALIDATE_RX_A6_TEST_FOR_REPEATER,
- H1_A45_AUTHENICATED,
+ H1_A45_AUTHENTICATED,
H1_A8_WAIT_FOR_READY,
H1_A9_READ_KSV_LIST,
HDCP1_STATE_END = H1_A9_READ_KSV_LIST
@@ -188,7 +188,7 @@ enum mod_hdcp_hdcp1_dp_state_id {
D1_A1_EXCHANGE_KSVS,
D1_A23_WAIT_FOR_R0_PRIME,
D1_A2_COMPUTATIONS_A3_VALIDATE_RX_A5_TEST_FOR_REPEATER,
- D1_A4_AUTHENICATED,
+ D1_A4_AUTHENTICATED,
D1_A6_WAIT_FOR_READY,
D1_A7_READ_KSV_LIST,
HDCP1_DP_STATE_END = D1_A7_READ_KSV_LIST,
diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp1_execution.c b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp1_execution.c
index 9e7302eac299..3db4a7da414f 100644
--- a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp1_execution.c
+++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp1_execution.c
@@ -476,7 +476,7 @@ enum mod_hdcp_status mod_hdcp_hdcp1_execution(struct mod_hdcp *hdcp,
status = computations_validate_rx_test_for_repeater(hdcp,
event_ctx, input);
break;
- case H1_A45_AUTHENICATED:
+ case H1_A45_AUTHENTICATED:
status = authenticated(hdcp, event_ctx, input);
break;
case H1_A8_WAIT_FOR_READY:
@@ -513,7 +513,7 @@ extern enum mod_hdcp_status mod_hdcp_hdcp1_dp_execution(struct mod_hdcp *hdcp,
status = computations_validate_rx_test_for_repeater(
hdcp, event_ctx, input);
break;
- case D1_A4_AUTHENICATED:
+ case D1_A4_AUTHENTICATED:
status = authenticated_dp(hdcp, event_ctx, input);
break;
case D1_A6_WAIT_FOR_READY:
diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp1_transition.c b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp1_transition.c
index 1d187809b709..136b8011ff3f 100644
--- a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp1_transition.c
+++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp1_transition.c
@@ -81,11 +81,11 @@ enum mod_hdcp_status mod_hdcp_hdcp1_transition(struct mod_hdcp *hdcp,
set_state_id(hdcp, output, H1_A8_WAIT_FOR_READY);
} else {
callback_in_ms(0, output);
- set_state_id(hdcp, output, H1_A45_AUTHENICATED);
+ set_state_id(hdcp, output, H1_A45_AUTHENTICATED);
HDCP_FULL_DDC_TRACE(hdcp);
}
break;
- case H1_A45_AUTHENICATED:
+ case H1_A45_AUTHENTICATED:
if (input->link_maintenance != PASS) {
/* 1A-07: consider invalid ri' a failure */
/* 1A-07a: consider read ri' not returned a failure */
@@ -129,7 +129,7 @@ enum mod_hdcp_status mod_hdcp_hdcp1_transition(struct mod_hdcp *hdcp,
break;
}
callback_in_ms(0, output);
- set_state_id(hdcp, output, H1_A45_AUTHENICATED);
+ set_state_id(hdcp, output, H1_A45_AUTHENTICATED);
HDCP_FULL_DDC_TRACE(hdcp);
break;
default:
@@ -224,11 +224,11 @@ enum mod_hdcp_status mod_hdcp_hdcp1_dp_transition(struct mod_hdcp *hdcp,
set_watchdog_in_ms(hdcp, 5000, output);
set_state_id(hdcp, output, D1_A6_WAIT_FOR_READY);
} else {
- set_state_id(hdcp, output, D1_A4_AUTHENICATED);
+ set_state_id(hdcp, output, D1_A4_AUTHENTICATED);
HDCP_FULL_DDC_TRACE(hdcp);
}
break;
- case D1_A4_AUTHENICATED:
+ case D1_A4_AUTHENTICATED:
if (input->link_integiry_check != PASS ||
input->reauth_request_check != PASS) {
/* 1A-07: restart hdcp on a link integrity failure */
@@ -295,7 +295,7 @@ enum mod_hdcp_status mod_hdcp_hdcp1_dp_transition(struct mod_hdcp *hdcp,
fail_and_restart_in_ms(0, &status, output);
break;
}
- set_state_id(hdcp, output, D1_A4_AUTHENICATED);
+ set_state_id(hdcp, output, D1_A4_AUTHENTICATED);
HDCP_FULL_DDC_TRACE(hdcp);
break;
default:
diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_log.c b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_log.c
index d868f556d180..3982ced5f969 100644
--- a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_log.c
+++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_log.c
@@ -136,8 +136,8 @@ char *mod_hdcp_state_id_to_str(int32_t id)
return "H1_A1_EXCHANGE_KSVS";
case H1_A2_COMPUTATIONS_A3_VALIDATE_RX_A6_TEST_FOR_REPEATER:
return "H1_A2_COMPUTATIONS_A3_VALIDATE_RX_A6_TEST_FOR_REPEATER";
- case H1_A45_AUTHENICATED:
- return "H1_A45_AUTHENICATED";
+ case H1_A45_AUTHENTICATED:
+ return "H1_A45_AUTHENTICATED";
case H1_A8_WAIT_FOR_READY:
return "H1_A8_WAIT_FOR_READY";
case H1_A9_READ_KSV_LIST:
@@ -150,8 +150,8 @@ char *mod_hdcp_state_id_to_str(int32_t id)
return "D1_A23_WAIT_FOR_R0_PRIME";
case D1_A2_COMPUTATIONS_A3_VALIDATE_RX_A5_TEST_FOR_REPEATER:
return "D1_A2_COMPUTATIONS_A3_VALIDATE_RX_A5_TEST_FOR_REPEATER";
- case D1_A4_AUTHENICATED:
- return "D1_A4_AUTHENICATED";
+ case D1_A4_AUTHENTICATED:
+ return "D1_A4_AUTHENTICATED";
case D1_A6_WAIT_FOR_READY:
return "D1_A6_WAIT_FOR_READY";
case D1_A7_READ_KSV_LIST:
diff --git a/drivers/gpu/drm/amd/display/modules/inc/mod_info_packet.h b/drivers/gpu/drm/amd/display/modules/inc/mod_info_packet.h
index d930bdecb117..ca8ce3c55337 100644
--- a/drivers/gpu/drm/amd/display/modules/inc/mod_info_packet.h
+++ b/drivers/gpu/drm/amd/display/modules/inc/mod_info_packet.h
@@ -35,4 +35,7 @@ struct mod_vrr_params;
void mod_build_vsc_infopacket(const struct dc_stream_state *stream,
struct dc_info_packet *info_packet);
+void mod_build_hf_vsif_infopacket(const struct dc_stream_state *stream,
+ struct dc_info_packet *info_packet, int ALLMEnabled, int ALLMValue);
+
#endif
diff --git a/drivers/gpu/drm/amd/display/modules/info_packet/info_packet.c b/drivers/gpu/drm/amd/display/modules/info_packet/info_packet.c
index d885d642ed7f..db6b08f6d093 100644
--- a/drivers/gpu/drm/amd/display/modules/info_packet/info_packet.c
+++ b/drivers/gpu/drm/amd/display/modules/info_packet/info_packet.c
@@ -31,6 +31,7 @@
#include "dc.h"
#define HDMI_INFOFRAME_TYPE_VENDOR 0x81
+#define HF_VSIF_VERSION 1
// VTEM Byte Offset
#define VTEM_PB0 0
@@ -395,3 +396,100 @@ void mod_build_vsc_infopacket(const struct dc_stream_state *stream,
}
+/**
+ *****************************************************************************
+ * Function: mod_build_hf_vsif_infopacket
+ *
+ * @brief
+ * Prepare HDMI Vendor Specific info frame.
+ * Follows HDMI Spec to build up Vendor Specific info frame
+ *
+ * @param [in] stream: contains data we may need to construct VSIF (i.e. timing_3d_format, etc.)
+ * @param [out] info_packet: output structure where to store VSIF
+ *****************************************************************************
+ */
+void mod_build_hf_vsif_infopacket(const struct dc_stream_state *stream,
+ struct dc_info_packet *info_packet, int ALLMEnabled, int ALLMValue)
+{
+ unsigned int length = 5;
+ bool hdmi_vic_mode = false;
+ uint8_t checksum = 0;
+ uint32_t i = 0;
+ enum dc_timing_3d_format format;
+ bool bALLM = (bool)ALLMEnabled;
+ bool bALLMVal = (bool)ALLMValue;
+
+ info_packet->valid = false;
+ format = stream->timing.timing_3d_format;
+ if (stream->view_format == VIEW_3D_FORMAT_NONE)
+ format = TIMING_3D_FORMAT_NONE;
+
+ if (stream->timing.hdmi_vic != 0
+ && stream->timing.h_total >= 3840
+ && stream->timing.v_total >= 2160
+ && format == TIMING_3D_FORMAT_NONE)
+ hdmi_vic_mode = true;
+
+ if ((format == TIMING_3D_FORMAT_NONE) && !hdmi_vic_mode && !bALLM)
+ return;
+
+ info_packet->sb[1] = 0x03;
+ info_packet->sb[2] = 0x0C;
+ info_packet->sb[3] = 0x00;
+
+ if (bALLM) {
+ info_packet->sb[1] = 0xD8;
+ info_packet->sb[2] = 0x5D;
+ info_packet->sb[3] = 0xC4;
+ info_packet->sb[4] = HF_VSIF_VERSION;
+ }
+
+ if (format != TIMING_3D_FORMAT_NONE)
+ info_packet->sb[4] = (2 << 5);
+
+ else if (hdmi_vic_mode)
+ info_packet->sb[4] = (1 << 5);
+
+ switch (format) {
+ case TIMING_3D_FORMAT_HW_FRAME_PACKING:
+ case TIMING_3D_FORMAT_SW_FRAME_PACKING:
+ info_packet->sb[5] = (0x0 << 4);
+ break;
+
+ case TIMING_3D_FORMAT_SIDE_BY_SIDE:
+ case TIMING_3D_FORMAT_SBS_SW_PACKED:
+ info_packet->sb[5] = (0x8 << 4);
+ length = 6;
+ break;
+
+ case TIMING_3D_FORMAT_TOP_AND_BOTTOM:
+ case TIMING_3D_FORMAT_TB_SW_PACKED:
+ info_packet->sb[5] = (0x6 << 4);
+ break;
+
+ default:
+ break;
+ }
+
+ if (hdmi_vic_mode)
+ info_packet->sb[5] = stream->timing.hdmi_vic;
+
+ info_packet->hb0 = HDMI_INFOFRAME_TYPE_VENDOR;
+ info_packet->hb1 = 0x01;
+ info_packet->hb2 = (uint8_t) (length);
+
+ if (bALLM)
+ info_packet->sb[5] = (info_packet->sb[5] & ~0x02) | (bALLMVal << 1);
+
+ checksum += info_packet->hb0;
+ checksum += info_packet->hb1;
+ checksum += info_packet->hb2;
+
+ for (i = 1; i <= length; i++)
+ checksum += info_packet->sb[i];
+
+ info_packet->sb[0] = (uint8_t) (0x100 - checksum);
+
+ info_packet->valid = true;
+}
+
diff --git a/drivers/gpu/drm/amd/dkms/m4/devcgroup-check-permission.m4 b/drivers/gpu/drm/amd/dkms/m4/devcgroup-check-permission.m4
index 5668ea575163..282014b14193 100644
--- a/drivers/gpu/drm/amd/dkms/m4/devcgroup-check-permission.m4
+++ b/drivers/gpu/drm/amd/dkms/m4/devcgroup-check-permission.m4
@@ -1,18 +1,16 @@
dnl #
-dnl # commit ecf8fecb7828
-dnl # device_cgroup: prepare code for bpf-based device controller
-dnl # commit 237d9c962b357
-dnl # device_cgroup: Export __devcgroup_check_permission
+dnl # commit d16020d7429fffd47cfb2f3ab3b6b5b362108a6e
+dnl # device_cgroup: Export devcgroup_check_permission
dnl #
-AC_DEFUN([AC_AMDGPU___DEVCGROUP_CHECK_PERMISSION],
- [AC_MSG_CHECKING([whether __devcgroup_check_permission() is available])
+AC_DEFUN([AC_AMDGPU_DEVCGROUP_CHECK_PERMISSION],
+ [AC_MSG_CHECKING([whether devcgroup_check_permission() is available])
AC_KERNEL_TRY_COMPILE_SYMBOL([
#include <linux/device_cgroup.h>
], [
- __devcgroup_check_permission(0, 0, 0, 0);
- ], [__devcgroup_check_permission], [security/device_cgroup.c], [
+ devcgroup_check_permission(0, 0, 0, 0);
+ ], [devcgroup_check_permission], [security/device_cgroup.c], [
AC_MSG_RESULT(yes)
- AC_DEFINE(HAVE___DEVCGROUP_CHECK_PERMISSION, 1, [__devcgroup_check_permission() is available])
+ AC_DEFINE(HAVE_DEVCGROUP_CHECK_PERMISSION, 1, [devcgroup_check_permission() is available])
], [
AC_MSG_RESULT(no)
])
diff --git a/drivers/gpu/drm/amd/dkms/m4/kernel.m4 b/drivers/gpu/drm/amd/dkms/m4/kernel.m4
index 9dae77444689..9af54ef27038 100644
--- a/drivers/gpu/drm/amd/dkms/m4/kernel.m4
+++ b/drivers/gpu/drm/amd/dkms/m4/kernel.m4
@@ -166,7 +166,7 @@ AC_DEFUN([AC_CONFIG_KERNEL], [
AC_AMDGPU_KTIME_GET_RAW_NS
AC_AMDGPU_KTIME_GET_BOOTTIME_NS
AC_AMDGPU_FB_OPS_FB_DEBUG_XX
- AC_AMDGPU___DEVCGROUP_CHECK_PERMISSION
+ AC_AMDGPU_DEVCGROUP_CHECK_PERMISSION
AC_AMDGPU_DRIVER_ATOMIC
AC_AMDGPU_DRM_DEVICE_DRIVER_FEATURES
AC_AMDGPU_IOMMU_GET_DOMAIN_FOR_DEV
diff --git a/drivers/gpu/drm/amd/include/asic_reg/gc/gc_9_0_offset.h b/drivers/gpu/drm/amd/include/asic_reg/gc/gc_9_0_offset.h
index ca16d9125fbc..2bfaaa8157d0 100644
--- a/drivers/gpu/drm/amd/include/asic_reg/gc/gc_9_0_offset.h
+++ b/drivers/gpu/drm/amd/include/asic_reg/gc/gc_9_0_offset.h
@@ -1146,7 +1146,14 @@
#define mmATC_L2_MEM_POWER_LS_BASE_IDX 0
#define mmATC_L2_CGTT_CLK_CTRL 0x080c
#define mmATC_L2_CGTT_CLK_CTRL_BASE_IDX 0
-
+#define mmATC_L2_CACHE_4K_EDC_INDEX 0x080e
+#define mmATC_L2_CACHE_4K_EDC_INDEX_BASE_IDX 0
+#define mmATC_L2_CACHE_2M_EDC_INDEX 0x080f
+#define mmATC_L2_CACHE_2M_EDC_INDEX_BASE_IDX 0
+#define mmATC_L2_CACHE_4K_EDC_CNT 0x0810
+#define mmATC_L2_CACHE_4K_EDC_CNT_BASE_IDX 0
+#define mmATC_L2_CACHE_2M_EDC_CNT 0x0811
+#define mmATC_L2_CACHE_2M_EDC_CNT_BASE_IDX 0
// addressBlock: gc_utcl2_vml2pfdec
// base address: 0xa100
@@ -1206,7 +1213,14 @@
#define mmVM_L2_CACHE_PARITY_CNTL_BASE_IDX 0
#define mmVM_L2_CGTT_CLK_CTRL 0x085e
#define mmVM_L2_CGTT_CLK_CTRL_BASE_IDX 0
-
+#define mmVM_L2_MEM_ECC_INDEX 0x0860
+#define mmVM_L2_MEM_ECC_INDEX_BASE_IDX 0
+#define mmVM_L2_WALKER_MEM_ECC_INDEX 0x0861
+#define mmVM_L2_WALKER_MEM_ECC_INDEX_BASE_IDX 0
+#define mmVM_L2_MEM_ECC_CNT 0x0862
+#define mmVM_L2_MEM_ECC_CNT_BASE_IDX 0
+#define mmVM_L2_WALKER_MEM_ECC_CNT 0x0863
+#define mmVM_L2_WALKER_MEM_ECC_CNT_BASE_IDX 0
