Merge tag 'drm-for-v4.15' of git://people.freedesktop.org/~airlied/linux

Pull drm updates from Dave Airlie:
 "This is the main drm pull request for v4.15.

  Core:
   - Atomic object lifetime fixes
   - Atomic iterator improvements
   - Sparse/smatch fixes
   - Legacy kms ioctls to be interruptible
   - EDID override improvements
   - fb/gem helper cleanups
   - Simple outreachy patches
   - Documentation improvements
   - Fix dma-buf rcu races
   - DRM mode object leasing for improving VR use cases.
   - vgaarb improvements for non-x86 platforms.

  New driver:
   - tve200: Faraday Technology TVE200 block.

     This "TV Encoder" encodes a ITU-T BT.656 stream and can be found in
     the StorLink SL3516 (later Cortina Systems CS3516) as well as the
     Grain Media GM8180.

  New bridges:
   - SiI9234 support

  New panels:
   - S6E63J0X03, OTM8009A, Seiko 43WVF1G, 7" rpi touch panel, Toshiba
     LT089AC19000, Innolux AT043TN24

  i915:
   - Remove Coffeelake from alpha support
   - Cannonlake workarounds
   - Infoframe refactoring for DisplayPort
   - VBT updates
   - DisplayPort vswing/emph/buffer translation refactoring
   - CCS fixes
   - Restore GPU clock boost on missed vblanks
   - Scatter list updates for userptr allocations
   - Gen9+ transition watermarks
   - Display IPC (Isochronous Priority Control)
   - Private PAT management
   - GVT: improved error handling and pci config sanitizing
   - Execlist refactoring
   - Transparent Huge Page support
   - User defined priorities support
   - HuC/GuC firmware refactoring
   - DP MST fixes
   - eDP power sequencing fixes
   - Use RCU instead of stop_machine
   - PSR state tracking support
   - Eviction fixes
   - BDW DP aux channel timeout fixes
   - LSPCON fixes
   - Cannonlake PLL fixes

  amdgpu:
   - Per VM BO support
   - Powerplay cleanups
   - CI powerplay support
   - PASID mgr for kfd
   - SR-IOV fixes
   - initial GPU reset for vega10
   - Prime mmap support
   - TTM updates
   - Clock query interface for Raven
   - Fence to handle ioctl
   - UVD encode ring support on Polaris
   - Transparent huge page DMA support
   - Compute LRU pipe tweaks
   - BO flag to allow buffers to opt out of implicit sync
   - CTX priority setting API
   - VRAM lost infrastructure plumbing

  qxl:
   - fix flicker since atomic rework

  amdkfd:
   - Further improvements from internal AMD tree
   - Usermode events
   - Drop radeon support

  nouveau:
   - Pascal temperature sensor support
   - Improved BAR2 handling
   - MMU rework to support Pascal MMU

  exynos:
   - Improved HDMI/mixer support
   - HDMI audio interface support

  tegra:
   - Prep work for tegra186
   - Cleanup/fixes

  msm:
   - Preemption support for a5xx
   - Display fixes for 8x96 (snapdragon 820)
   - Async cursor plane fixes
   - FW loading rework
   - GPU debugging improvements

  vc4:
   - Prep for DSI panels
   - fix T-format tiling scanout
   - New madvise ioctl

  Rockchip:
   - LVDS support

  omapdrm:
   - omap4 HDMI CEC support

  etnaviv:
   - GPU performance counters groundwork

  sun4i:
   - refactor driver load + TCON backend
   - HDMI improvements
   - A31 support
   - Misc fixes

  udl:
   - Probe/EDID read fixes.

  tilcdc:
   - Misc fixes.

  pl111:
   - Support more variants

  adv7511:
   - Improve EDID handling.
   - HDMI CEC support

  sii8620:
   - Add remote control support"

