Merge tag 'drm-fixes-for-v4.12-rc4' of git://people.freedesktop.org/~airlied/linux

Pull drm fixes from Dave Airlie:
"This is the main set of fixes for rc4, one amdgpu fix, some exynos
regression fixes, some msm fixes and some i915 and GVT fixes.

I've got a second regression fix for some DP chips that might be a
bit large, but I think we'd like to land it now, I'll send it along
tomorrow, once you are happy with this set"

* tag 'drm-fixes-for-v4.12-rc4' of git://people.freedesktop.org/~airlied/linux: (24 commits)
drm/amdgpu: Program ring for vce instance 1 at its register space
drm/exynos: clean up description of exynos_drm_crtc
drm/exynos: dsi: Remove bridge node reference in removal
drm/exynos: dsi: Fix the parse_dt function
drm/exynos: Merge pre/postclose hooks
drm/msm: Fix the check for the command size
drm/msm: Take the mutex before calling msm_gem_new_impl
drm/msm: for array in-fences, check if all backing fences are from our own context before waiting
drm/msm: constify irq_domain_ops
drm/msm/mdp5: release hwpipe(s) for unused planes
drm/msm: Reuse dma_fence_release.
drm/msm: Expose our reservation object when exporting a dmabuf.
drm/msm/gpu: check legacy clk names in get_clocks()
drm/msm/mdp5: use __drm_atomic_helper_plane_duplicate_state()
drm/msm: select PM_OPP
drm/i915: Stop pretending to mask/unmask LPE audio interrupts
drm/i915/selftests: Silence compiler warning in igt_ctx_exec
Revert "drm/i915: Restore lost "Initialized i915" welcome message"
drm/i915/gvt: clean up unsubmited workloads before destroying kmem cache
drm/i915/gvt: Disable compression workaround for Gen9
...

drivers/gpu/drm/amd/amdgpu/vce_v3_0.c

···
 static uint64_t vce_v3_0_ring_get_rptr(struct amdgpu_ring *ring)
 {
 	struct amdgpu_device *adev = ring->adev;
+	u32 v;
+
+	mutex_lock(&adev->grbm_idx_mutex);
+	if (adev->vce.harvest_config == 0 ||
+		adev->vce.harvest_config == AMDGPU_VCE_HARVEST_VCE1)
+		WREG32(mmGRBM_GFX_INDEX, GET_VCE_INSTANCE(0));
+	else if (adev->vce.harvest_config == AMDGPU_VCE_HARVEST_VCE0)
+		WREG32(mmGRBM_GFX_INDEX, GET_VCE_INSTANCE(1));
 
 	if (ring == &adev->vce.ring[0])
-		return RREG32(mmVCE_RB_RPTR);
+		v = RREG32(mmVCE_RB_RPTR);
 	else if (ring == &adev->vce.ring[1])
-		return RREG32(mmVCE_RB_RPTR2);
+		v = RREG32(mmVCE_RB_RPTR2);
 	else
-		return RREG32(mmVCE_RB_RPTR3);
+		v = RREG32(mmVCE_RB_RPTR3);
+
+	WREG32(mmGRBM_GFX_INDEX, mmGRBM_GFX_INDEX_DEFAULT);
+	mutex_unlock(&adev->grbm_idx_mutex);
+
+	return v;
 }
 
 /**
···
 static uint64_t vce_v3_0_ring_get_wptr(struct amdgpu_ring *ring)
 {
 	struct amdgpu_device *adev = ring->adev;
+	u32 v;
+
+	mutex_lock(&adev->grbm_idx_mutex);
+	if (adev->vce.harvest_config == 0 ||
+		adev->vce.harvest_config == AMDGPU_VCE_HARVEST_VCE1)
+		WREG32(mmGRBM_GFX_INDEX, GET_VCE_INSTANCE(0));
+	else if (adev->vce.harvest_config == AMDGPU_VCE_HARVEST_VCE0)
+		WREG32(mmGRBM_GFX_INDEX, GET_VCE_INSTANCE(1));
 
 	if (ring == &adev->vce.ring[0])
-		return RREG32(mmVCE_RB_WPTR);
+		v = RREG32(mmVCE_RB_WPTR);
 	else if (ring == &adev->vce.ring[1])
-		return RREG32(mmVCE_RB_WPTR2);
+		v = RREG32(mmVCE_RB_WPTR2);
 	else
-		return RREG32(mmVCE_RB_WPTR3);
+		v = RREG32(mmVCE_RB_WPTR3);
+
+	WREG32(mmGRBM_GFX_INDEX, mmGRBM_GFX_INDEX_DEFAULT);
+	mutex_unlock(&adev->grbm_idx_mutex);
+
+	return v;
 }
 