// addressBlock: gc_utcl2_vml2vcdec
// base address: 0xa200
diff --git a/drivers/gpu/drm/amd/include/asic_reg/gc/gc_9_0_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/gc/gc_9_0_sh_mask.h
index 064c4bb1dc62..d4c613a85352 100644
--- a/drivers/gpu/drm/amd/include/asic_reg/gc/gc_9_0_sh_mask.h
+++ b/drivers/gpu/drm/amd/include/asic_reg/gc/gc_9_0_sh_mask.h
@@ -6661,7 +6661,6 @@
#define ATC_L2_CGTT_CLK_CTRL__SOFT_STALL_OVERRIDE_MASK 0x00FF0000L
#define ATC_L2_CGTT_CLK_CTRL__SOFT_OVERRIDE_MASK 0xFF000000L
-
// addressBlock: gc_utcl2_vml2pfdec
//VM_L2_CNTL
#define VM_L2_CNTL__ENABLE_L2_CACHE__SHIFT 0x0
@@ -6991,7 +6990,22 @@
#define VM_L2_CGTT_CLK_CTRL__MGLS_OVERRIDE_MASK 0x00008000L
#define VM_L2_CGTT_CLK_CTRL__SOFT_STALL_OVERRIDE_MASK 0x00FF0000L
#define VM_L2_CGTT_CLK_CTRL__SOFT_OVERRIDE_MASK 0xFF000000L
-
+//VM_L2_MEM_ECC_INDEX
+#define VM_L2_MEM_ECC_INDEX__INDEX__SHIFT 0x0
+#define VM_L2_MEM_ECC_INDEX__INDEX_MASK 0x000000FFL
+//VM_L2_WALKER_MEM_ECC_INDEX
+#define VM_L2_WALKER_MEM_ECC_INDEX__INDEX__SHIFT 0x0
+#define VM_L2_WALKER_MEM_ECC_INDEX__INDEX_MASK 0x000000FFL
+//VM_L2_MEM_ECC_CNT
+#define VM_L2_MEM_ECC_CNT__SEC_COUNT__SHIFT 0xc
+#define VM_L2_MEM_ECC_CNT__DED_COUNT__SHIFT 0xe
+#define VM_L2_MEM_ECC_CNT__SEC_COUNT_MASK 0x00003000L
+#define VM_L2_MEM_ECC_CNT__DED_COUNT_MASK 0x0000C000L
+//VM_L2_WALKER_MEM_ECC_CNT
+#define VM_L2_WALKER_MEM_ECC_CNT__SEC_COUNT__SHIFT 0xc
+#define VM_L2_WALKER_MEM_ECC_CNT__DED_COUNT__SHIFT 0xe
+#define VM_L2_WALKER_MEM_ECC_CNT__SEC_COUNT_MASK 0x00003000L
+#define VM_L2_WALKER_MEM_ECC_CNT__DED_COUNT_MASK 0x0000C000L
// addressBlock: gc_utcl2_vml2vcdec
//VM_CONTEXT0_CNTL
diff --git a/drivers/gpu/drm/amd/include/asic_reg/oss/osssys_4_0_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/oss/osssys_4_0_sh_mask.h
index dc9895a684fe..096d878eb1de 100644
--- a/drivers/gpu/drm/amd/include/asic_reg/oss/osssys_4_0_sh_mask.h
+++ b/drivers/gpu/drm/amd/include/asic_reg/oss/osssys_4_0_sh_mask.h
@@ -588,11 +588,15 @@
#define IH_STORM_CLIENT_LIST_CNTL__CLIENT30_IS_STORM_CLIENT_MASK 0x40000000L
#define IH_STORM_CLIENT_LIST_CNTL__CLIENT31_IS_STORM_CLIENT_MASK 0x80000000L
//IH_CLK_CTRL
+#define IH_CLK_CTRL__IH_RETRY_INT_CAM_MEM_CLK_SOFT_OVERRIDE__SHIFT 0x19
+#define IH_CLK_CTRL__IH_BUFFER_MEM_CLK_SOFT_OVERRIDE__SHIFT 0x1a
#define IH_CLK_CTRL__DBUS_MUX_CLK_SOFT_OVERRIDE__SHIFT 0x1b
#define IH_CLK_CTRL__OSSSYS_SHARE_CLK_SOFT_OVERRIDE__SHIFT 0x1c
#define IH_CLK_CTRL__LIMIT_SMN_CLK_SOFT_OVERRIDE__SHIFT 0x1d
#define IH_CLK_CTRL__DYN_CLK_SOFT_OVERRIDE__SHIFT 0x1e
#define IH_CLK_CTRL__REG_CLK_SOFT_OVERRIDE__SHIFT 0x1f
+#define IH_CLK_CTRL__IH_RETRY_INT_CAM_MEM_CLK_SOFT_OVERRIDE_MASK 0x02000000L
+#define IH_CLK_CTRL__IH_BUFFER_MEM_CLK_SOFT_OVERRIDE_MASK 0x04000000L
#define IH_CLK_CTRL__DBUS_MUX_CLK_SOFT_OVERRIDE_MASK 0x08000000L
#define IH_CLK_CTRL__OSSSYS_SHARE_CLK_SOFT_OVERRIDE_MASK 0x10000000L
#define IH_CLK_CTRL__LIMIT_SMN_CLK_SOFT_OVERRIDE_MASK 0x20000000L
diff --git a/drivers/gpu/drm/amd/include/asic_reg/smuio/smuio_11_0_0_offset.h b/drivers/gpu/drm/amd/include/asic_reg/smuio/smuio_11_0_0_offset.h
index d3876052562b..687d6843c258 100644
--- a/drivers/gpu/drm/amd/include/asic_reg/smuio/smuio_11_0_0_offset.h
+++ b/drivers/gpu/drm/amd/include/asic_reg/smuio/smuio_11_0_0_offset.h
@@ -121,6 +121,98 @@
#define mmCKSVII2C_IC_COMP_VERSION_BASE_IDX 0
#define mmCKSVII2C_IC_COMP_TYPE 0x006d
#define mmCKSVII2C_IC_COMP_TYPE_BASE_IDX 0
+#define mmCKSVII2C1_IC_CON 0x0080
+#define mmCKSVII2C1_IC_CON_BASE_IDX 0
+#define mmCKSVII2C1_IC_TAR 0x0081
+#define mmCKSVII2C1_IC_TAR_BASE_IDX 0
+#define mmCKSVII2C1_IC_SAR 0x0082
+#define mmCKSVII2C1_IC_SAR_BASE_IDX 0
+#define mmCKSVII2C1_IC_HS_MADDR 0x0083
+#define mmCKSVII2C1_IC_HS_MADDR_BASE_IDX 0
+#define mmCKSVII2C1_IC_DATA_CMD 0x0084
+#define mmCKSVII2C1_IC_DATA_CMD_BASE_IDX 0
+#define mmCKSVII2C1_IC_SS_SCL_HCNT 0x0085
+#define mmCKSVII2C1_IC_SS_SCL_HCNT_BASE_IDX 0
+#define mmCKSVII2C1_IC_SS_SCL_LCNT 0x0086
+#define mmCKSVII2C1_IC_SS_SCL_LCNT_BASE_IDX 0
+#define mmCKSVII2C1_IC_FS_SCL_HCNT 0x0087
+#define mmCKSVII2C1_IC_FS_SCL_HCNT_BASE_IDX 0
+#define mmCKSVII2C1_IC_FS_SCL_LCNT 0x0088
+#define mmCKSVII2C1_IC_FS_SCL_LCNT_BASE_IDX 0
+#define mmCKSVII2C1_IC_HS_SCL_HCNT 0x0089
+#define mmCKSVII2C1_IC_HS_SCL_HCNT_BASE_IDX 0
+#define mmCKSVII2C1_IC_HS_SCL_LCNT 0x008a
+#define mmCKSVII2C1_IC_HS_SCL_LCNT_BASE_IDX 0
+#define mmCKSVII2C1_IC_INTR_STAT 0x008b
+#define mmCKSVII2C1_IC_INTR_STAT_BASE_IDX 0
+#define mmCKSVII2C1_IC_INTR_MASK 0x008c
+#define mmCKSVII2C1_IC_INTR_MASK_BASE_IDX 0
+#define mmCKSVII2C1_IC_RAW_INTR_STAT 0x008d
+#define mmCKSVII2C1_IC_RAW_INTR_STAT_BASE_IDX 0
+#define mmCKSVII2C1_IC_RX_TL 0x008e
+#define mmCKSVII2C1_IC_RX_TL_BASE_IDX 0
+#define mmCKSVII2C1_IC_TX_TL 0x008f
+#define mmCKSVII2C1_IC_TX_TL_BASE_IDX 0
+#define mmCKSVII2C1_IC_CLR_INTR 0x0090
+#define mmCKSVII2C1_IC_CLR_INTR_BASE_IDX 0
+#define mmCKSVII2C1_IC_CLR_RX_UNDER 0x0091
+#define mmCKSVII2C1_IC_CLR_RX_UNDER_BASE_IDX 0
+#define mmCKSVII2C1_IC_CLR_RX_OVER 0x0092
+#define mmCKSVII2C1_IC_CLR_RX_OVER_BASE_IDX 0
+#define mmCKSVII2C1_IC_CLR_TX_OVER 0x0093
+#define mmCKSVII2C1_IC_CLR_TX_OVER_BASE_IDX 0
+#define mmCKSVII2C1_IC_CLR_RD_REQ 0x0094
+#define mmCKSVII2C1_IC_CLR_RD_REQ_BASE_IDX 0
+#define mmCKSVII2C1_IC_CLR_TX_ABRT 0x0095
+#define mmCKSVII2C1_IC_CLR_TX_ABRT_BASE_IDX 0
+#define mmCKSVII2C1_IC_CLR_RX_DONE 0x0096
+#define mmCKSVII2C1_IC_CLR_RX_DONE_BASE_IDX 0
+#define mmCKSVII2C1_IC_CLR_ACTIVITY 0x0097
+#define mmCKSVII2C1_IC_CLR_ACTIVITY_BASE_IDX 0
+#define mmCKSVII2C1_IC_CLR_STOP_DET 0x0098
+#define mmCKSVII2C1_IC_CLR_STOP_DET_BASE_IDX 0
+#define mmCKSVII2C1_IC_CLR_START_DET 0x0099
+#define mmCKSVII2C1_IC_CLR_START_DET_BASE_IDX 0
+#define mmCKSVII2C1_IC_CLR_GEN_CALL 0x009a
+#define mmCKSVII2C1_IC_CLR_GEN_CALL_BASE_IDX 0
+#define mmCKSVII2C1_IC_ENABLE 0x009b
+#define mmCKSVII2C1_IC_ENABLE_BASE_IDX 0
+#define mmCKSVII2C1_IC_STATUS 0x009c
+#define mmCKSVII2C1_IC_STATUS_BASE_IDX 0
+#define mmCKSVII2C1_IC_TXFLR 0x009d
+#define mmCKSVII2C1_IC_TXFLR_BASE_IDX 0
+#define mmCKSVII2C1_IC_RXFLR 0x009e
+#define mmCKSVII2C1_IC_RXFLR_BASE_IDX 0
+#define mmCKSVII2C1_IC_SDA_HOLD 0x009f
+#define mmCKSVII2C1_IC_SDA_HOLD_BASE_IDX 0
+#define mmCKSVII2C1_IC_TX_ABRT_SOURCE 0x00a0
+#define mmCKSVII2C1_IC_TX_ABRT_SOURCE_BASE_IDX 0
+#define mmCKSVII2C1_IC_SLV_DATA_NACK_ONLY 0x00a1
+#define mmCKSVII2C1_IC_SLV_DATA_NACK_ONLY_BASE_IDX 0
+#define mmCKSVII2C1_IC_DMA_CR 0x00a2
+#define mmCKSVII2C1_IC_DMA_CR_BASE_IDX 0
+#define mmCKSVII2C1_IC_DMA_TDLR 0x00a3
+#define mmCKSVII2C1_IC_DMA_TDLR_BASE_IDX 0
+#define mmCKSVII2C1_IC_DMA_RDLR 0x00a4
+#define mmCKSVII2C1_IC_DMA_RDLR_BASE_IDX 0
+#define mmCKSVII2C1_IC_SDA_SETUP 0x00a5
+#define mmCKSVII2C1_IC_SDA_SETUP_BASE_IDX 0
+#define mmCKSVII2C1_IC_ACK_GENERAL_CALL 0x00a6
+#define mmCKSVII2C1_IC_ACK_GENERAL_CALL_BASE_IDX 0
+#define mmCKSVII2C1_IC_ENABLE_STATUS 0x00a7
+#define mmCKSVII2C1_IC_ENABLE_STATUS_BASE_IDX 0
+#define mmCKSVII2C1_IC_FS_SPKLEN 0x00a8
+#define mmCKSVII2C1_IC_FS_SPKLEN_BASE_IDX 0
+#define mmCKSVII2C1_IC_HS_SPKLEN 0x00a9
+#define mmCKSVII2C1_IC_HS_SPKLEN_BASE_IDX 0
+#define mmCKSVII2C1_IC_CLR_RESTART_DET 0x00aa
+#define mmCKSVII2C1_IC_CLR_RESTART_DET_BASE_IDX 0
+#define mmCKSVII2C1_IC_COMP_PARAM_1 0x00ab
+#define mmCKSVII2C1_IC_COMP_PARAM_1_BASE_IDX 0
+#define mmCKSVII2C1_IC_COMP_VERSION 0x00ac
+#define mmCKSVII2C1_IC_COMP_VERSION_BASE_IDX 0
+#define mmCKSVII2C1_IC_COMP_TYPE 0x00ad
+#define mmCKSVII2C1_IC_COMP_TYPE_BASE_IDX 0
#define mmSMUIO_MP_RESET_INTR 0x00c1
#define mmSMUIO_MP_RESET_INTR_BASE_IDX 0
#define mmSMUIO_SOC_HALT 0x00c2
diff --git a/drivers/gpu/drm/amd/include/asic_reg/smuio/smuio_11_0_0_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/smuio/smuio_11_0_0_sh_mask.h
index f8afa3518bf2..6905a9618127 100644
--- a/drivers/gpu/drm/amd/include/asic_reg/smuio/smuio_11_0_0_sh_mask.h
+++ b/drivers/gpu/drm/amd/include/asic_reg/smuio/smuio_11_0_0_sh_mask.h
@@ -268,6 +268,182 @@
//CKSVII2C_IC_COMP_TYPE
#define CKSVII2C_IC_COMP_TYPE__COMP_TYPE__SHIFT 0x0
#define CKSVII2C_IC_COMP_TYPE__COMP_TYPE_MASK 0xFFFFFFFFL
+//CKSVII2C1_IC_CON
+#define CKSVII2C1_IC_CON__IC1_MASTER_MODE__SHIFT 0x0
+#define CKSVII2C1_IC_CON__IC1_MAX_SPEED_MODE__SHIFT 0x1
+#define CKSVII2C1_IC_CON__IC1_10BITADDR_SLAVE__SHIFT 0x3
+#define CKSVII2C1_IC_CON__IC1_10BITADDR_MASTER__SHIFT 0x4
+#define CKSVII2C1_IC_CON__IC1_RESTART_EN__SHIFT 0x5
+#define CKSVII2C1_IC_CON__IC1_SLAVE_DISABLE__SHIFT 0x6
+#define CKSVII2C1_IC_CON__STOP1_DET_IFADDRESSED__SHIFT 0x7
+#define CKSVII2C1_IC_CON__TX1_EMPTY_CTRL__SHIFT 0x8
+#define CKSVII2C1_IC_CON__RX1_FIFO_FULL_HLD_CTRL__SHIFT 0x9
+#define CKSVII2C1_IC_CON__IC1_MASTER_MODE_MASK 0x00000001L
+#define CKSVII2C1_IC_CON__IC1_MAX_SPEED_MODE_MASK 0x00000006L
+#define CKSVII2C1_IC_CON__IC1_10BITADDR_SLAVE_MASK 0x00000008L
+#define CKSVII2C1_IC_CON__IC1_10BITADDR_MASTER_MASK 0x00000010L
+#define CKSVII2C1_IC_CON__IC1_RESTART_EN_MASK 0x00000020L
+#define CKSVII2C1_IC_CON__IC1_SLAVE_DISABLE_MASK 0x00000040L
+#define CKSVII2C1_IC_CON__STOP1_DET_IFADDRESSED_MASK 0x00000080L
+#define CKSVII2C1_IC_CON__TX1_EMPTY_CTRL_MASK 0x00000100L
+#define CKSVII2C1_IC_CON__RX1_FIFO_FULL_HLD_CTRL_MASK 0x00000200L
+//CKSVII2C1_IC_TAR
+#define CKSVII2C1_IC_TAR__IC1_TAR__SHIFT 0x0
+#define CKSVII2C1_IC_TAR__GC1_OR_START__SHIFT 0xa
+#define CKSVII2C1_IC_TAR__SPECIAL1__SHIFT 0xb
+#define CKSVII2C1_IC_TAR__IC1_10BITADDR_MASTER__SHIFT 0xc
+#define CKSVII2C1_IC_TAR__IC1_TAR_MASK 0x000003FFL
+#define CKSVII2C1_IC_TAR__GC1_OR_START_MASK 0x00000400L
+#define CKSVII2C1_IC_TAR__SPECIAL1_MASK 0x00000800L
+#define CKSVII2C1_IC_TAR__IC1_10BITADDR_MASTER_MASK 0x00001000L
+//CKSVII2C1_IC_SAR
+#define CKSVII2C1_IC_SAR__IC1_SAR__SHIFT 0x0
+#define CKSVII2C1_IC_SAR__IC1_SAR_MASK 0x000003FFL
+//CKSVII2C1_IC_HS_MADDR
+#define CKSVII2C1_IC_HS_MADDR__IC1_HS_MADDR__SHIFT 0x0
+#define CKSVII2C1_IC_HS_MADDR__IC1_HS_MADDR_MASK 0x00000007L
+//CKSVII2C1_IC_DATA_CMD
+#define CKSVII2C1_IC_DATA_CMD__DAT1__SHIFT 0x0
+#define CKSVII2C1_IC_DATA_CMD__CMD1__SHIFT 0x8
+#define CKSVII2C1_IC_DATA_CMD__STOP1__SHIFT 0x9
+#define CKSVII2C1_IC_DATA_CMD__RESTART1__SHIFT 0xa
+#define CKSVII2C1_IC_DATA_CMD__DAT1_MASK 0x000000FFL
+#define CKSVII2C1_IC_DATA_CMD__CMD1_MASK 0x00000100L
+#define CKSVII2C1_IC_DATA_CMD__STOP1_MASK 0x00000200L
+#define CKSVII2C1_IC_DATA_CMD__RESTART1_MASK 0x00000400L
+//CKSVII2C1_IC_SS_SCL_HCNT
+#define CKSVII2C1_IC_SS_SCL_HCNT__IC1_SS_SCL_HCNT__SHIFT 0x0
+#define CKSVII2C1_IC_SS_SCL_HCNT__IC1_SS_SCL_HCNT_MASK 0x0000FFFFL
+//CKSVII2C1_IC_SS_SCL_LCNT
+#define CKSVII2C1_IC_SS_SCL_LCNT__IC1_SS_SCL_LCNT__SHIFT 0x0
+#define CKSVII2C1_IC_SS_SCL_LCNT__IC1_SS_SCL_LCNT_MASK 0x0000FFFFL
+//CKSVII2C1_IC_FS_SCL_HCNT
+#define CKSVII2C1_IC_FS_SCL_HCNT__IC1_FS_SCL_HCNT__SHIFT 0x0
+#define CKSVII2C1_IC_FS_SCL_HCNT__IC1_FS_SCL_HCNT_MASK 0x0000FFFFL
+//CKSVII2C1_IC_FS_SCL_LCNT
+#define CKSVII2C1_IC_FS_SCL_LCNT__IC1_FS_SCL_LCNT__SHIFT 0x0
+#define CKSVII2C1_IC_FS_SCL_LCNT__IC1_FS_SCL_LCNT_MASK 0x0000FFFFL
+//CKSVII2C1_IC_HS_SCL_HCNT
+#define CKSVII2C1_IC_HS_SCL_HCNT__IC1_HS_SCL_HCNT__SHIFT 0x0
+#define CKSVII2C1_IC_HS_SCL_HCNT__IC1_HS_SCL_HCNT_MASK 0x0000FFFFL
+//CKSVII2C1_IC_HS_SCL_LCNT
+#define CKSVII2C1_IC_HS_SCL_LCNT__IC1_HS_SCL_LCNT__SHIFT 0x0
+#define CKSVII2C1_IC_HS_SCL_LCNT__IC1_HS_SCL_LCNT_MASK 0x0000FFFFL
+//CKSVII2C1_IC_INTR_STAT
+#define CKSVII2C1_IC_INTR_STAT__R1_RX_UNDER__SHIFT 0x0
+#define CKSVII2C1_IC_INTR_STAT__R1_RX_OVER__SHIFT 0x1
+#define CKSVII2C1_IC_INTR_STAT__R1_RX_FULL__SHIFT 0x2
+#define CKSVII2C1_IC_INTR_STAT__R1_TX_OVER__SHIFT 0x3