* tag 'drm-for-v4.15' of git://people.freedesktop.org/~airlied/linux: (1480 commits)
  drm/rockchip: analogix_dp: Use mutex rather than spinlock
  drm/mode_object: fix documentation for object lookups.
  drm/i915: Reorder context-close to avoid calling i915_vma_close() under RCU
  drm/i915: Move init_clock_gating() back to where it was
  drm/i915: Prune the reservation shared fence array
  drm/i915: Idle the GPU before shinking everything
  drm/i915: Lock llist_del_first() vs llist_del_all()
  drm/i915: Calculate ironlake intermediate watermarks correctly, v2.
  drm/i915: Disable lazy PPGTT page table optimization for vGPU
  drm/i915/execlists: Remove the priority "optimisation"
  drm/i915: Filter out spurious execlists context-switch interrupts
  drm/amdgpu: use irq-safe lock for kiq->ring_lock
  drm/amdgpu: bypass lru touch for KIQ ring submission
  drm/amdgpu: Potential uninitialized variable in amdgpu_vm_update_directories()
  drm/amdgpu: potential uninitialized variable in amdgpu_vce_ring_parse_cs()
  drm/amd/powerplay: initialize a variable before using it
  drm/amd/powerplay: suppress KASAN out of bounds warning in vega10_populate_all_memory_levels
  drm/amd/amdgpu: fix evicted VRAM bo adjudgement condition
  drm/vblank: Tune drm_crtc_accurate_vblank_count() WARN down to a debug
  drm/rockchip: add CONFIG_OF dependency for lvds
  ...

1 files changed, 13 insertions, 177 deletions

diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/nv44.c b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/nv44.c
index a648c2395545..598c53a27bde 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/nv44.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/nv44.c
@@ -21,176 +21,18 @@
  *
  * Authors: Ben Skeggs
  */
-#include "nv04.h"
+#include "mem.h"
+#include "vmm.h"
 
