Merge tag 'drm-for-v4.9' of git://people.freedesktop.org/~airlied/linux

Pull drm updates from Dave Airlie:
 "Core:
   - Fence destaging work
   - DRIVER_LEGACY to split off legacy drm drivers
   - drm_mm refactoring
   - Splitting drm_crtc.c into chunks and documenting better
   - Display info fixes
   - rbtree support for prime buffer lookup
   - Simple VGA DAC driver

  Panel:
   - Add Nexus 7 panel
   - More simple panels

  i915:
   - Refactoring GEM naming
   - Refactored vma/active tracking
   - Lockless request lookups
   - Better stolen memory support
   - FBC fixes
   - SKL watermark fixes
   - VGPU improvements
   - dma-buf fencing support
   - Better DP dongle support

  amdgpu:
   - Powerplay for Iceland asics
   - Improved GPU reset support
   - UVD/VEC powergating support for CZ/ST
   - Preinitialised VRAM buffer support
   - Virtual display support
   - Initial SI support
   - GTT rework
   - PCI shutdown callback support
   - HPD IRQ storm fixes

  amdkfd:
   - bugfixes

  tilcdc:
   - Atomic modesetting support

  mediatek:
   - AAL + GAMMA engine support
   - Hook up gamma LUT
   - Temporal dithering support

  imx:
   - Pixel clock from devicetree
   - drm bridge support for LVDS bridges
   - active plane reconfiguration
   - VDIC deinterlacer support
   - Frame synchronisation unit support
   - Color space conversion support

  analogix:
   - PSR support
   - Better panel on/off support

  rockchip:
   - rk3399 vop/crtc support
   - PSR support

  vc4:
   - Interlaced vblank timing
   - 3D rendering CPU overhead reduction
   - HDMI output fixes

  tda998x:
   - HDMI audio ASoC support

  sunxi:
   - Allwinner A33 support
   - better TCON support

  msm:
   - DT binding cleanups
   - Explicit fence-fd support

  sti:
   - remove sti415/416 support

  etnaviv:
   - MMUv2 refactoring
   - GC3000 support

  exynos:
   - Refactoring HDMI DCC/PHY
   - G2D pm regression fix
   - Page fault issues with wait for vblank

  There is no nouveau work in this tree, as Ben didn't get a pull
  request in, and he was fighting moving to atomic and adding mst
  support, so maybe best it waits for a cycle"

* tag 'drm-for-v4.9' of git://people.freedesktop.org/~airlied/linux: (1412 commits)
  drm/crtc: constify drm_crtc_index parameter
  drm/i915: Fix conflict resolution from backmerge of v4.8-rc8 to drm-next
  drm/i915/guc: Unwind GuC workqueue reservation if request construction fails
  drm/i915: Reset the breadcrumbs IRQ more carefully
  drm/i915: Force relocations via cpu if we run out of idle aperture
  drm/i915: Distinguish last emitted request from last submitted request
  drm/i915: Allow DP to work w/o EDID
  drm/i915: Move long hpd handling into the hotplug work
  drm/i915/execlists: Reinitialise context image after GPU hang
  drm/i915: Use correct index for backtracking HUNG semaphores
  drm/i915: Unalias obj->phys_handle and obj->userptr
  drm/i915: Just clear the mmiodebug before a register access
  drm/i915/gen9: only add the planes actually affected by ddb changes
  drm/i915: Allow PCH DPLL sharing regardless of DPLL_SDVO_HIGH_SPEED
  drm/i915/bxt: Fix HDMI DPLL configuration
  drm/i915/gen9: fix the watermark res_blocks value
  drm/i915/gen9: fix plane_blocks_per_line on watermarks calculations
  drm/i915/gen9: minimum scanlines for Y tile is not always 4
  drm/i915/gen9: fix the WaWmMemoryReadLatency implementation
  drm/i915/kbl: KBL also needs to run the SAGV code
  ...