+#define CKSVII2C1_IC_INTR_STAT__R1_TX_EMPTY__SHIFT 0x4
+#define CKSVII2C1_IC_INTR_STAT__R1_RD_REQ__SHIFT 0x5
+#define CKSVII2C1_IC_INTR_STAT__R1_TX_ABRT__SHIFT 0x6
+#define CKSVII2C1_IC_INTR_STAT__R1_RX_DONE__SHIFT 0x7
+#define CKSVII2C1_IC_INTR_STAT__R1_ACTIVITY__SHIFT 0x8
+#define CKSVII2C1_IC_INTR_STAT__R1_STOP_DET__SHIFT 0x9
+#define CKSVII2C1_IC_INTR_STAT__R1_START_DET__SHIFT 0xa
+#define CKSVII2C1_IC_INTR_STAT__R1_GEN_CALL__SHIFT 0xb
+#define CKSVII2C1_IC_INTR_STAT__R1_RESTART_DET__SHIFT 0xc
+#define CKSVII2C1_IC_INTR_STAT__R1_MST_ON_HOLD__SHIFT 0xd
+#define CKSVII2C1_IC_INTR_STAT__R1_RX_UNDER_MASK 0x00000001L
+#define CKSVII2C1_IC_INTR_STAT__R1_RX_OVER_MASK 0x00000002L
+#define CKSVII2C1_IC_INTR_STAT__R1_RX_FULL_MASK 0x00000004L
+#define CKSVII2C1_IC_INTR_STAT__R1_TX_OVER_MASK 0x00000008L
+#define CKSVII2C1_IC_INTR_STAT__R1_TX_EMPTY_MASK 0x00000010L
+#define CKSVII2C1_IC_INTR_STAT__R1_RD_REQ_MASK 0x00000020L
+#define CKSVII2C1_IC_INTR_STAT__R1_TX_ABRT_MASK 0x00000040L
+#define CKSVII2C1_IC_INTR_STAT__R1_RX_DONE_MASK 0x00000080L
+#define CKSVII2C1_IC_INTR_STAT__R1_ACTIVITY_MASK 0x00000100L
+#define CKSVII2C1_IC_INTR_STAT__R1_STOP_DET_MASK 0x00000200L
+#define CKSVII2C1_IC_INTR_STAT__R1_START_DET_MASK 0x00000400L
+#define CKSVII2C1_IC_INTR_STAT__R1_GEN_CALL_MASK 0x00000800L
+#define CKSVII2C1_IC_INTR_STAT__R1_RESTART_DET_MASK 0x00001000L
+#define CKSVII2C1_IC_INTR_STAT__R1_MST_ON_HOLD_MASK 0x00002000L
+//CKSVII2C1_IC_INTR_MASK
+#define CKSVII2C1_IC_INTR_MASK__M1_RX_UNDER__SHIFT 0x0
+#define CKSVII2C1_IC_INTR_MASK__M1_RX_OVER__SHIFT 0x1
+#define CKSVII2C1_IC_INTR_MASK__M1_RX_FULL__SHIFT 0x2
+#define CKSVII2C1_IC_INTR_MASK__M1_TX_OVER__SHIFT 0x3
+#define CKSVII2C1_IC_INTR_MASK__M1_TX_EMPTY__SHIFT 0x4
+#define CKSVII2C1_IC_INTR_MASK__M1_RD_REQ__SHIFT 0x5
+#define CKSVII2C1_IC_INTR_MASK__M1_TX_ABRT__SHIFT 0x6
+#define CKSVII2C1_IC_INTR_MASK__M1_RX_DONE__SHIFT 0x7
+#define CKSVII2C1_IC_INTR_MASK__M1_ACTIVITY__SHIFT 0x8
+#define CKSVII2C1_IC_INTR_MASK__M1_STOP_DET__SHIFT 0x9
+#define CKSVII2C1_IC_INTR_MASK__M1_START_DET__SHIFT 0xa
+#define CKSVII2C1_IC_INTR_MASK__M1_GEN_CALL__SHIFT 0xb
+#define CKSVII2C1_IC_INTR_MASK__M1_RESTART_DET__SHIFT 0xc
+#define CKSVII2C1_IC_INTR_MASK__M1_MST_ON_HOLD__SHIFT 0xd
+#define CKSVII2C1_IC_INTR_MASK__M1_RX_UNDER_MASK 0x00000001L
+#define CKSVII2C1_IC_INTR_MASK__M1_RX_OVER_MASK 0x00000002L
+#define CKSVII2C1_IC_INTR_MASK__M1_RX_FULL_MASK 0x00000004L
+#define CKSVII2C1_IC_INTR_MASK__M1_TX_OVER_MASK 0x00000008L
+#define CKSVII2C1_IC_INTR_MASK__M1_TX_EMPTY_MASK 0x00000010L
+#define CKSVII2C1_IC_INTR_MASK__M1_RD_REQ_MASK 0x00000020L
+#define CKSVII2C1_IC_INTR_MASK__M1_TX_ABRT_MASK 0x00000040L
+#define CKSVII2C1_IC_INTR_MASK__M1_RX_DONE_MASK 0x00000080L
+#define CKSVII2C1_IC_INTR_MASK__M1_ACTIVITY_MASK 0x00000100L
+#define CKSVII2C1_IC_INTR_MASK__M1_STOP_DET_MASK 0x00000200L
+#define CKSVII2C1_IC_INTR_MASK__M1_START_DET_MASK 0x00000400L
+#define CKSVII2C1_IC_INTR_MASK__M1_GEN_CALL_MASK 0x00000800L
+#define CKSVII2C1_IC_INTR_MASK__M1_RESTART_DET_MASK 0x00001000L
+#define CKSVII2C1_IC_INTR_MASK__M1_MST_ON_HOLD_MASK 0x00002000L
+//CKSVII2C1_IC_RAW_INTR_STAT
+//CKSVII2C1_IC_RX_TL
+//CKSVII2C1_IC_TX_TL
+//CKSVII2C1_IC_CLR_INTR
+//CKSVII2C1_IC_CLR_RX_UNDER
+//CKSVII2C1_IC_CLR_RX_OVER
+//CKSVII2C1_IC_CLR_TX_OVER
+//CKSVII2C1_IC_CLR_RD_REQ
+//CKSVII2C1_IC_CLR_TX_ABRT
+//CKSVII2C1_IC_CLR_RX_DONE
+//CKSVII2C1_IC_CLR_ACTIVITY
+//CKSVII2C1_IC_CLR_STOP_DET
+//CKSVII2C1_IC_CLR_START_DET
+//CKSVII2C1_IC_CLR_GEN_CALL
+//CKSVII2C1_IC_ENABLE
+#define CKSVII2C1_IC_ENABLE__ENABLE1__SHIFT 0x0
+#define CKSVII2C1_IC_ENABLE__ABORT1__SHIFT 0x1
+#define CKSVII2C1_IC_ENABLE__ENABLE1_MASK 0x00000001L
+#define CKSVII2C1_IC_ENABLE__ABORT1_MASK 0x00000002L
+//CKSVII2C1_IC_STATUS
+#define CKSVII2C1_IC_STATUS__ACTIVITY1__SHIFT 0x0
+#define CKSVII2C1_IC_STATUS__TFNF1__SHIFT 0x1
+#define CKSVII2C1_IC_STATUS__TFE1__SHIFT 0x2
+#define CKSVII2C1_IC_STATUS__RFNE1__SHIFT 0x3
+#define CKSVII2C1_IC_STATUS__RFF1__SHIFT 0x4
+#define CKSVII2C1_IC_STATUS__MST1_ACTIVITY__SHIFT 0x5
+#define CKSVII2C1_IC_STATUS__SLV1_ACTIVITY__SHIFT 0x6
+#define CKSVII2C1_IC_STATUS__ACTIVITY1_MASK 0x00000001L
+#define CKSVII2C1_IC_STATUS__TFNF1_MASK 0x00000002L
+#define CKSVII2C1_IC_STATUS__TFE1_MASK 0x00000004L
+#define CKSVII2C1_IC_STATUS__RFNE1_MASK 0x00000008L
+#define CKSVII2C1_IC_STATUS__RFF1_MASK 0x00000010L
+#define CKSVII2C1_IC_STATUS__MST1_ACTIVITY_MASK 0x00000020L
+#define CKSVII2C1_IC_STATUS__SLV1_ACTIVITY_MASK 0x00000040L
+//CKSVII2C1_IC_TXFLR
+//CKSVII2C1_IC_RXFLR
+//CKSVII2C1_IC_SDA_HOLD
+#define CKSVII2C1_IC_SDA_HOLD__IC1_SDA_HOLD__SHIFT 0x0
+#define CKSVII2C1_IC_SDA_HOLD__IC1_SDA_HOLD_MASK 0x00FFFFFFL
+//CKSVII2C1_IC_TX_ABRT_SOURCE
+//CKSVII2C1_IC_SLV_DATA_NACK_ONLY
+//CKSVII2C1_IC_DMA_CR
+//CKSVII2C1_IC_DMA_TDLR
+//CKSVII2C1_IC_DMA_RDLR
+//CKSVII2C1_IC_SDA_SETUP
+#define CKSVII2C1_IC_SDA_SETUP__SDA1_SETUP__SHIFT 0x0
+#define CKSVII2C1_IC_SDA_SETUP__SDA1_SETUP_MASK 0x000000FFL
+//CKSVII2C1_IC_ACK_GENERAL_CALL
+#define CKSVII2C1_IC_ACK_GENERAL_CALL__ACK1_GENERAL_CALL__SHIFT 0x0
+#define CKSVII2C1_IC_ACK_GENERAL_CALL__ACK1_GENERAL_CALL_MASK 0x00000001L
+//CKSVII2C1_IC_ENABLE_STATUS
+#define CKSVII2C1_IC_ENABLE_STATUS__IC1_EN__SHIFT 0x0
+#define CKSVII2C1_IC_ENABLE_STATUS__SLV1_RX_ABORTED__SHIFT 0x1
+#define CKSVII2C1_IC_ENABLE_STATUS__SLV1_FIFO_FILLED_AND_FLUSHED__SHIFT 0x2
+#define CKSVII2C1_IC_ENABLE_STATUS__IC1_EN_MASK 0x00000001L
+#define CKSVII2C1_IC_ENABLE_STATUS__SLV1_RX_ABORTED_MASK 0x00000002L
+#define CKSVII2C1_IC_ENABLE_STATUS__SLV1_FIFO_FILLED_AND_FLUSHED_MASK 0x00000004L
//SMUIO_MP_RESET_INTR
#define SMUIO_MP_RESET_INTR__SMUIO_MP_RESET_INTR__SHIFT 0x0
#define SMUIO_MP_RESET_INTR__SMUIO_MP_RESET_INTR_MASK 0x00000001L
diff --git a/drivers/gpu/drm/amd/include/kgd_kfd_interface.h b/drivers/gpu/drm/amd/include/kgd_kfd_interface.h
index 21eadd436778..3f8f63f965b3 100644
--- a/drivers/gpu/drm/amd/include/kgd_kfd_interface.h
+++ b/drivers/gpu/drm/amd/include/kgd_kfd_interface.h
@@ -300,12 +300,10 @@ struct kfd2kgd_calls {
uint32_t (*address_watch_get_offset)(struct kgd_dev *kgd,
unsigned int watch_point_id,
unsigned int reg_offset);
- bool (*get_atc_vmid_pasid_mapping_valid)(
+ bool (*get_atc_vmid_pasid_mapping_info)(
struct kgd_dev *kgd,
- uint8_t vmid);
- uint16_t (*get_atc_vmid_pasid_mapping_pasid)(
- struct kgd_dev *kgd,
- uint8_t vmid);
+ uint8_t vmid,
+ uint16_t *p_pasid);
/* No longer needed from GFXv9 onward. The scratch base address is
* passed to the shader by the CP. It's the user mode driver's
diff --git a/drivers/gpu/drm/amd/include/kgd_pp_interface.h b/drivers/gpu/drm/amd/include/kgd_pp_interface.h
index 27cf0afaa0b4..5902f80d1fce 100644
--- a/drivers/gpu/drm/amd/include/kgd_pp_interface.h
+++ b/drivers/gpu/drm/amd/include/kgd_pp_interface.h
@@ -179,6 +179,11 @@ enum pp_mp1_state {
PP_MP1_STATE_RESET,
};
+enum pp_df_cstate {
+ DF_CSTATE_DISALLOW = 0,
+ DF_CSTATE_ALLOW,
+};
+
#define PP_GROUP_MASK 0xF0000000
#define PP_GROUP_SHIFT 28
@@ -312,6 +317,7 @@ struct amd_pm_funcs {
int (*get_ppfeature_status)(void *handle, char *buf);
int (*set_ppfeature_status)(void *handle, uint64_t ppfeature_masks);
int (*asic_reset_mode_2)(void *handle);
+ int (*set_df_cstate)(void *handle, enum pp_df_cstate state);
};
#endif
diff --git a/drivers/gpu/drm/amd/powerplay/amd_powerplay.c b/drivers/gpu/drm/amd/powerplay/amd_powerplay.c
index fa8ad7db2b3a..83196b79edd5 100644
--- a/drivers/gpu/drm/amd/powerplay/amd_powerplay.c
+++ b/drivers/gpu/drm/amd/powerplay/amd_powerplay.c
@@ -1548,6 +1548,23 @@ static int pp_smu_i2c_bus_access(void *handle, bool acquire)
return ret;
}
+static int pp_set_df_cstate(void *handle, enum pp_df_cstate state)
+{
+ struct pp_hwmgr *hwmgr = handle;
+
+ if (!hwmgr)
+ return -EINVAL;
+
+ if (!hwmgr->pm_en || !hwmgr->hwmgr_func->set_df_cstate)
+ return 0;
+
+ mutex_lock(&hwmgr->smu_lock);
+ hwmgr->hwmgr_func->set_df_cstate(hwmgr, state);
+ mutex_unlock(&hwmgr->smu_lock);
+
+ return 0;
+}
+
static const struct amd_pm_funcs pp_dpm_funcs = {
.load_firmware = pp_dpm_load_fw,
.wait_for_fw_loading_complete = pp_dpm_fw_loading_complete,
@@ -1606,4 +1623,5 @@ static const struct amd_pm_funcs pp_dpm_funcs = {
.set_ppfeature_status = pp_set_ppfeature_status,
.asic_reset_mode_2 = pp_asic_reset_mode_2,
.smu_i2c_bus_access = pp_smu_i2c_bus_access,
+ .set_df_cstate = pp_set_df_cstate,
};
diff --git a/drivers/gpu/drm/amd/powerplay/amdgpu_smu.c b/drivers/gpu/drm/amd/powerplay/amdgpu_smu.c
index 0237e4d06aca..d0a25dd8fcfc 100644
--- a/drivers/gpu/drm/amd/powerplay/amdgpu_smu.c
+++ b/drivers/gpu/drm/amd/powerplay/amdgpu_smu.c
@@ -205,8 +205,7 @@ int smu_get_smc_version(struct smu_context *smu, uint32_t *if_version, uint32_t
int smu_set_soft_freq_range(struct smu_context *smu, enum smu_clk_type clk_type,
uint32_t min, uint32_t max)
{
- int ret = 0, clk_id = 0;
- uint32_t param;
+ int ret = 0;
if (min <= 0 && max <= 0)
return -EINVAL;
@@ -214,27 +213,7 @@ int smu_set_soft_freq_range(struct smu_context *smu, enum smu_clk_type clk_type,
if (!smu_clk_dpm_is_enabled(smu, clk_type))
return 0;
- clk_id = smu_clk_get_index(smu, clk_type);
- if (clk_id < 0)
- return clk_id;
-
- if (max > 0) {
- param = (uint32_t)((clk_id << 16) | (max & 0xffff));
- ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetSoftMaxByFreq,
- param);
- if (ret)
- return ret;
- }
-
- if (min > 0) {
- param = (uint32_t)((clk_id << 16) | (min & 0xffff));
- ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetSoftMinByFreq,
- param);
- if (ret)
- return ret;
- }
-
-
+ ret = smu_set_soft_freq_limited_range(smu, clk_type, min, max);
return ret;
}
@@ -733,6 +712,7 @@ static int smu_early_init(void *handle)
smu->adev = adev;
smu->pm_enabled = !!amdgpu_dpm;
+ smu->is_apu = false;
mutex_init(&smu->mutex);
return smu_set_funcs(adev);
@@ -840,6 +820,8 @@ static int smu_sw_init(void *handle)
smu->smu_baco.state = SMU_BACO_STATE_EXIT;
smu->smu_baco.platform_support = false;
+ mutex_init(&smu->sensor_lock);
+
smu->watermarks_bitmap = 0;
smu->power_profile_mode = PP_SMC_POWER_PROFILE_BOOTUP_DEFAULT;
smu->default_power_profile_mode = PP_SMC_POWER_PROFILE_BOOTUP_DEFAULT;
@@ -914,8 +896,7 @@ static int smu_init_fb_allocations(struct smu_context *smu)
struct amdgpu_device *adev = smu->adev;
struct smu_table_context *smu_table = &smu->smu_table;
struct smu_table *tables = smu_table->tables;
- uint32_t i = 0;
- int32_t ret = 0;
+ int ret, i;
for (i = 0; i < SMU_TABLE_COUNT; i++) {
if (tables[i].size == 0)
@@ -933,7 +914,7 @@ static int smu_init_fb_allocations(struct smu_context *smu)
return 0;
failed:
- for (; i > 0; i--) {
+ while (--i >= 0) {
if (tables[i].size == 0)
continue;
amdgpu_bo_free_kernel(&tables[i].bo,
@@ -964,50 +945,6 @@ static int smu_fini_fb_allocations(struct smu_context *smu)
return 0;
}
-static int smu_override_pcie_parameters(struct smu_context *smu)
-{
- struct amdgpu_device *adev = smu->adev;
- uint32_t pcie_gen = 0, pcie_width = 0, smu_pcie_arg;
- int ret;
-
- if (adev->flags & AMD_IS_APU)
- return 0;
-
- if (adev->pm.pcie_gen_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_GEN4)
- pcie_gen = 3;
- else if (adev->pm.pcie_gen_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_GEN3)
- pcie_gen = 2;
- else if (adev->pm.pcie_gen_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_GEN2)
- pcie_gen = 1;
- else if (adev->pm.pcie_gen_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_GEN1)
- pcie_gen = 0;
-
- /* Bit 31:16: LCLK DPM level. 0 is DPM0, and 1 is DPM1
- * Bit 15:8: PCIE GEN, 0 to 3 corresponds to GEN1 to GEN4