-#include <core/gpuobj.h>
 #include <core/option.h>
-#include <subdev/timer.h>
 
-#define NV44_GART_SIZE (512 * 1024 * 1024)
-#define NV44_GART_PAGE (  4 * 1024)
-
-/*******************************************************************************
- * VM map/unmap callbacks
- ******************************************************************************/
-
-static void
-nv44_vm_fill(struct nvkm_memory *pgt, dma_addr_t null,
-	     dma_addr_t *list, u32 pte, u32 cnt)
-{
-	u32 base = (pte << 2) & ~0x0000000f;
-	u32 tmp[4];
-
-	tmp[0] = nvkm_ro32(pgt, base + 0x0);
-	tmp[1] = nvkm_ro32(pgt, base + 0x4);
-	tmp[2] = nvkm_ro32(pgt, base + 0x8);
-	tmp[3] = nvkm_ro32(pgt, base + 0xc);
-
-	while (cnt--) {
-		u32 addr = list ? (*list++ >> 12) : (null >> 12);
-		switch (pte++ & 0x3) {
-		case 0:
-			tmp[0] &= ~0x07ffffff;
-			tmp[0] |= addr;
-			break;
-		case 1:
-			tmp[0] &= ~0xf8000000;
-			tmp[0] |= addr << 27;
-			tmp[1] &= ~0x003fffff;
-			tmp[1] |= addr >> 5;
-			break;
-		case 2:
-			tmp[1] &= ~0xffc00000;
-			tmp[1] |= addr << 22;
-			tmp[2] &= ~0x0001ffff;
-			tmp[2] |= addr >> 10;
-			break;
-		case 3:
-			tmp[2] &= ~0xfffe0000;
-			tmp[2] |= addr << 17;
-			tmp[3] &= ~0x00000fff;
-			tmp[3] |= addr >> 15;
-			break;
-		}
-	}
-
-	nvkm_wo32(pgt, base + 0x0, tmp[0]);
-	nvkm_wo32(pgt, base + 0x4, tmp[1]);
-	nvkm_wo32(pgt, base + 0x8, tmp[2]);
-	nvkm_wo32(pgt, base + 0xc, tmp[3] | 0x40000000);
-}
-
-static void
-nv44_vm_map_sg(struct nvkm_vma *vma, struct nvkm_memory *pgt,
-	       struct nvkm_mem *mem, u32 pte, u32 cnt, dma_addr_t *list)
-{
-	struct nv04_mmu *mmu = nv04_mmu(vma->vm->mmu);
-	u32 tmp[4];
-	int i;
-
-	nvkm_kmap(pgt);
-	if (pte & 3) {
-		u32  max = 4 - (pte & 3);
-		u32 part = (cnt > max) ? max : cnt;
-		nv44_vm_fill(pgt, mmu->null, list, pte, part);
-		pte  += part;
-		list += part;
-		cnt  -= part;
-	}
-
-	while (cnt >= 4) {
-		for (i = 0; i < 4; i++)
-			tmp[i] = *list++ >> 12;
-		nvkm_wo32(pgt, pte++ * 4, tmp[0] >>  0 | tmp[1] << 27);
-		nvkm_wo32(pgt, pte++ * 4, tmp[1] >>  5 | tmp[2] << 22);
-		nvkm_wo32(pgt, pte++ * 4, tmp[2] >> 10 | tmp[3] << 17);
-		nvkm_wo32(pgt, pte++ * 4, tmp[3] >> 15 | 0x40000000);
-		cnt -= 4;
-	}
-
-	if (cnt)
-		nv44_vm_fill(pgt, mmu->null, list, pte, cnt);
-	nvkm_done(pgt);
-}
-
-static void
-nv44_vm_unmap(struct nvkm_vma *vma, struct nvkm_memory *pgt, u32 pte, u32 cnt)
-{
-	struct nv04_mmu *mmu = nv04_mmu(vma->vm->mmu);
-
-	nvkm_kmap(pgt);
-	if (pte & 3) {
-		u32  max = 4 - (pte & 3);
-		u32 part = (cnt > max) ? max : cnt;
-		nv44_vm_fill(pgt, mmu->null, NULL, pte, part);
-		pte  += part;
-		cnt  -= part;
-	}
-
-	while (cnt >= 4) {
-		nvkm_wo32(pgt, pte++ * 4, 0x00000000);
-		nvkm_wo32(pgt, pte++ * 4, 0x00000000);
-		nvkm_wo32(pgt, pte++ * 4, 0x00000000);
-		nvkm_wo32(pgt, pte++ * 4, 0x00000000);
-		cnt -= 4;
-	}
-
-	if (cnt)
-		nv44_vm_fill(pgt, mmu->null, NULL, pte, cnt);
-	nvkm_done(pgt);
-}
-
-static void
-nv44_vm_flush(struct nvkm_vm *vm)
-{
-	struct nv04_mmu *mmu = nv04_mmu(vm->mmu);
-	struct nvkm_device *device = mmu->base.subdev.device;
-	nvkm_wr32(device, 0x100814, mmu->base.limit - NV44_GART_PAGE);
-	nvkm_wr32(device, 0x100808, 0x00000020);
-	nvkm_msec(device, 2000,
-		if (nvkm_rd32(device, 0x100808) & 0x00000001)
-			break;
-	);
-	nvkm_wr32(device, 0x100808, 0x00000000);
-}
-
-/*******************************************************************************
- * MMU subdev
- ******************************************************************************/
-
-static int
-nv44_mmu_oneinit(struct nvkm_mmu *base)
-{
-	struct nv04_mmu *mmu = nv04_mmu(base);
-	struct nvkm_device *device = mmu->base.subdev.device;
-	int ret;
-
-	mmu->nullp = dma_alloc_coherent(device->dev, 16 * 1024,
-					&mmu->null, GFP_KERNEL);
-	if (!mmu->nullp) {
-		nvkm_warn(&mmu->base.subdev, "unable to allocate dummy pages\n");
-		mmu->null = 0;
-	}
-
-	ret = nvkm_vm_create(&mmu->base, 0, NV44_GART_SIZE, 0, 4096, NULL,
-			     &mmu->vm);
-	if (ret)
-		return ret;
-
-	ret = nvkm_memory_new(device, NVKM_MEM_TARGET_INST,
-			      (NV44_GART_SIZE / NV44_GART_PAGE) * 4,
-			      512 * 1024, true,
-			      &mmu->vm->pgt[0].mem[0]);
-	mmu->vm->pgt[0].refcount[0] = 1;
-	return ret;
-}
+#include <nvif/class.h>
 