 /**
···
 {
 	struct amdgpu_device *adev = ring->adev;
 
+	mutex_lock(&adev->grbm_idx_mutex);
+	if (adev->vce.harvest_config == 0 ||
+		adev->vce.harvest_config == AMDGPU_VCE_HARVEST_VCE1)
+		WREG32(mmGRBM_GFX_INDEX, GET_VCE_INSTANCE(0));
+	else if (adev->vce.harvest_config == AMDGPU_VCE_HARVEST_VCE0)
+		WREG32(mmGRBM_GFX_INDEX, GET_VCE_INSTANCE(1));
+
 	if (ring == &adev->vce.ring[0])
 		WREG32(mmVCE_RB_WPTR, lower_32_bits(ring->wptr));
 	else if (ring == &adev->vce.ring[1])
 		WREG32(mmVCE_RB_WPTR2, lower_32_bits(ring->wptr));
 	else
 		WREG32(mmVCE_RB_WPTR3, lower_32_bits(ring->wptr));
+
+	WREG32(mmGRBM_GFX_INDEX, mmGRBM_GFX_INDEX_DEFAULT);
+	mutex_unlock(&adev->grbm_idx_mutex);
 }
 
 static void vce_v3_0_override_vce_clock_gating(struct amdgpu_device *adev, bool override)
···
 	struct amdgpu_ring *ring;
 	int idx, r;
 
-	ring = &adev->vce.ring[0];
-	WREG32(mmVCE_RB_RPTR, lower_32_bits(ring->wptr));
-	WREG32(mmVCE_RB_WPTR, lower_32_bits(ring->wptr));
-	WREG32(mmVCE_RB_BASE_LO, ring->gpu_addr);
-	WREG32(mmVCE_RB_BASE_HI, upper_32_bits(ring->gpu_addr));
-	WREG32(mmVCE_RB_SIZE, ring->ring_size / 4);
-
-	ring = &adev->vce.ring[1];
-	WREG32(mmVCE_RB_RPTR2, lower_32_bits(ring->wptr));
-	WREG32(mmVCE_RB_WPTR2, lower_32_bits(ring->wptr));
-	WREG32(mmVCE_RB_BASE_LO2, ring->gpu_addr);
-	WREG32(mmVCE_RB_BASE_HI2, upper_32_bits(ring->gpu_addr));
-	WREG32(mmVCE_RB_SIZE2, ring->ring_size / 4);
-
-	ring = &adev->vce.ring[2];
-	WREG32(mmVCE_RB_RPTR3, lower_32_bits(ring->wptr));
-	WREG32(mmVCE_RB_WPTR3, lower_32_bits(ring->wptr));
-	WREG32(mmVCE_RB_BASE_LO3, ring->gpu_addr);
-	WREG32(mmVCE_RB_BASE_HI3, upper_32_bits(ring->gpu_addr));
-	WREG32(mmVCE_RB_SIZE3, ring->ring_size / 4);
-
 	mutex_lock(&adev->grbm_idx_mutex);
 	for (idx = 0; idx < 2; ++idx) {
 		if (adev->vce.harvest_config & (1 << idx))
 			continue;
 