- * Bit 7:0: PCIE lane width, 1 to 7 corresponds is x1 to x32
- */
- if (adev->pm.pcie_mlw_mask & CAIL_PCIE_LINK_WIDTH_SUPPORT_X16)
- pcie_width = 6;
- else if (adev->pm.pcie_mlw_mask & CAIL_PCIE_LINK_WIDTH_SUPPORT_X12)
- pcie_width = 5;
- else if (adev->pm.pcie_mlw_mask & CAIL_PCIE_LINK_WIDTH_SUPPORT_X8)
- pcie_width = 4;
- else if (adev->pm.pcie_mlw_mask & CAIL_PCIE_LINK_WIDTH_SUPPORT_X4)
- pcie_width = 3;
- else if (adev->pm.pcie_mlw_mask & CAIL_PCIE_LINK_WIDTH_SUPPORT_X2)
- pcie_width = 2;
- else if (adev->pm.pcie_mlw_mask & CAIL_PCIE_LINK_WIDTH_SUPPORT_X1)
- pcie_width = 1;
-
- smu_pcie_arg = (1 << 16) | (pcie_gen << 8) | pcie_width;
- ret = smu_send_smc_msg_with_param(smu,
- SMU_MSG_OverridePcieParameters,
- smu_pcie_arg);
- if (ret)
- pr_err("[%s] Attempt to override pcie params failed!\n", __func__);
- return ret;
-}
-
static int smu_smc_table_hw_init(struct smu_context *smu,
bool initialize)
{
@@ -1251,6 +1188,7 @@ static int smu_hw_init(void *handle)
if (adev->flags & AMD_IS_APU) {
smu_powergate_sdma(&adev->smu, false);
smu_powergate_vcn(&adev->smu, false);
+ smu_set_gfx_cgpg(&adev->smu, true);
}
if (!smu->pm_enabled)
@@ -1402,7 +1340,7 @@ static int smu_resume(void *handle)
ret = smu_start_smc_engine(smu);
if (ret) {
pr_err("SMU is not ready yet!\n");
- return ret;
+ goto failed;
}
ret = smu_smc_table_hw_init(smu, false);
@@ -1413,6 +1351,9 @@ static int smu_resume(void *handle)
if (ret)
goto failed;
+ if (smu->is_apu)
+ smu_set_gfx_cgpg(&adev->smu, true);
+
mutex_unlock(&smu->mutex);
pr_info("SMU is resumed successfully!\n");
@@ -1557,7 +1498,8 @@ static int smu_enable_umd_pstate(void *handle,
struct smu_context *smu = (struct smu_context*)(handle);
struct smu_dpm_context *smu_dpm_ctx = &(smu->smu_dpm);
- if (!smu->pm_enabled || !smu_dpm_ctx->dpm_context)
+
+ if (!smu->is_apu && (!smu->pm_enabled || !smu_dpm_ctx->dpm_context))
return -EINVAL;
if (!(smu_dpm_ctx->dpm_level & profile_mode_mask)) {
@@ -1755,7 +1697,7 @@ enum amd_dpm_forced_level smu_get_performance_level(struct smu_context *smu)
struct smu_dpm_context *smu_dpm_ctx = &(smu->smu_dpm);
enum amd_dpm_forced_level level;
- if (!smu_dpm_ctx->dpm_context)
+ if (!smu->is_apu && !smu_dpm_ctx->dpm_context)
return -EINVAL;
mutex_lock(&(smu->mutex));
@@ -1770,7 +1712,7 @@ int smu_force_performance_level(struct smu_context *smu, enum amd_dpm_forced_lev
struct smu_dpm_context *smu_dpm_ctx = &(smu->smu_dpm);
int ret = 0;
- if (!smu_dpm_ctx->dpm_context)
+ if (!smu->is_apu && !smu_dpm_ctx->dpm_context)
return -EINVAL;
ret = smu_enable_umd_pstate(smu, &level);
@@ -1852,6 +1794,64 @@ int smu_set_mp1_state(struct smu_context *smu,
return ret;
}
+int smu_set_df_cstate(struct smu_context *smu,
+ enum pp_df_cstate state)
+{
+ int ret = 0;
+
+ /*
+ * The SMC is not fully ready. That may be
+ * expected as the IP may be masked.
+ * So, just return without error.
+ */
+ if (!smu->pm_enabled)
+ return 0;
+
+ if (!smu->ppt_funcs || !smu->ppt_funcs->set_df_cstate)
+ return 0;
+
+ ret = smu->ppt_funcs->set_df_cstate(smu, state);
+ if (ret)
+ pr_err("[SetDfCstate] failed!\n");
+
+ return ret;
+}
+
+int smu_write_watermarks_table(struct smu_context *smu)
+{
+ int ret = 0;
+ struct smu_table_context *smu_table = &smu->smu_table;
+ struct smu_table *table = NULL;
+
+ table = &smu_table->tables[SMU_TABLE_WATERMARKS];
+
+ if (!table->cpu_addr)
+ return -EINVAL;
+
+ ret = smu_update_table(smu, SMU_TABLE_WATERMARKS, 0, table->cpu_addr,
+ true);
+
+ return ret;
+}
+
+int smu_set_watermarks_for_clock_ranges(struct smu_context *smu,
+ struct dm_pp_wm_sets_with_clock_ranges_soc15 *clock_ranges)
+{
+ int ret = 0;
+ struct smu_table *watermarks = &smu->smu_table.tables[SMU_TABLE_WATERMARKS];
+ void *table = watermarks->cpu_addr;
+
+ if (!smu->disable_watermark &&
+ smu_feature_is_enabled(smu, SMU_FEATURE_DPM_DCEFCLK_BIT) &&
+ smu_feature_is_enabled(smu, SMU_FEATURE_DPM_SOCCLK_BIT)) {
+ smu_set_watermarks_table(smu, table, clock_ranges);
+ smu->watermarks_bitmap |= WATERMARKS_EXIST;
+ smu->watermarks_bitmap &= ~WATERMARKS_LOADED;
+ }
+
+ return ret;
+}
+
const struct amd_ip_funcs smu_ip_funcs = {
.name = "smu",
.early_init = smu_early_init,
diff --git a/drivers/gpu/drm/amd/powerplay/arcturus_ppt.c b/drivers/gpu/drm/amd/powerplay/arcturus_ppt.c
index a047a7ec3698..90d871af8e58 100644
--- a/drivers/gpu/drm/amd/powerplay/arcturus_ppt.c
+++ b/drivers/gpu/drm/amd/powerplay/arcturus_ppt.c
@@ -1025,6 +1025,7 @@ static int arcturus_read_sensor(struct smu_context *smu,
if (!data || !size)
return -EINVAL;
+ mutex_lock(&smu->sensor_lock);
switch (sensor) {
case AMDGPU_PP_SENSOR_MAX_FAN_RPM:
*(uint32_t *)data = pptable->FanMaximumRpm;
@@ -1051,6 +1052,7 @@ static int arcturus_read_sensor(struct smu_context *smu,
default:
ret = smu_smc_read_sensor(smu, sensor, data, size);
}
+ mutex_unlock(&smu->sensor_lock);
return ret;
}
@@ -1896,6 +1898,35 @@ static bool arcturus_is_dpm_running(struct smu_context *smu)
return !!(feature_enabled & SMC_DPM_FEATURE);
}
+static int arcturus_dpm_set_uvd_enable(struct smu_context *smu, bool enable)
+{
+ struct smu_power_context *smu_power = &smu->smu_power;
+ struct smu_power_gate *power_gate = &smu_power->power_gate;
+ int ret = 0;
+
+ if (enable) {
+ if (!smu_feature_is_enabled(smu, SMU_FEATURE_VCN_PG_BIT)) {
+ ret = smu_feature_set_enabled(smu, SMU_FEATURE_VCN_PG_BIT, 1);
+ if (ret) {
+ pr_err("[EnableVCNDPM] failed!\n");
+ return ret;
+ }
+ }
+ power_gate->vcn_gated = false;
+ } else {
+ if (smu_feature_is_enabled(smu, SMU_FEATURE_VCN_PG_BIT)) {
+ ret = smu_feature_set_enabled(smu, SMU_FEATURE_VCN_PG_BIT, 0);
+ if (ret) {
+ pr_err("[DisableVCNDPM] failed!\n");
+ return ret;
+ }
+ }
+ power_gate->vcn_gated = true;
+ }
+
+ return ret;
+}
+
static const struct pptable_funcs arcturus_ppt_funcs = {
/* translate smu index into arcturus specific index */
.get_smu_msg_index = arcturus_get_smu_msg_index,
@@ -1934,6 +1965,7 @@ static const struct pptable_funcs arcturus_ppt_funcs = {
.dump_pptable = arcturus_dump_pptable,
.get_power_limit = arcturus_get_power_limit,
.is_dpm_running = arcturus_is_dpm_running,
+ .dpm_set_uvd_enable = arcturus_dpm_set_uvd_enable,
};
void arcturus_set_ppt_funcs(struct smu_context *smu)
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c
index 25e68f245dba..897fd494fe33 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c
@@ -1993,7 +1993,6 @@ static int smu7_sort_lookup_table(struct pp_hwmgr *hwmgr,
struct phm_ppt_v1_voltage_lookup_table *lookup_table)
{
uint32_t table_size, i, j;
- struct phm_ppt_v1_voltage_lookup_record tmp_voltage_lookup_record;
table_size = lookup_table->count;
PP_ASSERT_WITH_CODE(0 != lookup_table->count,
@@ -2004,9 +2003,8 @@ static int smu7_sort_lookup_table(struct pp_hwmgr *hwmgr,
for (j = i + 1; j > 0; j--) {
if (lookup_table->entries[j].us_vdd <
lookup_table->entries[j - 1].us_vdd) {
- tmp_voltage_lookup_record = lookup_table->entries[j - 1];
- lookup_table->entries[j - 1] = lookup_table->entries[j];
- lookup_table->entries[j] = tmp_voltage_lookup_record;
+ swap(lookup_table->entries[j - 1],
+ lookup_table->entries[j]);
}
}
}
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c
index fd41b63a5436..b39f5268b4db 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c
@@ -711,7 +711,6 @@ static int vega10_sort_lookup_table(struct pp_hwmgr *hwmgr,
struct phm_ppt_v1_voltage_lookup_table *lookup_table)
{
uint32_t table_size, i, j;
- struct phm_ppt_v1_voltage_lookup_record tmp_voltage_lookup_record;
PP_ASSERT_WITH_CODE(lookup_table && lookup_table->count,
"Lookup table is empty", return -EINVAL);
@@ -723,9 +722,8 @@ static int vega10_sort_lookup_table(struct pp_hwmgr *hwmgr,
for (j = i + 1; j > 0; j--) {
if (lookup_table->entries[j].us_vdd <
lookup_table->entries[j - 1].us_vdd) {
- tmp_voltage_lookup_record = lookup_table->entries[j - 1];
- lookup_table->entries[j - 1] = lookup_table->entries[j];
- lookup_table->entries[j] = tmp_voltage_lookup_record;
+ swap(lookup_table->entries[j - 1],
+ lookup_table->entries[j]);
}
}
}
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_baco.c b/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_baco.c
index df6ff9252401..9b5e72bdceca 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_baco.c
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_baco.c
@@ -29,7 +29,7 @@
#include "vega20_baco.h"
#include "vega20_smumgr.h"
-
+#include "amdgpu_ras.h"
static const struct soc15_baco_cmd_entry clean_baco_tbl[] =
{
@@ -74,6 +74,7 @@ int vega20_baco_get_state(struct pp_hwmgr *hwmgr, enum BACO_STATE *state)
int vega20_baco_set_state(struct pp_hwmgr *hwmgr, enum BACO_STATE state)
{
struct amdgpu_device *adev = (struct amdgpu_device *)(hwmgr->adev);
+ struct amdgpu_ras *ras = amdgpu_ras_get_context(adev);
enum BACO_STATE cur_state;
uint32_t data;
@@ -84,13 +85,19 @@ int vega20_baco_set_state(struct pp_hwmgr *hwmgr, enum BACO_STATE state)
return 0;
if (state == BACO_STATE_IN) {
- data = RREG32_SOC15(THM, 0, mmTHM_BACO_CNTL);
- data |= 0x80000000;
- WREG32_SOC15(THM, 0, mmTHM_BACO_CNTL, data);
-
-
- if(smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_EnterBaco, 0))
- return -EINVAL;
+ if (!ras || !ras->supported) {
+ data = RREG32_SOC15(THM, 0, mmTHM_BACO_CNTL);
+ data |= 0x80000000;
+ WREG32_SOC15(THM, 0, mmTHM_BACO_CNTL, data);
+
+ if(smum_send_msg_to_smc_with_parameter(hwmgr,
+ PPSMC_MSG_EnterBaco, 0))
+ return -EINVAL;
+ } else {
+ if(smum_send_msg_to_smc_with_parameter(hwmgr,
+ PPSMC_MSG_EnterBaco, 1))
+ return -EINVAL;
+ }
} else if (state == BACO_STATE_OUT) {
if (smum_send_msg_to_smc(hwmgr, PPSMC_MSG_ExitBaco))
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.c
index f5915308e643..3d3c647a63ff 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.c
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.c
@@ -183,6 +183,9 @@ static int vega20_set_features_platform_caps(struct pp_hwmgr *hwmgr)
PHM_PlatformCaps_TablelessHardwareInterface);
phm_cap_set(hwmgr->platform_descriptor.platformCaps,
+ PHM_PlatformCaps_BACO);
+
+ phm_cap_set(hwmgr->platform_descriptor.platformCaps,
PHM_PlatformCaps_EnableSMU7ThermalManagement);
if (adev->pg_flags & AMD_PG_SUPPORT_UVD)
@@ -4155,6 +4158,24 @@ static int vega20_smu_i2c_bus_access(struct pp_hwmgr *hwmgr, bool acquire)
return res;
}
+static int vega20_set_df_cstate(struct pp_hwmgr *hwmgr,
+ enum pp_df_cstate state)
+{
+ int ret;
+
+ /* PPSMC_MSG_DFCstateControl is supported with 40.50 and later fws */
+ if (hwmgr->smu_version < 0x283200) {
+ pr_err("Df cstate control is supported with 40.50 and later SMC fw!\n");
+ return -EINVAL;
+ }
+
+ ret = smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_DFCstateControl, state);
+ if (ret)
+ pr_err("SetDfCstate failed!\n");
+
+ return ret;
+}
+
static const struct pp_hwmgr_func vega20_hwmgr_funcs = {
/* init/fini related */
.backend_init = vega20_hwmgr_backend_init,
@@ -4223,6 +4244,7 @@ static const struct pp_hwmgr_func vega20_hwmgr_funcs = {
.set_asic_baco_state = vega20_baco_set_state,
.set_mp1_state = vega20_set_mp1_state,
.smu_i2c_bus_access = vega20_smu_i2c_bus_access,
+ .set_df_cstate = vega20_set_df_cstate,
};
int vega20_hwmgr_init(struct pp_hwmgr *hwmgr)
diff --git a/drivers/gpu/drm/amd/powerplay/inc/amdgpu_smu.h b/drivers/gpu/drm/amd/powerplay/inc/amdgpu_smu.h
index 5c898444ff97..bf13bf33ba0c 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/amdgpu_smu.h
+++ b/drivers/gpu/drm/amd/powerplay/inc/amdgpu_smu.h
@@ -350,6 +350,7 @@ struct smu_context
const struct smu_funcs *funcs;
const struct pptable_funcs *ppt_funcs;
struct mutex mutex;
+ struct mutex sensor_lock;