1 files changed, 233 insertions, 24 deletions

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
index 425413fcaf02..71116da9e782 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
@@ -1645,6 +1645,147 @@ static u32 gfx_v7_0_get_rb_active_bitmap(struct amdgpu_device *adev)
 	return (~data) & mask;
 }
 
+static void
+gfx_v7_0_raster_config(struct amdgpu_device *adev, u32 *rconf, u32 *rconf1)
+{
+	switch (adev->asic_type) {
+	case CHIP_BONAIRE:
+		*rconf |= RB_MAP_PKR0(2) | RB_XSEL2(1) | SE_MAP(2) |
+			  SE_XSEL(1) | SE_YSEL(1);
+		*rconf1 |= 0x0;
+		break;
+	case CHIP_HAWAII:
+		*rconf |= RB_MAP_PKR0(2) | RB_MAP_PKR1(2) |
+			  RB_XSEL2(1) | PKR_MAP(2) | PKR_XSEL(1) |
+			  PKR_YSEL(1) | SE_MAP(2) | SE_XSEL(2) |
+			  SE_YSEL(3);
+		*rconf1 |= SE_PAIR_MAP(2) | SE_PAIR_XSEL(3) |
+			   SE_PAIR_YSEL(2);
+		break;
+	case CHIP_KAVERI:
+		*rconf |= RB_MAP_PKR0(2);
+		*rconf1 |= 0x0;
+		break;
+	case CHIP_KABINI:
+	case CHIP_MULLINS:
+		*rconf |= 0x0;
+		*rconf1 |= 0x0;
+		break;
+	default:
+		DRM_ERROR("unknown asic: 0x%x\n", adev->asic_type);
+		break;
+	}
+}
+
+static void
+gfx_v7_0_write_harvested_raster_configs(struct amdgpu_device *adev,
+					u32 raster_config, u32 raster_config_1,
+					unsigned rb_mask, unsigned num_rb)
+{
+	unsigned sh_per_se = max_t(unsigned, adev->gfx.config.max_sh_per_se, 1);
+	unsigned num_se = max_t(unsigned, adev->gfx.config.max_shader_engines, 1);
+	unsigned rb_per_pkr = min_t(unsigned, num_rb / num_se / sh_per_se, 2);
+	unsigned rb_per_se = num_rb / num_se;
+	unsigned se_mask[4];
+	unsigned se;
+
+	se_mask[0] = ((1 << rb_per_se) - 1) & rb_mask;
+	se_mask[1] = (se_mask[0] << rb_per_se) & rb_mask;
+	se_mask[2] = (se_mask[1] << rb_per_se) & rb_mask;
+	se_mask[3] = (se_mask[2] << rb_per_se) & rb_mask;
+
+	WARN_ON(!(num_se == 1 || num_se == 2 || num_se == 4));
+	WARN_ON(!(sh_per_se == 1 || sh_per_se == 2));
+	WARN_ON(!(rb_per_pkr == 1 || rb_per_pkr == 2));
+
+	if ((num_se > 2) && ((!se_mask[0] && !se_mask[1]) ||
+			     (!se_mask[2] && !se_mask[3]))) {
+		raster_config_1 &= ~SE_PAIR_MAP_MASK;
+
+		if (!se_mask[0] && !se_mask[1]) {
+			raster_config_1 |=
+				SE_PAIR_MAP(RASTER_CONFIG_SE_PAIR_MAP_3);
+		} else {
+			raster_config_1 |=
+				SE_PAIR_MAP(RASTER_CONFIG_SE_PAIR_MAP_0);
+		}
+	}
+
+	for (se = 0; se < num_se; se++) {
+		unsigned raster_config_se = raster_config;
+		unsigned pkr0_mask = ((1 << rb_per_pkr) - 1) << (se * rb_per_se);
+		unsigned pkr1_mask = pkr0_mask << rb_per_pkr;
+		int idx = (se / 2) * 2;
+
+		if ((num_se > 1) && (!se_mask[idx] || !se_mask[idx + 1])) {
+			raster_config_se &= ~SE_MAP_MASK;
+
+			if (!se_mask[idx]) {
+				raster_config_se |= SE_MAP(RASTER_CONFIG_SE_MAP_3);
+			} else {
+				raster_config_se |= SE_MAP(RASTER_CONFIG_SE_MAP_0);
+			}
+		}
+
+		pkr0_mask &= rb_mask;
+		pkr1_mask &= rb_mask;
+		if (rb_per_se > 2 && (!pkr0_mask || !pkr1_mask)) {
+			raster_config_se &= ~PKR_MAP_MASK;
+
+			if (!pkr0_mask) {
+				raster_config_se |= PKR_MAP(RASTER_CONFIG_PKR_MAP_3);
+			} else {
+				raster_config_se |= PKR_MAP(RASTER_CONFIG_PKR_MAP_0);
+			}
+		}
+
+		if (rb_per_se >= 2) {
+			unsigned rb0_mask = 1 << (se * rb_per_se);
+			unsigned rb1_mask = rb0_mask << 1;
+
+			rb0_mask &= rb_mask;
+			rb1_mask &= rb_mask;
+			if (!rb0_mask || !rb1_mask) {
+				raster_config_se &= ~RB_MAP_PKR0_MASK;
+
+				if (!rb0_mask) {
+					raster_config_se |=
+						RB_MAP_PKR0(RASTER_CONFIG_RB_MAP_3);
+				} else {
+					raster_config_se |=
+						RB_MAP_PKR0(RASTER_CONFIG_RB_MAP_0);
+				}
+			}
+
+			if (rb_per_se > 2) {
+				rb0_mask = 1 << (se * rb_per_se + rb_per_pkr);
+				rb1_mask = rb0_mask << 1;
+				rb0_mask &= rb_mask;
+				rb1_mask &= rb_mask;
+				if (!rb0_mask || !rb1_mask) {
+					raster_config_se &= ~RB_MAP_PKR1_MASK;
+
+					if (!rb0_mask) {
+						raster_config_se |=
+							RB_MAP_PKR1(RASTER_CONFIG_RB_MAP_3);
+					} else {
+						raster_config_se |=
+							RB_MAP_PKR1(RASTER_CONFIG_RB_MAP_0);
+					}
+				}
+			}
+		}
+
+		/* GRBM_GFX_INDEX has a different offset on CI+ */
+		gfx_v7_0_select_se_sh(adev, se, 0xffffffff, 0xffffffff);
+		WREG32(mmPA_SC_RASTER_CONFIG, raster_config_se);
+		WREG32(mmPA_SC_RASTER_CONFIG_1, raster_config_1);
+	}
+
+	/* GRBM_GFX_INDEX has a different offset on CI+ */
+	gfx_v7_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+}
+
 /**
  * gfx_v7_0_setup_rb - setup the RBs on the asic
  *
@@ -1658,9 +1799,11 @@ static void gfx_v7_0_setup_rb(struct amdgpu_device *adev)
 {
 	int i, j;
 	u32 data;
+	u32 raster_config = 0, raster_config_1 = 0;
 	u32 active_rbs = 0;
 	u32 rb_bitmap_width_per_sh = adev->gfx.config.max_backends_per_se /
 					adev->gfx.config.max_sh_per_se;
+	unsigned num_rb_pipes;
 