 		WREG32(mmGRBM_GFX_INDEX, GET_VCE_INSTANCE(idx));
+
+		/* Program instance 0 reg space for two instances or instance 0 case
+		program instance 1 reg space for only instance 1 available case */
+		if (idx != 1 || adev->vce.harvest_config == AMDGPU_VCE_HARVEST_VCE0) {
+			ring = &adev->vce.ring[0];
+			WREG32(mmVCE_RB_RPTR, lower_32_bits(ring->wptr));
+			WREG32(mmVCE_RB_WPTR, lower_32_bits(ring->wptr));
+			WREG32(mmVCE_RB_BASE_LO, ring->gpu_addr);
+			WREG32(mmVCE_RB_BASE_HI, upper_32_bits(ring->gpu_addr));
+			WREG32(mmVCE_RB_SIZE, ring->ring_size / 4);
+
+			ring = &adev->vce.ring[1];
+			WREG32(mmVCE_RB_RPTR2, lower_32_bits(ring->wptr));
+			WREG32(mmVCE_RB_WPTR2, lower_32_bits(ring->wptr));
+			WREG32(mmVCE_RB_BASE_LO2, ring->gpu_addr);
+			WREG32(mmVCE_RB_BASE_HI2, upper_32_bits(ring->gpu_addr));
+			WREG32(mmVCE_RB_SIZE2, ring->ring_size / 4);
+
+			ring = &adev->vce.ring[2];
+			WREG32(mmVCE_RB_RPTR3, lower_32_bits(ring->wptr));
+			WREG32(mmVCE_RB_WPTR3, lower_32_bits(ring->wptr));
+			WREG32(mmVCE_RB_BASE_LO3, ring->gpu_addr);
+			WREG32(mmVCE_RB_BASE_HI3, upper_32_bits(ring->gpu_addr));
+			WREG32(mmVCE_RB_SIZE3, ring->ring_size / 4);
+		}
+
 		vce_v3_0_mc_resume(adev, idx);
 		WREG32_FIELD(VCE_STATUS, JOB_BUSY, 1);
 

drivers/gpu/drm/exynos/exynos_drm_drv.c

···
 	return ret;
 }
 
-static void exynos_drm_preclose(struct drm_device *dev,
-					struct drm_file *file)
-{
-	exynos_drm_subdrv_close(dev, file);
-}
-
 static void exynos_drm_postclose(struct drm_device *dev, struct drm_file *file)
 {
+	exynos_drm_subdrv_close(dev, file);
 	kfree(file->driver_priv);
 	file->driver_priv = NULL;
 }
···
 	.driver_features	= DRIVER_MODESET | DRIVER_GEM | DRIVER_PRIME
 				  | DRIVER_ATOMIC | DRIVER_RENDER,
 	.open			= exynos_drm_open,
-	.preclose		= exynos_drm_preclose,
 	.lastclose		= exynos_drm_lastclose,
 	.postclose		= exynos_drm_postclose,
 	.gem_free_object_unlocked = exynos_drm_gem_free_object,

drivers/gpu/drm/exynos/exynos_drm_drv.h

···
  *	drm framework doesn't support multiple irq yet.
  *	we can refer to the crtc to current hardware interrupt occurred through
  *	this pipe value.
- * @enabled: if the crtc is enabled or not
- * @event: vblank event that is currently queued for flip
- * @wait_update: wait all pending planes updates to finish
- * @pending_update: number of pending plane updates in this crtc
  * @ops: pointer to callbacks for exynos drm specific functionality
  * @ctx: A pointer to the crtc's implementation specific context
+ * @pipe_clk: A pointer to the crtc's pipeline clock.
  */
 struct exynos_drm_crtc {
 	struct drm_crtc			base;

drivers/gpu/drm/exynos/exynos_drm_dsi.c

···
 {
 	struct device *dev = dsi->dev;
 	struct device_node *node = dev->of_node;
-	struct device_node *ep;
 	int ret;
 
 	ret = exynos_dsi_of_read_u32(node, "samsung,pll-clock-frequency",
···
 	if (ret < 0)
 		return ret;
 
-	ep = of_graph_get_endpoint_by_regs(node, DSI_PORT_OUT, 0);
-	if (!ep) {
-		dev_err(dev, "no output port with endpoint specified\n");
-		return -EINVAL;
-	}
-
-	ret = exynos_dsi_of_read_u32(ep, "samsung,burst-clock-frequency",
+	ret = exynos_dsi_of_read_u32(node, "samsung,burst-clock-frequency",
 				     &dsi->burst_clk_rate);
 	if (ret < 0)
-		goto end;
+		return ret;
 