uint64_t pool_size;
struct smu_table_context smu_table;
@@ -387,6 +388,7 @@ struct smu_context
uint32_t power_profile_mode;
uint32_t default_power_profile_mode;
bool pm_enabled;
+ bool is_apu;
uint32_t smc_if_version;
@@ -464,7 +466,11 @@ struct pptable_funcs {
int (*display_disable_memory_clock_switch)(struct smu_context *smu, bool disable_memory_clock_switch);
void (*dump_pptable)(struct smu_context *smu);
int (*get_power_limit)(struct smu_context *smu, uint32_t *limit, bool asic_default);
- int (*get_dpm_uclk_limited)(struct smu_context *smu, uint32_t *clock, bool max);
+ int (*get_dpm_clk_limited)(struct smu_context *smu, enum smu_clk_type clk_type,
+ uint32_t dpm_level, uint32_t *freq);
+ int (*set_df_cstate)(struct smu_context *smu, enum pp_df_cstate state);
+ int (*update_pcie_parameters)(struct smu_context *smu, uint32_t pcie_gen_cap, uint32_t pcie_width_cap);
+ int (*get_dpm_clock_table)(struct smu_context *smu, struct dpm_clocks *clock_table);
};
struct smu_funcs
@@ -490,7 +496,6 @@ struct smu_funcs
int (*set_min_dcef_deep_sleep)(struct smu_context *smu);
int (*set_tool_table_location)(struct smu_context *smu);
int (*notify_memory_pool_location)(struct smu_context *smu);
- int (*write_watermarks_table)(struct smu_context *smu);
int (*set_last_dcef_min_deep_sleep_clk)(struct smu_context *smu);
int (*system_features_control)(struct smu_context *smu, bool en);
int (*send_smc_msg)(struct smu_context *smu, uint16_t msg);
@@ -528,8 +533,6 @@ struct smu_funcs
int (*get_current_shallow_sleep_clocks)(struct smu_context *smu,
struct smu_clock_info *clocks);
int (*notify_smu_enable_pwe)(struct smu_context *smu);
- int (*set_watermarks_for_clock_ranges)(struct smu_context *smu,
- struct dm_pp_wm_sets_with_clock_ranges_soc15 *clock_ranges);
int (*conv_power_profile_to_pplib_workload)(int power_profile);
uint32_t (*get_fan_control_mode)(struct smu_context *smu);
int (*set_fan_control_mode)(struct smu_context *smu, uint32_t mode);
@@ -546,6 +549,8 @@ struct smu_funcs
int (*baco_reset)(struct smu_context *smu);
int (*mode2_reset)(struct smu_context *smu);
int (*get_dpm_ultimate_freq)(struct smu_context *smu, enum smu_clk_type clk_type, uint32_t *min, uint32_t *max);
+ int (*set_soft_freq_limited_range)(struct smu_context *smu, enum smu_clk_type clk_type, uint32_t min, uint32_t max);
+ int (*override_pcie_parameters)(struct smu_context *smu);
};
#define smu_init_microcode(smu) \
@@ -592,9 +597,6 @@ struct smu_funcs
((smu)->funcs->notify_memory_pool_location ? (smu)->funcs->notify_memory_pool_location((smu)) : 0)
#define smu_gfx_off_control(smu, enable) \
((smu)->funcs->gfx_off_control ? (smu)->funcs->gfx_off_control((smu), (enable)) : 0)
-
-#define smu_write_watermarks_table(smu) \
- ((smu)->funcs->write_watermarks_table ? (smu)->funcs->write_watermarks_table((smu)) : 0)
#define smu_set_last_dcef_min_deep_sleep_clk(smu) \
((smu)->funcs->set_last_dcef_min_deep_sleep_clk ? (smu)->funcs->set_last_dcef_min_deep_sleep_clk((smu)) : 0)
#define smu_system_features_control(smu, en) \
@@ -734,8 +736,6 @@ struct smu_funcs
((smu)->funcs->get_current_shallow_sleep_clocks ? (smu)->funcs->get_current_shallow_sleep_clocks((smu), (clocks)) : 0)
#define smu_notify_smu_enable_pwe(smu) \
((smu)->funcs->notify_smu_enable_pwe ? (smu)->funcs->notify_smu_enable_pwe((smu)) : 0)
-#define smu_set_watermarks_for_clock_ranges(smu, clock_ranges) \
- ((smu)->funcs->set_watermarks_for_clock_ranges ? (smu)->funcs->set_watermarks_for_clock_ranges((smu), (clock_ranges)) : 0)
#define smu_dpm_set_uvd_enable(smu, enable) \
((smu)->ppt_funcs->dpm_set_uvd_enable ? (smu)->ppt_funcs->dpm_set_uvd_enable((smu), (enable)) : 0)
#define smu_dpm_set_vce_enable(smu, enable) \
@@ -772,9 +772,18 @@ struct smu_funcs
((smu)->ppt_funcs->set_performance_level? (smu)->ppt_funcs->set_performance_level((smu), (level)) : -EINVAL);
#define smu_dump_pptable(smu) \
((smu)->ppt_funcs->dump_pptable ? (smu)->ppt_funcs->dump_pptable((smu)) : 0)
-#define smu_get_dpm_uclk_limited(smu, clock, max) \
- ((smu)->ppt_funcs->get_dpm_uclk_limited ? (smu)->ppt_funcs->get_dpm_uclk_limited((smu), (clock), (max)) : -EINVAL)
+#define smu_get_dpm_clk_limited(smu, clk_type, dpm_level, freq) \
+ ((smu)->ppt_funcs->get_dpm_clk_limited ? (smu)->ppt_funcs->get_dpm_clk_limited((smu), (clk_type), (dpm_level), (freq)) : -EINVAL)
+#define smu_set_soft_freq_limited_range(smu, clk_type, min, max) \
+ ((smu)->funcs->set_soft_freq_limited_range ? (smu)->funcs->set_soft_freq_limited_range((smu), (clk_type), (min), (max)) : -EINVAL)
+#define smu_get_dpm_clock_table(smu, clock_table) \
+ ((smu)->ppt_funcs->get_dpm_clock_table ? (smu)->ppt_funcs->get_dpm_clock_table((smu), (clock_table)) : -EINVAL)
+
+#define smu_override_pcie_parameters(smu) \
+ ((smu)->funcs->override_pcie_parameters ? (smu)->funcs->override_pcie_parameters((smu)) : 0)
+#define smu_update_pcie_parameters(smu, pcie_gen_cap, pcie_width_cap) \
+ ((smu)->ppt_funcs->update_pcie_parameters ? (smu)->ppt_funcs->update_pcie_parameters((smu), (pcie_gen_cap), (pcie_width_cap)) : 0)
extern int smu_get_atom_data_table(struct smu_context *smu, uint32_t table,
uint16_t *size, uint8_t *frev, uint8_t *crev,
@@ -808,6 +817,10 @@ int smu_sys_get_pp_table(struct smu_context *smu, void **table);
int smu_sys_set_pp_table(struct smu_context *smu, void *buf, size_t size);
int smu_get_power_num_states(struct smu_context *smu, struct pp_states_info *state_info);
enum amd_pm_state_type smu_get_current_power_state(struct smu_context *smu);
+int smu_write_watermarks_table(struct smu_context *smu);
+int smu_set_watermarks_for_clock_ranges(
+ struct smu_context *smu,
+ struct dm_pp_wm_sets_with_clock_ranges_soc15 *clock_ranges);
/* smu to display interface */
extern int smu_display_configuration_change(struct smu_context *smu, const
@@ -846,5 +859,7 @@ int smu_force_clk_levels(struct smu_context *smu,
uint32_t mask);
int smu_set_mp1_state(struct smu_context *smu,
enum pp_mp1_state mp1_state);
+int smu_set_df_cstate(struct smu_context *smu,
+ enum pp_df_cstate state);
#endif
diff --git a/drivers/gpu/drm/amd/powerplay/inc/hwmgr.h b/drivers/gpu/drm/amd/powerplay/inc/hwmgr.h
index 7bf9a14bfa0b..bd8c922dfd3e 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/hwmgr.h
+++ b/drivers/gpu/drm/amd/powerplay/inc/hwmgr.h
@@ -355,6 +355,7 @@ struct pp_hwmgr_func {
int (*set_mp1_state)(struct pp_hwmgr *hwmgr, enum pp_mp1_state mp1_state);
int (*asic_reset)(struct pp_hwmgr *hwmgr, enum SMU_ASIC_RESET_MODE mode);
int (*smu_i2c_bus_access)(struct pp_hwmgr *hwmgr, bool aquire);
+ int (*set_df_cstate)(struct pp_hwmgr *hwmgr, enum pp_df_cstate state);
};
struct pp_table_func {
diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu_types.h b/drivers/gpu/drm/amd/powerplay/inc/smu_types.h
index 12a1de55ce3c..d8c9b7f91fcc 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/smu_types.h
+++ b/drivers/gpu/drm/amd/powerplay/inc/smu_types.h
@@ -169,6 +169,7 @@
__SMU_DUMMY_MAP(PowerGateAtHub), \
__SMU_DUMMY_MAP(SetSoftMinJpeg), \
__SMU_DUMMY_MAP(SetHardMinFclkByFreq), \
+ __SMU_DUMMY_MAP(DFCstateControl), \
#undef __SMU_DUMMY_MAP
#define __SMU_DUMMY_MAP(type) SMU_MSG_##type
diff --git a/drivers/gpu/drm/amd/powerplay/inc/vega20_ppsmc.h b/drivers/gpu/drm/amd/powerplay/inc/vega20_ppsmc.h
index a0883038f3c3..0c66f0fe1aaf 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/vega20_ppsmc.h
+++ b/drivers/gpu/drm/amd/powerplay/inc/vega20_ppsmc.h
@@ -120,7 +120,8 @@
#define PPSMC_MSG_SetMGpuFanBoostLimitRpm 0x5D
#define PPSMC_MSG_GetAVFSVoltageByDpm 0x5F
#define PPSMC_MSG_BacoWorkAroundFlushVDCI 0x60
-#define PPSMC_Message_Count 0x61
+#define PPSMC_MSG_DFCstateControl 0x63
+#define PPSMC_Message_Count 0x64
typedef uint32_t PPSMC_Result;
typedef uint32_t PPSMC_Msg;
diff --git a/drivers/gpu/drm/amd/powerplay/navi10_ppt.c b/drivers/gpu/drm/amd/powerplay/navi10_ppt.c
index 088e80b1a757..b88aae9bb242 100644
--- a/drivers/gpu/drm/amd/powerplay/navi10_ppt.c
+++ b/drivers/gpu/drm/amd/powerplay/navi10_ppt.c
@@ -327,11 +327,7 @@ navi10_get_allowed_feature_mask(struct smu_context *smu,
memset(feature_mask, 0, sizeof(uint32_t) * num);
*(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DPM_PREFETCHER_BIT)
- | FEATURE_MASK(FEATURE_DPM_GFXCLK_BIT)
- | FEATURE_MASK(FEATURE_DPM_SOCCLK_BIT)
| FEATURE_MASK(FEATURE_DPM_MP0CLK_BIT)
- | FEATURE_MASK(FEATURE_DPM_LINK_BIT)
- | FEATURE_MASK(FEATURE_GFX_ULV_BIT)
| FEATURE_MASK(FEATURE_RSMU_SMN_CG_BIT)
| FEATURE_MASK(FEATURE_DS_SOCCLK_BIT)
| FEATURE_MASK(FEATURE_PPT_BIT)
@@ -342,8 +338,6 @@ navi10_get_allowed_feature_mask(struct smu_context *smu,
| FEATURE_MASK(FEATURE_FAN_CONTROL_BIT)
| FEATURE_MASK(FEATURE_THERMAL_BIT)
| FEATURE_MASK(FEATURE_LED_DISPLAY_BIT)
- | FEATURE_MASK(FEATURE_DPM_DCEFCLK_BIT)
- | FEATURE_MASK(FEATURE_DS_GFXCLK_BIT)
| FEATURE_MASK(FEATURE_DS_LCLK_BIT)
| FEATURE_MASK(FEATURE_DS_DCEFCLK_BIT)
| FEATURE_MASK(FEATURE_FW_DSTATE_BIT)
@@ -354,11 +348,29 @@ navi10_get_allowed_feature_mask(struct smu_context *smu,
| FEATURE_MASK(FEATURE_FW_CTF_BIT)
| FEATURE_MASK(FEATURE_OUT_OF_BAND_MONITOR_BIT);
+ if (adev->pm.pp_feature & PP_SOCCLK_DPM_MASK)
+ *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DPM_SOCCLK_BIT);
+
+ if (adev->pm.pp_feature & PP_SCLK_DPM_MASK)
+ *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DPM_GFXCLK_BIT);
+
+ if (adev->pm.pp_feature & PP_PCIE_DPM_MASK)
+ *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DPM_LINK_BIT);
+
+ if (adev->pm.pp_feature & PP_DCEFCLK_DPM_MASK)
+ *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DPM_DCEFCLK_BIT);
+
if (adev->pm.pp_feature & PP_MCLK_DPM_MASK)
*(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DPM_UCLK_BIT)
| FEATURE_MASK(FEATURE_MEM_VDDCI_SCALING_BIT)
| FEATURE_MASK(FEATURE_MEM_MVDD_SCALING_BIT);
+ if (adev->pm.pp_feature & PP_ULV_MASK)
+ *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_GFX_ULV_BIT);
+
+ if (adev->pm.pp_feature & PP_SCLK_DEEP_SLEEP_MASK)
+ *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DS_GFXCLK_BIT);
+
if (adev->pm.pp_feature & PP_GFXOFF_MASK)
*(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_GFXOFF_BIT);
@@ -547,7 +559,7 @@ static int navi10_get_metrics_table(struct smu_context *smu,
struct smu_table_context *smu_table= &smu->smu_table;
int ret = 0;
- if (!smu_table->metrics_time || time_after(jiffies, smu_table->metrics_time + HZ / 1000)) {
+ if (!smu_table->metrics_time || time_after(jiffies, smu_table->metrics_time + msecs_to_jiffies(100))) {
ret = smu_update_table(smu, SMU_TABLE_SMU_METRICS, 0,
(void *)smu_table->metrics_table, false);
if (ret) {
@@ -1386,6 +1398,7 @@ static int navi10_read_sensor(struct smu_context *smu,
if(!data || !size)
return -EINVAL;
+ mutex_lock(&smu->sensor_lock);
switch (sensor) {
case AMDGPU_PP_SENSOR_MAX_FAN_RPM:
*(uint32_t *)data = pptable->FanMaximumRpm;
@@ -1409,6 +1422,7 @@ static int navi10_read_sensor(struct smu_context *smu,
default:
ret = smu_smc_read_sensor(smu, sensor, data, size);
}
+ mutex_unlock(&smu->sensor_lock);
return ret;
}
@@ -1614,6 +1628,28 @@ static int navi10_get_power_limit(struct smu_context *smu,
return 0;
}
+static int navi10_update_pcie_parameters(struct smu_context *smu,
+ uint32_t pcie_gen_cap,
+ uint32_t pcie_width_cap)
+{
+ PPTable_t *pptable = smu->smu_table.driver_pptable;
+ int ret, i;
+ uint32_t smu_pcie_arg;
+
+ for (i = 0; i < NUM_LINK_LEVELS; i++) {
+ smu_pcie_arg = (i << 16) |
+ ((pptable->PcieGenSpeed[i] <= pcie_gen_cap) ? (pptable->PcieGenSpeed[i] << 8) :
+ (pcie_gen_cap << 8)) | ((pptable->PcieLaneCount[i] <= pcie_width_cap) ?