 	mutex_lock(&adev->grbm_idx_mutex);
 	for (i = 0; i < adev->gfx.config.max_shader_engines; i++) {
@@ -1672,10 +1815,25 @@ static void gfx_v7_0_setup_rb(struct amdgpu_device *adev)
 		}
 	}
 	gfx_v7_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
-	mutex_unlock(&adev->grbm_idx_mutex);
 
 	adev->gfx.config.backend_enable_mask = active_rbs;
 	adev->gfx.config.num_rbs = hweight32(active_rbs);
+
+	num_rb_pipes = min_t(unsigned, adev->gfx.config.max_backends_per_se *
+			     adev->gfx.config.max_shader_engines, 16);
+
+	gfx_v7_0_raster_config(adev, &raster_config, &raster_config_1);
+
+	if (!adev->gfx.config.backend_enable_mask ||
+			adev->gfx.config.num_rbs >= num_rb_pipes) {
+		WREG32(mmPA_SC_RASTER_CONFIG, raster_config);
+		WREG32(mmPA_SC_RASTER_CONFIG_1, raster_config_1);
+	} else {
+		gfx_v7_0_write_harvested_raster_configs(adev, raster_config, raster_config_1,
+							adev->gfx.config.backend_enable_mask,
+							num_rb_pipes);
+	}
+	mutex_unlock(&adev->grbm_idx_mutex);
 }
 
 /**
@@ -2096,6 +2254,25 @@ 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)
+{
+	uint32_t dw2 = 0;
+
+	dw2 |= 0x80000000; /* set load_enable otherwise this package is just NOPs */
+	if (flags & AMDGPU_HAVE_CTX_SWITCH) {
+		/* set load_global_config & load_global_uconfig */
+		dw2 |= 0x8001;
+		/* set load_cs_sh_regs */
+		dw2 |= 0x01000000;
+		/* set load_per_context_state & load_gfx_sh_regs */
+		dw2 |= 0x10002;
+	}
+
+	amdgpu_ring_write(ring, PACKET3(PACKET3_CONTEXT_CONTROL, 1));
+	amdgpu_ring_write(ring, dw2);
+	amdgpu_ring_write(ring, 0);
+}
+
 /**
  * gfx_v7_0_ring_test_ib - basic ring IB test
  *
@@ -2443,7 +2620,7 @@ static int gfx_v7_0_cp_gfx_resume(struct amdgpu_device *adev)
 	return 0;
 }
 