-	ret = exynos_dsi_of_read_u32(ep, "samsung,esc-clock-frequency",
+	ret = exynos_dsi_of_read_u32(node, "samsung,esc-clock-frequency",
 				     &dsi->esc_clk_rate);
 	if (ret < 0)
-		goto end;
-
-	of_node_put(ep);
+		return ret;
 
 	dsi->bridge_node = of_graph_get_remote_node(node, DSI_PORT_OUT, 0);
 	if (!dsi->bridge_node)
 		return -EINVAL;
 
-end:
-	of_node_put(ep);
-
-	return ret;
+	return 0;
 }
 
 static int exynos_dsi_bind(struct device *dev, struct device *master,
···
 
 static int exynos_dsi_remove(struct platform_device *pdev)
 {
+	struct exynos_dsi *dsi = platform_get_drvdata(pdev);
+
+	of_node_put(dsi->bridge_node);
+
 	pm_runtime_disable(&pdev->dev);
 
 	component_del(&pdev->dev, &exynos_dsi_component_ops);

drivers/gpu/drm/i915/gvt/execlist.c

···
 	vgpu_vreg(vgpu, ctx_status_ptr_reg) = ctx_status_ptr.dw;
 }
 
+static void clean_workloads(struct intel_vgpu *vgpu, unsigned long engine_mask)
+{
+	struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv;
+	struct intel_engine_cs *engine;
+	struct intel_vgpu_workload *pos, *n;
+	unsigned int tmp;
+
+	/* free the unsubmited workloads in the queues. */
+	for_each_engine_masked(engine, dev_priv, engine_mask, tmp) {
+		list_for_each_entry_safe(pos, n,
+			&vgpu->workload_q_head[engine->id], list) {
+			list_del_init(&pos->list);
+			free_workload(pos);
+		}
+	}
+}
+
 void intel_vgpu_clean_execlist(struct intel_vgpu *vgpu)
 {
+	clean_workloads(vgpu, ALL_ENGINES);
 	kmem_cache_destroy(vgpu->workloads);
 }
 
···
 {
 	struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv;
 	struct intel_engine_cs *engine;
-	struct intel_vgpu_workload *pos, *n;
 	unsigned int tmp;
 
-	for_each_engine_masked(engine, dev_priv, engine_mask, tmp) {
-		/* free the unsubmited workload in the queue */
-		list_for_each_entry_safe(pos, n,
-			&vgpu->workload_q_head[engine->id], list) {
-			list_del_init(&pos->list);
-			free_workload(pos);
-		}
-
+	clean_workloads(vgpu, engine_mask);
+	for_each_engine_masked(engine, dev_priv, engine_mask, tmp)
 		init_vgpu_execlist(vgpu, engine->id);
-	}
 }

drivers/gpu/drm/i915/gvt/handlers.c

···
 		void *p_data, unsigned int bytes)
 {
 	struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv;
-	i915_reg_t reg = {.reg = offset};
+	u32 v = *(u32 *)p_data;
+
+	if (!IS_SKYLAKE(dev_priv) && !IS_KABYLAKE(dev_priv))
+		return intel_vgpu_default_mmio_write(vgpu,
+				offset, p_data, bytes);
 