+ pptable->PcieLaneCount[i] : pcie_width_cap);
+ ret = smu_send_smc_msg_with_param(smu,
+ SMU_MSG_OverridePcieParameters,
+ smu_pcie_arg);
+ }
+
+ return ret;
+}
+
+
static const struct pptable_funcs navi10_ppt_funcs = {
.tables_init = navi10_tables_init,
.alloc_dpm_context = navi10_allocate_dpm_context,
@@ -1652,6 +1688,7 @@ static const struct pptable_funcs navi10_ppt_funcs = {
.get_thermal_temperature_range = navi10_get_thermal_temperature_range,
.display_disable_memory_clock_switch = navi10_display_disable_memory_clock_switch,
.get_power_limit = navi10_get_power_limit,
+ .update_pcie_parameters = navi10_update_pcie_parameters,
};
void navi10_set_ppt_funcs(struct smu_context *smu)
diff --git a/drivers/gpu/drm/amd/powerplay/renoir_ppt.c b/drivers/gpu/drm/amd/powerplay/renoir_ppt.c
index 9311b6acc34d..fa314c275a82 100644
--- a/drivers/gpu/drm/amd/powerplay/renoir_ppt.c
+++ b/drivers/gpu/drm/amd/powerplay/renoir_ppt.c
@@ -160,21 +160,17 @@ static int renoir_tables_init(struct smu_context *smu, struct smu_table *tables)
* This interface just for getting uclk ultimate freq and should't introduce
* other likewise function result in overmuch callback.
*/
-static int renoir_get_dpm_uclk_limited(struct smu_context *smu, uint32_t *clock, bool max)
+static int renoir_get_dpm_clk_limited(struct smu_context *smu, enum smu_clk_type clk_type,
+ uint32_t dpm_level, uint32_t *freq)
{
+ DpmClocks_t *clk_table = smu->smu_table.clocks_table;
- DpmClocks_t *table = smu->smu_table.clocks_table;
-
- if (!clock || !table)
+ if (!clk_table || clk_type >= SMU_CLK_COUNT)
return -EINVAL;
- if (max)
- *clock = table->FClocks[NUM_FCLK_DPM_LEVELS-1].Freq;
- else
- *clock = table->FClocks[0].Freq;
+ GET_DPM_CUR_FREQ(clk_table, clk_type, dpm_level, *freq);
return 0;
-
}
static int renoir_print_clk_levels(struct smu_context *smu,
@@ -304,19 +300,369 @@ static int renoir_dpm_set_uvd_enable(struct smu_context *smu, bool enable)
return ret;
}
+static int renoir_force_dpm_limit_value(struct smu_context *smu, bool highest)
+{
+ int ret = 0, i = 0;
+ uint32_t min_freq, max_freq, force_freq;
+ enum smu_clk_type clk_type;
+
+ enum smu_clk_type clks[] = {
+ SMU_GFXCLK,
+ SMU_MCLK,
+ SMU_SOCCLK,
+ };
+
+ for (i = 0; i < ARRAY_SIZE(clks); i++) {
+ clk_type = clks[i];
+ ret = smu_get_dpm_freq_range(smu, clk_type, &min_freq, &max_freq);
+ if (ret)
+ return ret;
+
+ force_freq = highest ? max_freq : min_freq;
+ ret = smu_set_soft_freq_range(smu, clk_type, force_freq, force_freq);
+ if (ret)
+ return ret;
+ }
+
+ return ret;
+}
+
+static int renoir_unforce_dpm_levels(struct smu_context *smu) {
+
+ int ret = 0, i = 0;
+ uint32_t min_freq, max_freq;
+ enum smu_clk_type clk_type;
+
+ struct clk_feature_map {
+ enum smu_clk_type clk_type;
+ uint32_t feature;
+ } clk_feature_map[] = {
+ {SMU_GFXCLK, SMU_FEATURE_DPM_GFXCLK_BIT},
+ {SMU_MCLK, SMU_FEATURE_DPM_UCLK_BIT},
+ {SMU_SOCCLK, SMU_FEATURE_DPM_SOCCLK_BIT},
+ };
+
+ for (i = 0; i < ARRAY_SIZE(clk_feature_map); i++) {
+ if (!smu_feature_is_enabled(smu, clk_feature_map[i].feature))
+ continue;
+
+ clk_type = clk_feature_map[i].clk_type;
+
+ ret = smu_get_dpm_freq_range(smu, clk_type, &min_freq, &max_freq);
+ if (ret)
+ return ret;
+
+ ret = smu_set_soft_freq_range(smu, clk_type, min_freq, max_freq);
+ if (ret)
+ return ret;
+ }
+
+ return ret;
+}
+
+static int renoir_get_workload_type(struct smu_context *smu, uint32_t profile)
+{
+
+ uint32_t pplib_workload = 0;
+
+ switch (profile) {
+ case PP_SMC_POWER_PROFILE_FULLSCREEN3D:
+ pplib_workload = WORKLOAD_PPLIB_FULL_SCREEN_3D_BIT;
+ break;
+ case PP_SMC_POWER_PROFILE_CUSTOM:
+ pplib_workload = WORKLOAD_PPLIB_COUNT;
+ break;
+ case PP_SMC_POWER_PROFILE_VIDEO:
+ pplib_workload = WORKLOAD_PPLIB_VIDEO_BIT;
+ break;
+ case PP_SMC_POWER_PROFILE_VR:
+ pplib_workload = WORKLOAD_PPLIB_VR_BIT;
+ break;
+ case PP_SMC_POWER_PROFILE_COMPUTE:
+ pplib_workload = WORKLOAD_PPLIB_COMPUTE_BIT;
+ break;
+ default:
+ return -EINVAL;
+ }
+
+ return pplib_workload;
+}
+
+static int renoir_get_profiling_clk_mask(struct smu_context *smu,
+ enum amd_dpm_forced_level level,
+ uint32_t *sclk_mask,
+ uint32_t *mclk_mask,
+ uint32_t *soc_mask)
+{
+
+ if (level == AMD_DPM_FORCED_LEVEL_PROFILE_MIN_SCLK) {
+ if (sclk_mask)
+ *sclk_mask = 0;
+ } else if (level == AMD_DPM_FORCED_LEVEL_PROFILE_MIN_MCLK) {
+ if (mclk_mask)
+ *mclk_mask = 0;
+ } else if (level == AMD_DPM_FORCED_LEVEL_PROFILE_PEAK) {
+ if(sclk_mask)
+ /* The sclk as gfxclk and has three level about max/min/current */
+ *sclk_mask = 3 - 1;
+
+ if(mclk_mask)
+ *mclk_mask = NUM_MEMCLK_DPM_LEVELS - 1;
+
+ if(soc_mask)
+ *soc_mask = NUM_SOCCLK_DPM_LEVELS - 1;
+ }
+
+ return 0;
+}
+
+/**
+ * This interface get dpm clock table for dc
+ */
+static int renoir_get_dpm_clock_table(struct smu_context *smu, struct dpm_clocks *clock_table)
+{
+ DpmClocks_t *table = smu->smu_table.clocks_table;
+ int i;
+
+ if (!clock_table || !table)
+ return -EINVAL;
+
+ for (i = 0; i < PP_SMU_NUM_DCFCLK_DPM_LEVELS; i++) {
+ clock_table->DcfClocks[i].Freq = table->DcfClocks[i].Freq;
+ clock_table->DcfClocks[i].Vol = table->DcfClocks[i].Vol;
+ }
+
+ for (i = 0; i < PP_SMU_NUM_SOCCLK_DPM_LEVELS; i++) {
+ clock_table->SocClocks[i].Freq = table->SocClocks[i].Freq;
+ clock_table->SocClocks[i].Vol = table->SocClocks[i].Vol;
+ }
+
+ for (i = 0; i < PP_SMU_NUM_FCLK_DPM_LEVELS; i++) {
+ clock_table->FClocks[i].Freq = table->FClocks[i].Freq;
+ clock_table->FClocks[i].Vol = table->FClocks[i].Vol;
+ }
+
+ for (i = 0; i< PP_SMU_NUM_MEMCLK_DPM_LEVELS; i++) {
+ clock_table->MemClocks[i].Freq = table->MemClocks[i].Freq;
+ clock_table->MemClocks[i].Vol = table->MemClocks[i].Vol;
+ }
+
+ return 0;
+}
+
+static int renoir_force_clk_levels(struct smu_context *smu,
+ enum smu_clk_type clk_type, uint32_t mask)
+{
+
+ int ret = 0 ;
+ uint32_t soft_min_level = 0, soft_max_level = 0, min_freq = 0, max_freq = 0;
+ DpmClocks_t *clk_table = smu->smu_table.clocks_table;
+
+ soft_min_level = mask ? (ffs(mask) - 1) : 0;
+ soft_max_level = mask ? (fls(mask) - 1) : 0;
+
+ switch (clk_type) {
+ case SMU_GFXCLK:
+ case SMU_SCLK:
+ if (soft_min_level > 2 || soft_max_level > 2) {
+ pr_info("Currently sclk only support 3 levels on APU\n");
+ return -EINVAL;
+ }
+
+ ret = smu_get_dpm_freq_range(smu, SMU_GFXCLK, &min_freq, &max_freq);
+ if (ret)
+ return ret;
+ ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetSoftMaxGfxClk,
+ soft_max_level == 0 ? min_freq :
+ soft_max_level == 1 ? RENOIR_UMD_PSTATE_GFXCLK : max_freq);
+ if (ret)
+ return ret;
+ ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetHardMinGfxClk,
+ soft_min_level == 2 ? max_freq :
+ soft_min_level == 1 ? RENOIR_UMD_PSTATE_GFXCLK : min_freq);
+ if (ret)
+ return ret;
+ break;
+ case SMU_SOCCLK:
+ GET_DPM_CUR_FREQ(clk_table, clk_type, soft_min_level, min_freq);
+ GET_DPM_CUR_FREQ(clk_table, clk_type, soft_max_level, max_freq);
+ ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetSoftMaxSocclkByFreq, max_freq);
+ if (ret)
+ return ret;
+ ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetHardMinSocclkByFreq, min_freq);
+ if (ret)
+ return ret;
+ break;
+ case SMU_MCLK:
+ case SMU_FCLK:
+ GET_DPM_CUR_FREQ(clk_table, clk_type, soft_min_level, min_freq);
+ GET_DPM_CUR_FREQ(clk_table, clk_type, soft_max_level, max_freq);
+ ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetSoftMaxFclkByFreq, max_freq);
+ if (ret)
+ return ret;
+ ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetHardMinFclkByFreq, min_freq);
+ if (ret)
+ return ret;
+ break;
+ default:
+ break;
+ }
+
+ return ret;
+}
+
+static int renoir_set_power_profile_mode(struct smu_context *smu, long *input, uint32_t size)
+{
+ int workload_type, ret;
+ uint32_t profile_mode = input[size];
+
+ if (profile_mode > PP_SMC_POWER_PROFILE_CUSTOM) {
+ pr_err("Invalid power profile mode %d\n", smu->power_profile_mode);
+ return -EINVAL;
+ }
+
+ /* conv PP_SMC_POWER_PROFILE* to WORKLOAD_PPLIB_*_BIT */
+ workload_type = smu_workload_get_type(smu, smu->power_profile_mode);
+ if (workload_type < 0) {
+ pr_err("Unsupported power profile mode %d on RENOIR\n",smu->power_profile_mode);
+ return -EINVAL;
+ }
+
+ ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetWorkloadMask,
+ 1 << workload_type);
+ if (ret) {
+ pr_err("Fail to set workload type %d\n", workload_type);
+ return ret;
+ }
+
+ smu->power_profile_mode = profile_mode;
+
+ return 0;
+}
+
+static int renoir_set_peak_clock_by_device(struct smu_context *smu)
+{
+ int ret = 0;
+ uint32_t sclk_freq = 0, uclk_freq = 0;
+
+ ret = smu_get_dpm_freq_range(smu, SMU_SCLK, NULL, &sclk_freq);
+ if (ret)
+ return ret;
+
+ ret = smu_set_soft_freq_range(smu, SMU_SCLK, sclk_freq, sclk_freq);
+ if (ret)
+ return ret;
+
+ ret = smu_get_dpm_freq_range(smu, SMU_UCLK, NULL, &uclk_freq);
+ if (ret)
+ return ret;
+
+ ret = smu_set_soft_freq_range(smu, SMU_UCLK, uclk_freq, uclk_freq);
+ if (ret)
+ return ret;
+
+ return ret;
+}
+
+static int renoir_set_performance_level(struct smu_context *smu, enum amd_dpm_forced_level level)
+{
+ int ret = 0;
+
+ switch (level) {
+ case AMD_DPM_FORCED_LEVEL_PROFILE_PEAK:
+ ret = renoir_set_peak_clock_by_device(smu);
+ break;
+ default:
+ ret = -EINVAL;
+ break;
+ }
+
+ return ret;
+}
+
+/* save watermark settings into pplib smu structure,
+ * also pass data to smu controller
+ */
+static int renoir_set_watermarks_table(
+ struct smu_context *smu,
+ void *watermarks,
+ struct dm_pp_wm_sets_with_clock_ranges_soc15 *clock_ranges)
+{
+ int i;
+ int ret = 0;
+ Watermarks_t *table = watermarks;
+
+ if (!table || !clock_ranges)
+ return -EINVAL;
+
+ if (clock_ranges->num_wm_dmif_sets > 4 ||
+ clock_ranges->num_wm_mcif_sets > 4)
+ return -EINVAL;
+
+ /* save into smu->smu_table.tables[SMU_TABLE_WATERMARKS]->cpu_addr*/
+ for (i = 0; i < clock_ranges->num_wm_dmif_sets; i++) {
+ table->WatermarkRow[WM_DCFCLK][i].MinClock =
+ cpu_to_le16((uint16_t)
+ (clock_ranges->wm_dmif_clocks_ranges[i].wm_min_dcfclk_clk_in_khz));
+ table->WatermarkRow[WM_DCFCLK][i].MaxClock =
+ cpu_to_le16((uint16_t)
+ (clock_ranges->wm_dmif_clocks_ranges[i].wm_max_dcfclk_clk_in_khz));
+ table->WatermarkRow[WM_DCFCLK][i].MinMclk =
+ cpu_to_le16((uint16_t)
+ (clock_ranges->wm_dmif_clocks_ranges[i].wm_min_mem_clk_in_khz));
+ table->WatermarkRow[WM_DCFCLK][i].MaxMclk =
+ cpu_to_le16((uint16_t)
+ (clock_ranges->wm_dmif_clocks_ranges[i].wm_max_mem_clk_in_khz));
+ table->WatermarkRow[WM_DCFCLK][i].WmSetting = (uint8_t)
+ clock_ranges->wm_dmif_clocks_ranges[i].wm_set_id;
+ }
+
+ for (i = 0; i < clock_ranges->num_wm_mcif_sets; i++) {
+ table->WatermarkRow[WM_SOCCLK][i].MinClock =
+ cpu_to_le16((uint16_t)
+ (clock_ranges->wm_mcif_clocks_ranges[i].wm_min_socclk_clk_in_khz));
+ table->WatermarkRow[WM_SOCCLK][i].MaxClock =
+ cpu_to_le16((uint16_t)
+ (clock_ranges->wm_mcif_clocks_ranges[i].wm_max_socclk_clk_in_khz));
+ table->WatermarkRow[WM_SOCCLK][i].MinMclk =
+ cpu_to_le16((uint16_t)
+ (clock_ranges->wm_mcif_clocks_ranges[i].wm_min_mem_clk_in_khz));
+ table->WatermarkRow[WM_SOCCLK][i].MaxMclk =
+ cpu_to_le16((uint16_t)
+ (clock_ranges->wm_mcif_clocks_ranges[i].wm_max_mem_clk_in_khz));
+ table->WatermarkRow[WM_SOCCLK][i].WmSetting = (uint8_t)
+ clock_ranges->wm_mcif_clocks_ranges[i].wm_set_id;
+ }
+
+ /* pass data to smu controller */
+ ret = smu_write_watermarks_table(smu);
+
+ return ret;
+}
+