-static u32 gfx_v7_0_ring_get_rptr_gfx(struct amdgpu_ring *ring)
+static u32 gfx_v7_0_ring_get_rptr(struct amdgpu_ring *ring)
 {
 	return ring->adev->wb.wb[ring->rptr_offs];
 }
@@ -2463,11 +2640,6 @@ static void gfx_v7_0_ring_set_wptr_gfx(struct amdgpu_ring *ring)
 	(void)RREG32(mmCP_RB0_WPTR);
 }
 
-static u32 gfx_v7_0_ring_get_rptr_compute(struct amdgpu_ring *ring)
-{
-	return ring->adev->wb.wb[ring->rptr_offs];
-}
-
 static u32 gfx_v7_0_ring_get_wptr_compute(struct amdgpu_ring *ring)
 {
 	/* XXX check if swapping is necessary on BE */
@@ -4182,6 +4354,41 @@ static void gfx_v7_0_ring_emit_gds_switch(struct amdgpu_ring *ring,
 	amdgpu_ring_write(ring, (1 << (oa_size + oa_base)) - (1 << oa_base));
 }
 
+static unsigned gfx_v7_0_ring_get_emit_ib_size_gfx(struct amdgpu_ring *ring)
+{
+	return
+		4; /* gfx_v7_0_ring_emit_ib_gfx */
+}
+
+static unsigned gfx_v7_0_ring_get_dma_frame_size_gfx(struct amdgpu_ring *ring)
+{
+	return
+		20 + /* gfx_v7_0_ring_emit_gds_switch */
+		7 + /* gfx_v7_0_ring_emit_hdp_flush */
+		5 + /* gfx_v7_0_ring_emit_hdp_invalidate */
+		12 + 12 + 12 + /* gfx_v7_0_ring_emit_fence_gfx x3 for user fence, vm fence */
+		7 + 4 + /* gfx_v7_0_ring_emit_pipeline_sync */
+		17 + 6 + /* gfx_v7_0_ring_emit_vm_flush */
+		3; /* gfx_v7_ring_emit_cntxcntl */
+}
+
+static unsigned gfx_v7_0_ring_get_emit_ib_size_compute(struct amdgpu_ring *ring)
+{
+	return
+		4; /* gfx_v7_0_ring_emit_ib_compute */
+}
+
+static unsigned gfx_v7_0_ring_get_dma_frame_size_compute(struct amdgpu_ring *ring)
+{
+	return
+		20 + /* gfx_v7_0_ring_emit_gds_switch */
+		7 + /* gfx_v7_0_ring_emit_hdp_flush */
+		5 + /* gfx_v7_0_ring_emit_hdp_invalidate */
+		7 + /* gfx_v7_0_ring_emit_pipeline_sync */
+		17 + /* gfx_v7_0_ring_emit_vm_flush */
+		7 + 7 + 7; /* gfx_v7_0_ring_emit_fence_compute x3 for user fence, vm fence */
+}
+
 static const struct amdgpu_gfx_funcs gfx_v7_0_gfx_funcs = {
 	.get_gpu_clock_counter = &gfx_v7_0_get_gpu_clock_counter,
 	.select_se_sh = &gfx_v7_0_select_se_sh,
@@ -4471,24 +4678,21 @@ static int gfx_v7_0_sw_init(void *handle)
 	}
 
 	/* reserve GDS, GWS and OA resource for gfx */
-	r = amdgpu_bo_create(adev, adev->gds.mem.gfx_partition_size,
-			PAGE_SIZE, true,
-			AMDGPU_GEM_DOMAIN_GDS, 0,
-			NULL, NULL, &adev->gds.gds_gfx_bo);
+	r = amdgpu_bo_create_kernel(adev, adev->gds.mem.gfx_partition_size,
+				    PAGE_SIZE, AMDGPU_GEM_DOMAIN_GDS,
+				    &adev->gds.gds_gfx_bo, NULL, NULL);
 	if (r)
 		return r;
 