 static void
-nv44_mmu_init(struct nvkm_mmu *base)
+nv44_mmu_init(struct nvkm_mmu *mmu)
 {
-	struct nv04_mmu *mmu = nv04_mmu(base);
-	struct nvkm_device *device = mmu->base.subdev.device;
-	struct nvkm_memory *gart = mmu->vm->pgt[0].mem[0];
+	struct nvkm_device *device = mmu->subdev.device;
+	struct nvkm_memory *pt = mmu->vmm->pd->pt[0]->memory;
 	u32 addr;
 
 	/* calculate vram address of this PRAMIN block, object must be
@@ -198,11 +40,11 @@ nv44_mmu_init(struct nvkm_mmu *base)
 	 * of 512KiB for this to work correctly
 	 */
 	addr  = nvkm_rd32(device, 0x10020c);
-	addr -= ((nvkm_memory_addr(gart) >> 19) + 1) << 19;
+	addr -= ((nvkm_memory_addr(pt) >> 19) + 1) << 19;
 
 	nvkm_wr32(device, 0x100850, 0x80000000);
-	nvkm_wr32(device, 0x100818, mmu->null);
-	nvkm_wr32(device, 0x100804, NV44_GART_SIZE);
+	nvkm_wr32(device, 0x100818, mmu->vmm->null);
+	nvkm_wr32(device, 0x100804, (nvkm_memory_size(pt) / 4) * 4096);
 	nvkm_wr32(device, 0x100850, 0x00008000);
 	nvkm_mask(device, 0x10008c, 0x00000200, 0x00000200);
 	nvkm_wr32(device, 0x100820, 0x00000000);
@@ -212,17 +54,11 @@ nv44_mmu_init(struct nvkm_mmu *base)
 
 static const struct nvkm_mmu_func
 nv44_mmu = {
-	.dtor = nv04_mmu_dtor,
-	.oneinit = nv44_mmu_oneinit,
 	.init = nv44_mmu_init,
-	.limit = NV44_GART_SIZE,
 	.dma_bits = 39,
-	.pgt_bits = 32 - 12,
-	.spg_shift = 12,
-	.lpg_shift = 12,
-	.map_sg = nv44_vm_map_sg,
-	.unmap = nv44_vm_unmap,
-	.flush = nv44_vm_flush,
+	.mmu = {{ -1, -1, NVIF_CLASS_MMU_NV04}},
+	.mem = {{ -1, -1, NVIF_CLASS_MEM_NV04}, nv04_mem_new, nv04_mem_map },
+	.vmm = {{ -1, -1, NVIF_CLASS_VMM_NV04}, nv44_vmm_new, true },
 };
 
 int
@@ -232,5 +68,5 @@ nv44_mmu_new(struct nvkm_device *device, int index, struct nvkm_mmu **pmmu)
 	    !nvkm_boolopt(device->cfgopt, "NvPCIE", true))
 		return nv04_mmu_new(device, index, pmmu);
 
-	return nv04_mmu_new_(&nv44_mmu, device, index, pmmu);
+	return nvkm_mmu_new_(&nv44_mmu, device, index, pmmu);
 }