static const struct pptable_funcs renoir_ppt_funcs = {
.get_smu_msg_index = renoir_get_smu_msg_index,
.get_smu_table_index = renoir_get_smu_table_index,
.tables_init = renoir_tables_init,
.set_power_state = NULL,
- .get_dpm_uclk_limited = renoir_get_dpm_uclk_limited,
+ .get_dpm_clk_limited = renoir_get_dpm_clk_limited,
.print_clk_levels = renoir_print_clk_levels,
.get_current_power_state = renoir_get_current_power_state,
.dpm_set_uvd_enable = renoir_dpm_set_uvd_enable,
+ .force_dpm_limit_value = renoir_force_dpm_limit_value,
+ .unforce_dpm_levels = renoir_unforce_dpm_levels,
+ .get_workload_type = renoir_get_workload_type,
+ .get_profiling_clk_mask = renoir_get_profiling_clk_mask,
+ .force_clk_levels = renoir_force_clk_levels,
+ .set_power_profile_mode = renoir_set_power_profile_mode,
+ .set_performance_level = renoir_set_performance_level,
+ .get_dpm_clock_table = renoir_get_dpm_clock_table,
+ .set_watermarks_table = renoir_set_watermarks_table,
};
void renoir_set_ppt_funcs(struct smu_context *smu)
{
smu->ppt_funcs = &renoir_ppt_funcs;
smu->smc_if_version = SMU12_DRIVER_IF_VERSION;
+ smu->is_apu = true;
}
diff --git a/drivers/gpu/drm/amd/powerplay/smu_v11_0.c b/drivers/gpu/drm/amd/powerplay/smu_v11_0.c
index 022b5c8672a1..ac02bcd24da0 100644
--- a/drivers/gpu/drm/amd/powerplay/smu_v11_0.c
+++ b/drivers/gpu/drm/amd/powerplay/smu_v11_0.c
@@ -32,6 +32,7 @@
#include "vega20_ppt.h"
#include "arcturus_ppt.h"
#include "navi10_ppt.h"
+#include "amd_pcie.h"
#include "asic_reg/thm/thm_11_0_2_offset.h"
#include "asic_reg/thm/thm_11_0_2_sh_mask.h"
@@ -770,23 +771,6 @@ static int smu_v11_0_write_pptable(struct smu_context *smu)
return ret;
}
-static int smu_v11_0_write_watermarks_table(struct smu_context *smu)
-{
- int ret = 0;
- struct smu_table_context *smu_table = &smu->smu_table;
- struct smu_table *table = NULL;
-
- table = &smu_table->tables[SMU_TABLE_WATERMARKS];
-
- if (!table->cpu_addr)
- return -EINVAL;
-
- ret = smu_update_table(smu, SMU_TABLE_WATERMARKS, 0, table->cpu_addr,
- true);
-
- return ret;
-}
-
static int smu_v11_0_set_deep_sleep_dcefclk(struct smu_context *smu, uint32_t clk)
{
int ret;
@@ -1336,26 +1320,6 @@ failed:
return ret;
}
-static int
-smu_v11_0_set_watermarks_for_clock_ranges(struct smu_context *smu, struct
- dm_pp_wm_sets_with_clock_ranges_soc15
- *clock_ranges)
-{
- int ret = 0;
- struct smu_table *watermarks = &smu->smu_table.tables[SMU_TABLE_WATERMARKS];
- void *table = watermarks->cpu_addr;
-
- if (!smu->disable_watermark &&
- smu_feature_is_enabled(smu, SMU_FEATURE_DPM_DCEFCLK_BIT) &&
- smu_feature_is_enabled(smu, SMU_FEATURE_DPM_SOCCLK_BIT)) {
- smu_set_watermarks_table(smu, table, clock_ranges);
- smu->watermarks_bitmap |= WATERMARKS_EXIST;
- smu->watermarks_bitmap &= ~WATERMARKS_LOADED;
- }
-
- return ret;
-}
-
static int smu_v11_0_gfx_off_control(struct smu_context *smu, bool enable)
{
int ret = 0;
@@ -1762,6 +1726,77 @@ failed:
return ret;
}
+static int smu_v11_0_set_soft_freq_limited_range(struct smu_context *smu, enum smu_clk_type clk_type,
+ uint32_t min, uint32_t max)
+{
+ int ret = 0, clk_id = 0;
+ uint32_t param;
+
+ clk_id = smu_clk_get_index(smu, clk_type);
+ if (clk_id < 0)
+ return clk_id;
+
+ if (max > 0) {
+ param = (uint32_t)((clk_id << 16) | (max & 0xffff));
+ ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetSoftMaxByFreq,
+ param);
+ if (ret)
+ return ret;
+ }
+
+ if (min > 0) {
+ param = (uint32_t)((clk_id << 16) | (min & 0xffff));
+ ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetSoftMinByFreq,
+ param);
+ if (ret)
+ return ret;
+ }
+
+ return ret;
+}
+
+static int smu_v11_0_override_pcie_parameters(struct smu_context *smu)
+{
+ struct amdgpu_device *adev = smu->adev;
+ uint32_t pcie_gen = 0, pcie_width = 0;
+ int ret;
+
+ if (adev->pm.pcie_gen_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_GEN4)
+ pcie_gen = 3;
+ else if (adev->pm.pcie_gen_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_GEN3)
+ pcie_gen = 2;
+ else if (adev->pm.pcie_gen_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_GEN2)
+ pcie_gen = 1;
+ else if (adev->pm.pcie_gen_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_GEN1)
+ pcie_gen = 0;
+
+ /* Bit 31:16: LCLK DPM level. 0 is DPM0, and 1 is DPM1
+ * Bit 15:8: PCIE GEN, 0 to 3 corresponds to GEN1 to GEN4
+ * Bit 7:0: PCIE lane width, 1 to 7 corresponds is x1 to x32
+ */
+ if (adev->pm.pcie_mlw_mask & CAIL_PCIE_LINK_WIDTH_SUPPORT_X16)
+ pcie_width = 6;
+ else if (adev->pm.pcie_mlw_mask & CAIL_PCIE_LINK_WIDTH_SUPPORT_X12)
+ pcie_width = 5;
+ else if (adev->pm.pcie_mlw_mask & CAIL_PCIE_LINK_WIDTH_SUPPORT_X8)
+ pcie_width = 4;
+ else if (adev->pm.pcie_mlw_mask & CAIL_PCIE_LINK_WIDTH_SUPPORT_X4)
+ pcie_width = 3;
+ else if (adev->pm.pcie_mlw_mask & CAIL_PCIE_LINK_WIDTH_SUPPORT_X2)
+ pcie_width = 2;
+ else if (adev->pm.pcie_mlw_mask & CAIL_PCIE_LINK_WIDTH_SUPPORT_X1)
+ pcie_width = 1;
+
+ ret = smu_update_pcie_parameters(smu, pcie_gen, pcie_width);
+
+ if (ret)
+ pr_err("[%s] Attempt to override pcie params failed!\n", __func__);
+
+ return ret;
+
+}
+
+
static const struct smu_funcs smu_v11_0_funcs = {
.init_microcode = smu_v11_0_init_microcode,
.load_microcode = smu_v11_0_load_microcode,
@@ -1782,7 +1817,6 @@ static const struct smu_funcs smu_v11_0_funcs = {
.parse_pptable = smu_v11_0_parse_pptable,
.populate_smc_tables = smu_v11_0_populate_smc_pptable,
.write_pptable = smu_v11_0_write_pptable,
- .write_watermarks_table = smu_v11_0_write_watermarks_table,
.set_min_dcef_deep_sleep = smu_v11_0_set_min_dcef_deep_sleep,
.set_tool_table_location = smu_v11_0_set_tool_table_location,
.init_display_count = smu_v11_0_init_display_count,
@@ -1798,7 +1832,6 @@ static const struct smu_funcs smu_v11_0_funcs = {
.read_sensor = smu_v11_0_read_sensor,
.set_deep_sleep_dcefclk = smu_v11_0_set_deep_sleep_dcefclk,
.display_clock_voltage_request = smu_v11_0_display_clock_voltage_request,
- .set_watermarks_for_clock_ranges = smu_v11_0_set_watermarks_for_clock_ranges,
.get_fan_control_mode = smu_v11_0_get_fan_control_mode,
.set_fan_control_mode = smu_v11_0_set_fan_control_mode,
.set_fan_speed_percent = smu_v11_0_set_fan_speed_percent,
@@ -1813,6 +1846,8 @@ static const struct smu_funcs smu_v11_0_funcs = {
.baco_set_state = smu_v11_0_baco_set_state,
.baco_reset = smu_v11_0_baco_reset,
.get_dpm_ultimate_freq = smu_v11_0_get_dpm_ultimate_freq,
+ .set_soft_freq_limited_range = smu_v11_0_set_soft_freq_limited_range,
+ .override_pcie_parameters = smu_v11_0_override_pcie_parameters,
};
void smu_v11_0_set_smu_funcs(struct smu_context *smu)
diff --git a/drivers/gpu/drm/amd/powerplay/smu_v12_0.c b/drivers/gpu/drm/amd/powerplay/smu_v12_0.c
index d9d947375557..c9691d0fb523 100644
--- a/drivers/gpu/drm/amd/powerplay/smu_v12_0.c
+++ b/drivers/gpu/drm/amd/powerplay/smu_v12_0.c
@@ -323,10 +323,18 @@ static int smu_v12_0_get_dpm_ultimate_freq(struct smu_context *smu, enum smu_clk
uint32_t *min, uint32_t *max)
{
int ret = 0;
+ uint32_t mclk_mask, soc_mask;
mutex_lock(&smu->mutex);
if (max) {
+ ret = smu_get_profiling_clk_mask(smu, AMD_DPM_FORCED_LEVEL_PROFILE_PEAK,
+ NULL,
+ &mclk_mask,
+ &soc_mask);
+ if (ret)
+ goto failed;
+
switch (clk_type) {
case SMU_GFXCLK:
case SMU_SCLK:
@@ -340,14 +348,20 @@ static int smu_v12_0_get_dpm_ultimate_freq(struct smu_context *smu, enum smu_clk
goto failed;
break;
case SMU_UCLK:
- ret = smu_get_dpm_uclk_limited(smu, max, true);
+ case SMU_FCLK:
+ case SMU_MCLK:
+ ret = smu_get_dpm_clk_limited(smu, clk_type, mclk_mask, max);
+ if (ret)
+ goto failed;
+ break;
+ case SMU_SOCCLK:
+ ret = smu_get_dpm_clk_limited(smu, clk_type, soc_mask, max);
if (ret)
goto failed;
break;
default:
ret = -EINVAL;
goto failed;
-
}
}
@@ -365,7 +379,14 @@ static int smu_v12_0_get_dpm_ultimate_freq(struct smu_context *smu, enum smu_clk
goto failed;
break;
case SMU_UCLK:
- ret = smu_get_dpm_uclk_limited(smu, min, false);
+ case SMU_FCLK:
+ case SMU_MCLK:
+ ret = smu_get_dpm_clk_limited(smu, clk_type, 0, min);
+ if (ret)
+ goto failed;
+ break;
+ case SMU_SOCCLK:
+ ret = smu_get_dpm_clk_limited(smu, clk_type, 0, min);
if (ret)
goto failed;
break;
@@ -373,7 +394,6 @@ static int smu_v12_0_get_dpm_ultimate_freq(struct smu_context *smu, enum smu_clk
ret = -EINVAL;
goto failed;
}
-
}
failed:
mutex_unlock(&smu->mutex);
@@ -383,6 +403,61 @@ failed:
static int smu_v12_0_mode2_reset(struct smu_context *smu){
return smu_v12_0_send_msg_with_param(smu, SMU_MSG_GfxDeviceDriverReset, SMU_RESET_MODE_2);
}
+
+static int smu_v12_0_set_soft_freq_limited_range(struct smu_context *smu, enum smu_clk_type clk_type,
+ uint32_t min, uint32_t max)
+{
+ int ret = 0;
+
+ if (max < min)
+ return -EINVAL;
+
+ switch (clk_type) {
+ case SMU_GFXCLK:
+ case SMU_SCLK:
+ ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetHardMinGfxClk, min);
+ if (ret)
+ return ret;
+
+ ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetSoftMaxGfxClk, max);
+ if (ret)
+ return ret;
+ break;
+ case SMU_FCLK:
+ case SMU_MCLK:
+ ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetHardMinFclkByFreq, min);
+ if (ret)
+ return ret;
+
+ ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetSoftMaxFclkByFreq, max);
+ if (ret)
+ return ret;
+ break;
+ case SMU_SOCCLK:
+ ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetHardMinSocclkByFreq, min);
+ if (ret)
+ return ret;
+
+ ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetSoftMaxSocclkByFreq, max);
+ if (ret)
+ return ret;
+ break;
+ case SMU_VCLK:
+ ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetHardMinVcn, min);
+ if (ret)
+ return ret;
+
+ ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetSoftMaxVcn, max);
+ if (ret)
+ return ret;
+ break;
+ default:
+ return -EINVAL;
+ }
+
+ return ret;
+}
+
static const struct smu_funcs smu_v12_0_funcs = {
.check_fw_status = smu_v12_0_check_fw_status,
.check_fw_version = smu_v12_0_check_fw_version,
@@ -398,6 +473,7 @@ static const struct smu_funcs smu_v12_0_funcs = {
.populate_smc_tables = smu_v12_0_populate_smc_tables,
.get_dpm_ultimate_freq = smu_v12_0_get_dpm_ultimate_freq,
.mode2_reset = smu_v12_0_mode2_reset,
+ .set_soft_freq_limited_range = smu_v12_0_set_soft_freq_limited_range,
};
void smu_v12_0_set_smu_funcs(struct smu_context *smu)
diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/polaris10_smumgr.c b/drivers/gpu/drm/amd/powerplay/smumgr/polaris10_smumgr.c
index a1a9f6196009..2ab589e33b7b 100644
--- a/drivers/gpu/drm/amd/powerplay/smumgr/polaris10_smumgr.c
+++ b/drivers/gpu/drm/amd/powerplay/smumgr/polaris10_smumgr.c
@@ -653,7 +653,7 @@ static int polaris10_populate_smc_mvdd_table(struct pp_hwmgr *hwmgr,
count = SMU_MAX_SMIO_LEVELS;
for (level = 0; level < count; level++) {
table->SmioTable2.Pattern[level].Voltage =
- PP_HOST_TO_SMC_US(data->mvdd_voltage_table.entries[count].value * VOLTAGE_SCALE);
+ PP_HOST_TO_SMC_US(data->mvdd_voltage_table.entries[level].value * VOLTAGE_SCALE);
/* Index into DpmTable.Smio. Drive bits from Smio entry to get this voltage level.*/
table->SmioTable2.Pattern[level].Smio =
(uint8_t) level;
diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/smu8_smumgr.c b/drivers/gpu/drm/amd/powerplay/smumgr/smu8_smumgr.c
index 4728aa23a818..7dca04a89217 100644
--- a/drivers/gpu/drm/amd/powerplay/smumgr/smu8_smumgr.c
+++ b/drivers/gpu/drm/amd/powerplay/smumgr/smu8_smumgr.c