-	r = amdgpu_bo_create(adev, adev->gds.gws.gfx_partition_size,
-		PAGE_SIZE, true,
-		AMDGPU_GEM_DOMAIN_GWS, 0,
-		NULL, NULL, &adev->gds.gws_gfx_bo);
+	r = amdgpu_bo_create_kernel(adev, adev->gds.gws.gfx_partition_size,
+				    PAGE_SIZE, AMDGPU_GEM_DOMAIN_GWS,
+				    &adev->gds.gws_gfx_bo, NULL, NULL);
 	if (r)
 		return r;
 
-	r = amdgpu_bo_create(adev, adev->gds.oa.gfx_partition_size,
-			PAGE_SIZE, true,
-			AMDGPU_GEM_DOMAIN_OA, 0,
-			NULL, NULL, &adev->gds.oa_gfx_bo);
+	r = amdgpu_bo_create_kernel(adev, adev->gds.oa.gfx_partition_size,
+				    PAGE_SIZE, AMDGPU_GEM_DOMAIN_OA,
+				    &adev->gds.oa_gfx_bo, NULL, NULL);
 	if (r)
 		return r;
 
@@ -4504,9 +4708,9 @@ static int gfx_v7_0_sw_fini(void *handle)
 	int i;
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
-	amdgpu_bo_unref(&adev->gds.oa_gfx_bo);
-	amdgpu_bo_unref(&adev->gds.gws_gfx_bo);
-	amdgpu_bo_unref(&adev->gds.gds_gfx_bo);
+	amdgpu_bo_free_kernel(&adev->gds.oa_gfx_bo, NULL, NULL);
+	amdgpu_bo_free_kernel(&adev->gds.gws_gfx_bo, NULL, NULL);
+	amdgpu_bo_free_kernel(&adev->gds.gds_gfx_bo, NULL, NULL);
 
 	for (i = 0; i < adev->gfx.num_gfx_rings; i++)
 		amdgpu_ring_fini(&adev->gfx.gfx_ring[i]);
@@ -4937,7 +5141,7 @@ const struct amd_ip_funcs gfx_v7_0_ip_funcs = {
 };
 
 static const struct amdgpu_ring_funcs gfx_v7_0_ring_funcs_gfx = {
-	.get_rptr = gfx_v7_0_ring_get_rptr_gfx,
+	.get_rptr = gfx_v7_0_ring_get_rptr,
 	.get_wptr = gfx_v7_0_ring_get_wptr_gfx,
 	.set_wptr = gfx_v7_0_ring_set_wptr_gfx,
 	.parse_cs = NULL,
@@ -4952,10 +5156,13 @@ static const struct amdgpu_ring_funcs gfx_v7_0_ring_funcs_gfx = {
 	.test_ib = gfx_v7_0_ring_test_ib,
 	.insert_nop = amdgpu_ring_insert_nop,
 	.pad_ib = amdgpu_ring_generic_pad_ib,
+	.emit_cntxcntl = gfx_v7_ring_emit_cntxcntl,
+	.get_emit_ib_size = gfx_v7_0_ring_get_emit_ib_size_gfx,
+	.get_dma_frame_size = gfx_v7_0_ring_get_dma_frame_size_gfx,
 };
 
 static const struct amdgpu_ring_funcs gfx_v7_0_ring_funcs_compute = {
-	.get_rptr = gfx_v7_0_ring_get_rptr_compute,
+	.get_rptr = gfx_v7_0_ring_get_rptr,
 	.get_wptr = gfx_v7_0_ring_get_wptr_compute,
 	.set_wptr = gfx_v7_0_ring_set_wptr_compute,
 	.parse_cs = NULL,
@@ -4970,6 +5177,8 @@ static const struct amdgpu_ring_funcs gfx_v7_0_ring_funcs_compute = {
 	.test_ib = gfx_v7_0_ring_test_ib,
 	.insert_nop = amdgpu_ring_insert_nop,
 	.pad_ib = amdgpu_ring_generic_pad_ib,
+	.get_emit_ib_size = gfx_v7_0_ring_get_emit_ib_size_compute,
+	.get_dma_frame_size = gfx_v7_0_ring_get_dma_frame_size_compute,
 };
 
 static void gfx_v7_0_set_ring_funcs(struct amdgpu_device *adev)