@@ -177,12 +177,10 @@ static int smu8_load_mec_firmware(struct pp_hwmgr *hwmgr)
uint32_t tmp;
int ret = 0;
struct cgs_firmware_info info = {0};
- struct smu8_smumgr *smu8_smu;
if (hwmgr == NULL || hwmgr->device == NULL)
return -EINVAL;
- smu8_smu = hwmgr->smu_backend;
ret = cgs_get_firmware_info(hwmgr->device,
CGS_UCODE_ID_CP_MEC, &info);
diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/vegam_smumgr.c b/drivers/gpu/drm/amd/powerplay/smumgr/vegam_smumgr.c
index 7c960b07746f..ae18fbcb26fb 100644
--- a/drivers/gpu/drm/amd/powerplay/smumgr/vegam_smumgr.c
+++ b/drivers/gpu/drm/amd/powerplay/smumgr/vegam_smumgr.c
@@ -456,7 +456,7 @@ static int vegam_populate_smc_mvdd_table(struct pp_hwmgr *hwmgr,
count = SMU_MAX_SMIO_LEVELS;
for (level = 0; level < count; level++) {
table->SmioTable2.Pattern[level].Voltage = PP_HOST_TO_SMC_US(
- data->mvdd_voltage_table.entries[count].value * VOLTAGE_SCALE);
+ data->mvdd_voltage_table.entries[level].value * VOLTAGE_SCALE);
/* Index into DpmTable.Smio. Drive bits from Smio entry to get this voltage level.*/
table->SmioTable2.Pattern[level].Smio =
(uint8_t) level;
diff --git a/drivers/gpu/drm/amd/powerplay/vega20_ppt.c b/drivers/gpu/drm/amd/powerplay/vega20_ppt.c
index e9ecc3f7aab1..a76ffd58404e 100644
--- a/drivers/gpu/drm/amd/powerplay/vega20_ppt.c
+++ b/drivers/gpu/drm/amd/powerplay/vega20_ppt.c
@@ -143,6 +143,7 @@ static struct smu_11_0_cmn2aisc_mapping vega20_message_map[SMU_MSG_MAX_COUNT] =
MSG_MAP(PrepareMp1ForShutdown),
MSG_MAP(SetMGpuFanBoostLimitRpm),
MSG_MAP(GetAVFSVoltageByDpm),
+ MSG_MAP(DFCstateControl),
};
static struct smu_11_0_cmn2aisc_mapping vega20_clk_map[SMU_CLK_COUNT] = {
@@ -3017,6 +3018,7 @@ static int vega20_read_sensor(struct smu_context *smu,
if(!data || !size)
return -EINVAL;
+ mutex_lock(&smu->sensor_lock);
switch (sensor) {
case AMDGPU_PP_SENSOR_MAX_FAN_RPM:
*(uint32_t *)data = pptable->FanMaximumRpm;
@@ -3042,6 +3044,7 @@ static int vega20_read_sensor(struct smu_context *smu,
default:
ret = smu_smc_read_sensor(smu, sensor, data, size);
}
+ mutex_unlock(&smu->sensor_lock);
return ret;
}
@@ -3133,6 +3136,49 @@ static int vega20_get_thermal_temperature_range(struct smu_context *smu,
return 0;
}
+static int vega20_set_df_cstate(struct smu_context *smu,
+ enum pp_df_cstate state)
+{
+ uint32_t smu_version;
+ int ret;
+
+ ret = smu_get_smc_version(smu, NULL, &smu_version);
+ if (ret) {
+ pr_err("Failed to get smu version!\n");
+ return ret;
+ }
+
+ /* PPSMC_MSG_DFCstateControl is supported with 40.50 and later fws */
+ if (smu_version < 0x283200) {
+ pr_err("Df cstate control is supported with 40.50 and later SMC fw!\n");
+ return -EINVAL;
+ }
+
+ return smu_send_smc_msg_with_param(smu, SMU_MSG_DFCstateControl, state);
+}
+
+static int vega20_update_pcie_parameters(struct smu_context *smu,
+ uint32_t pcie_gen_cap,
+ uint32_t pcie_width_cap)
+{
+ PPTable_t *pptable = smu->smu_table.driver_pptable;
+ int ret, i;
+ uint32_t smu_pcie_arg;
+
+ for (i = 0; i < NUM_LINK_LEVELS; i++) {
+ smu_pcie_arg = (i << 16) |
+ ((pptable->PcieGenSpeed[i] <= pcie_gen_cap) ? (pptable->PcieGenSpeed[i] << 8) :
+ (pcie_gen_cap << 8)) | ((pptable->PcieLaneCount[i] <= pcie_width_cap) ?
+ pptable->PcieLaneCount[i] : pcie_width_cap);
+ ret = smu_send_smc_msg_with_param(smu,
+ SMU_MSG_OverridePcieParameters,
+ smu_pcie_arg);
+ }
+
+ return ret;
+}
+
+
static const struct pptable_funcs vega20_ppt_funcs = {
.tables_init = vega20_tables_init,
.alloc_dpm_context = vega20_allocate_dpm_context,
@@ -3175,7 +3221,9 @@ static const struct pptable_funcs vega20_ppt_funcs = {
.get_fan_speed_percent = vega20_get_fan_speed_percent,
.get_fan_speed_rpm = vega20_get_fan_speed_rpm,
.set_watermarks_table = vega20_set_watermarks_table,
- .get_thermal_temperature_range = vega20_get_thermal_temperature_range
+ .get_thermal_temperature_range = vega20_get_thermal_temperature_range,
+ .set_df_cstate = vega20_set_df_cstate,
+ .update_pcie_parameters = vega20_update_pcie_parameters
};
void vega20_set_ppt_funcs(struct smu_context *smu)
diff --git a/drivers/gpu/drm/drm_mode_object.c b/drivers/gpu/drm/drm_mode_object.c
index a9005c1c2384..b312abb07fe0 100644
--- a/drivers/gpu/drm/drm_mode_object.c
+++ b/drivers/gpu/drm/drm_mode_object.c
@@ -37,6 +37,8 @@ int __drm_mode_object_add(struct drm_device *dev, struct drm_mode_object *obj,
{
int ret;
+ WARN_ON(!dev->driver->load && dev->registered && !obj_free_cb);
+
mutex_lock(&dev->mode_config.idr_mutex);
ret = idr_alloc(&dev->mode_config.object_idr, register_obj ? obj : NULL,
1, 0, GFP_KERNEL);
@@ -97,6 +99,8 @@ void drm_mode_object_register(struct drm_device *dev,
void drm_mode_object_unregister(struct drm_device *dev,
struct drm_mode_object *object)
{
+ WARN_ON(!dev->driver->load && dev->registered && !object->free_cb);
+
mutex_lock(&dev->mode_config.idr_mutex);
if (object->id) {
idr_remove(&dev->mode_config.object_idr, object->id);
diff --git a/include/drm/drm_dp_helper.h b/include/drm/drm_dp_helper.h
index cadf29281861..3369b9f417e3 100644
--- a/include/drm/drm_dp_helper.h
+++ b/include/drm/drm_dp_helper.h
@@ -949,6 +949,32 @@
#define DP_HDCP_2_2_REG_STREAM_TYPE_OFFSET 0x69494
#define DP_HDCP_2_2_REG_DBG_OFFSET 0x69518
+/* Link Training (LT)-tunable PHY Repeaters */
+#define DP_LT_TUNABLE_PHY_REPEATER_FIELD_DATA_STRUCTURE_REV 0xf0000 /* 1.3 */
+#define DP_MAX_LINK_RATE_PHY_REPEATER 0xf0001 /* 1.4a */
+#define DP_PHY_REPEATER_CNT 0xf0002 /* 1.3 */
+#define DP_PHY_REPEATER_MODE 0xf0003 /* 1.3 */
+#define DP_MAX_LANE_COUNT_PHY_REPEATER 0xf0004 /* 1.4a */
+#define DP_Repeater_FEC_CAPABILITY 0xf0004 /* 1.4 */
+#define DP_PHY_REPEATER_EXTENDED_WAIT_TIMEOUT 0xf0005 /* 1.4a */
+#define DP_TRAINING_PATTERN_SET_PHY_REPEATER1 0xf0010 /* 1.3 */
+#define DP_TRAINING_LANE0_SET_PHY_REPEATER1 0xf0011 /* 1.3 */
+#define DP_TRAINING_LANE1_SET_PHY_REPEATER1 0xf0012 /* 1.3 */
+#define DP_TRAINING_LANE2_SET_PHY_REPEATER1 0xf0013 /* 1.3 */
+#define DP_TRAINING_LANE3_SET_PHY_REPEATER1 0xf0014 /* 1.3 */
+#define DP_TRAINING_AUX_RD_INTERVAL_PHY_REPEATER1 0xf0020 /* 1.4a */
+#define DP_TRANSMITTER_CAPABILITY_PHY_REPEATER1 0xf0021 /* 1.4a */
+#define DP_LANE0_1_STATUS_PHY_REPEATER1 0xf0030 /* 1.3 */
+#define DP_LANE2_3_STATUS_PHY_REPEATER1 0xf0031 /* 1.3 */
+#define DP_LANE_ALIGN_STATUS_UPDATED_PHY_REPEATER1 0xf0032 /* 1.3 */
+#define DP_ADJUST_REQUEST_LANE0_1_PHY_REPEATER1 0xf0033 /* 1.3 */
+#define DP_ADJUST_REQUEST_LANE2_3_PHY_REPEATER1 0xf0034 /* 1.3 */
+#define DP_SYMBOL_ERROR_COUNT_LANE0_PHY_REPEATER1 0xf0035 /* 1.3 */
+#define DP_SYMBOL_ERROR_COUNT_LANE1_PHY_REPEATER1 0xf0037 /* 1.3 */
+#define DP_SYMBOL_ERROR_COUNT_LANE2_PHY_REPEATER1 0xf0039 /* 1.3 */
+#define DP_SYMBOL_ERROR_COUNT_LANE3_PHY_REPEATER1 0xf003b /* 1.3 */
+#define DP_FEC_STATUS_PHY_REPEATER1 0xf0290 /* 1.4 */
+
/* DP HDCP message start offsets in DPCD address space */
#define DP_HDCP_2_2_AKE_INIT_OFFSET DP_HDCP_2_2_REG_RTX_OFFSET
#define DP_HDCP_2_2_AKE_SEND_CERT_OFFSET DP_HDCP_2_2_REG_CERT_RX_OFFSET
diff --git a/include/kcl/kcl_device_cgroup.h b/include/kcl/kcl_device_cgroup.h
index 585427ea75c0..6bb6e0ef6809 100644
--- a/include/kcl/kcl_device_cgroup.h
+++ b/include/kcl/kcl_device_cgroup.h
@@ -13,8 +13,8 @@
#define DEVCG_ACC_WRITE 4
#endif
-#if defined(CONFIG_CGROUP_DEVICE) || defined(CONFIG_CGROUP_BPF)
-#ifndef HAVE___DEVCGROUP_CHECK_PERMISSION
+#if defined(CONFIG_CGROUP_DEVICE)
+#ifndef HAVE_DEVCGROUP_CHECK_PERMISSION
extern int (*__kcl_devcgroup_check_permission)(short type, u32 major, u32 minor,
short access);
#endif
@@ -22,7 +22,7 @@ extern int (*__kcl_devcgroup_check_permission)(short type, u32 major, u32 minor,
static inline int kcl_devcgroup_check_permission(short type, u32 major, u32 minor,
short access)
{
-#ifndef HAVE___DEVCGROUP_CHECK_PERMISSION
+#ifndef HAVE_DEVCGROUP_CHECK_PERMISSION
#ifdef BPF_CGROUP_RUN_PROG_DEVICE_CGROUP
int rc = BPF_CGROUP_RUN_PROG_DEVICE_CGROUP(type, major, minor, access);
diff --git a/include/linux/device_cgroup.h b/include/linux/device_cgroup.h
index 8557efe096dc..fa35b52e0002 100644
--- a/include/linux/device_cgroup.h
+++ b/include/linux/device_cgroup.h
@@ -12,26 +12,15 @@
#define DEVCG_DEV_ALL 4 /* this represents all devices */
#ifdef CONFIG_CGROUP_DEVICE
-extern int __devcgroup_check_permission(short type, u32 major, u32 minor,
- short access);
+int devcgroup_check_permission(short type, u32 major, u32 minor,
+ short access);
#else
-static inline int __devcgroup_check_permission(short type, u32 major, u32 minor,
- short access)
+static inline int devcgroup_check_permission(short type, u32 major, u32 minor,
+ short access)
{ return 0; }
#endif
#if defined(CONFIG_CGROUP_DEVICE) || defined(CONFIG_CGROUP_BPF)
-static inline int devcgroup_check_permission(short type, u32 major, u32 minor,
- short access)
-{
- int rc = BPF_CGROUP_RUN_PROG_DEVICE_CGROUP(type, major, minor, access);
-
- if (rc)
- return -EPERM;
-
- return __devcgroup_check_permission(type, major, minor, access);
-}
-
static inline int devcgroup_inode_permission(struct inode *inode, int mask)
{
short type, access = 0;
diff --git a/include/uapi/drm/amdgpu_drm.h b/include/uapi/drm/amdgpu_drm.h
index 04d4b4e4c832..ee4a95fb0c3e 100644
--- a/include/uapi/drm/amdgpu_drm.h
+++ b/include/uapi/drm/amdgpu_drm.h
@@ -148,6 +148,11 @@ extern "C" {
* releasing the memory
*/
#define AMDGPU_GEM_CREATE_VRAM_WIPE_ON_RELEASE (1 << 9)
+/* Flag that BO will be encrypted and that the TMZ bit should be
+ * set in the PTEs when mapping this buffer via GPUVM or
+ * accessing it with various hw blocks
+ */
+#define AMDGPU_GEM_CREATE_ENCRYPTED (1 << 10)
/* hybrid specific */
/* Flag that the memory should be in SPARSE resource */
@@ -223,6 +228,9 @@ union drm_amdgpu_bo_list {
#define AMDGPU_CTX_OP_QUERY_STATE 3
#define AMDGPU_CTX_OP_QUERY_STATE2 4
+/* Flag the command submission as secure */
+#define AMDGPU_CS_FLAGS_SECURE (1 << 0)
+
/* GPU reset status */
#define AMDGPU_CTX_NO_RESET 0
/* this the context caused it */
@@ -616,7 +624,7 @@ struct drm_amdgpu_cs_in {
/** Handle of resource list associated with CS */
__u32 bo_list_handle;
__u32 num_chunks;
- __u32 _pad;
+ __u32 flags;
/** this points to __u64 * which point to cs chunks */
__u64 chunks;
};
@@ -1086,6 +1094,8 @@ struct drm_amdgpu_info_device {
__u64 high_va_max;
/* gfx10 pa_sc_tile_steering_override */
__u32 pa_sc_tile_steering_override;
+ /* disabled TCCs */
+ __u64 tcc_disabled_mask;
};
struct drm_amdgpu_info_hw_ip {
diff --git a/security/device_cgroup.c b/security/device_cgroup.c
index e3a9ad5db5a0..3c57e05bf73b 100644
--- a/security/device_cgroup.c
+++ b/security/device_cgroup.c
@@ -801,8 +801,8 @@ struct cgroup_subsys devices_cgrp_subsys = {
*
* returns 0 on success, -EPERM case the operation is not permitted
*/
-int __devcgroup_check_permission(short type, u32 major, u32 minor,
- short access)
+static int __devcgroup_check_permission(short type, u32 major, u32 minor,
+ short access)
{
struct dev_cgroup *dev_cgroup;
bool rc;
@@ -824,4 +824,14 @@ int __devcgroup_check_permission(short type, u32 major, u32 minor,
return 0;
}
-EXPORT_SYMBOL(__devcgroup_check_permission);
+
+int devcgroup_check_permission(short type, u32 major, u32 minor, short access)
+{
+ int rc = BPF_CGROUP_RUN_PROG_DEVICE_CGROUP(type, major, minor, access);
+
+ if (rc)
+ return -EPERM;
+
+ return __devcgroup_check_permission(type, major, minor, access);
+}
+EXPORT_SYMBOL(devcgroup_check_permission);