 	switch (offset) {
 	case 0x4ddc:
-		vgpu_vreg(vgpu, offset) = 0x8000003c;
-		/* WaCompressedResourceSamplerPbeMediaNewHashMode:skl */
-		I915_WRITE(reg, vgpu_vreg(vgpu, offset));
+		/* bypass WaCompressedResourceSamplerPbeMediaNewHashMode */
+		vgpu_vreg(vgpu, offset) = v & ~(1 << 31);
 		break;
 	case 0x42080:
-		vgpu_vreg(vgpu, offset) = 0x8000;
-		/* WaCompressedResourceDisplayNewHashMode:skl */
-		I915_WRITE(reg, vgpu_vreg(vgpu, offset));
+		/* bypass WaCompressedResourceDisplayNewHashMode */
+		vgpu_vreg(vgpu, offset) = v & ~(1 << 15);
+		break;
+	case 0xe194:
+		/* bypass WaCompressedResourceSamplerPbeMediaNewHashMode */
+		vgpu_vreg(vgpu, offset) = v & ~(1 << 8);
+		break;
+	case 0x7014:
+		/* bypass WaCompressedResourceSamplerPbeMediaNewHashMode */
+		vgpu_vreg(vgpu, offset) = v & ~(1 << 13);
 		break;
 	default:
 		return -EINVAL;
···
 	MMIO_DFH(GAM_ECOCHK, D_ALL, F_CMD_ACCESS, NULL, NULL);
 	MMIO_DFH(GEN7_COMMON_SLICE_CHICKEN1, D_ALL, F_MODE_MASK | F_CMD_ACCESS,
 		NULL, NULL);
-	MMIO_DFH(COMMON_SLICE_CHICKEN2, D_ALL, F_MODE_MASK | F_CMD_ACCESS, NULL, NULL);
+	MMIO_DFH(COMMON_SLICE_CHICKEN2, D_ALL, F_MODE_MASK | F_CMD_ACCESS, NULL,
+		 skl_misc_ctl_write);
 	MMIO_DFH(0x9030, D_ALL, F_CMD_ACCESS, NULL, NULL);
 	MMIO_DFH(0x20a0, D_ALL, F_CMD_ACCESS, NULL, NULL);
 	MMIO_DFH(0x2420, D_ALL, F_CMD_ACCESS, NULL, NULL);
···
 	MMIO_D(0x6e570, D_BDW_PLUS);
 	MMIO_D(0x65f10, D_BDW_PLUS);
 
-	MMIO_DFH(0xe194, D_BDW_PLUS, F_MODE_MASK | F_CMD_ACCESS, NULL, NULL);
+	MMIO_DFH(0xe194, D_BDW_PLUS, F_MODE_MASK | F_CMD_ACCESS, NULL,
+		 skl_misc_ctl_write);
 	MMIO_DFH(0xe188, D_BDW_PLUS, F_MODE_MASK | F_CMD_ACCESS, NULL, NULL);
 	MMIO_DFH(HALF_SLICE_CHICKEN2, D_BDW_PLUS, F_MODE_MASK | F_CMD_ACCESS, NULL, NULL);
 	MMIO_DFH(0x2580, D_BDW_PLUS, F_MODE_MASK | F_CMD_ACCESS, NULL, NULL);

drivers/gpu/drm/i915/i915_drv.c

···
 
 	dev_priv->ipc_enabled = false;
 
-	/* Everything is in place, we can now relax! */
-	DRM_INFO("Initialized %s %d.%d.%d %s for %s on minor %d\n",
-		 driver.name, driver.major, driver.minor, driver.patchlevel,
-		 driver.date, pci_name(pdev), dev_priv->drm.primary->index);
 	if (IS_ENABLED(CONFIG_DRM_I915_DEBUG))
 		DRM_INFO("DRM_I915_DEBUG enabled\n");
 	if (IS_ENABLED(CONFIG_DRM_I915_DEBUG_GEM))

drivers/gpu/drm/i915/i915_gem_gtt.c

···
 		    appgtt->base.allocate_va_range) {
 			ret = appgtt->base.allocate_va_range(&appgtt->base,
 							     vma->node.start,
-							     vma->node.size);
+							     vma->size);
 			if (ret)
 				goto err_pages;
 		}

drivers/gpu/drm/i915/i915_gem_shrinker.c

···
 		return;
 
 	mutex_unlock(&dev->struct_mutex);
-
-	/* expedite the RCU grace period to free some request slabs */
-	synchronize_rcu_expedited();
 }
 
 static bool any_vma_pinned(struct drm_i915_gem_object *obj)
···
 				I915_SHRINK_UNBOUND |
 				I915_SHRINK_ACTIVE);
 	intel_runtime_pm_put(dev_priv);
-
-	synchronize_rcu(); /* wait for our earlier RCU delayed slab frees */
 
 	return freed;
 }

drivers/gpu/drm/i915/i915_irq.c

···
 	u32 pipestat_mask;
 	u32 enable_mask;
 	enum pipe pipe;
-	u32 val;
 
 	pipestat_mask = PLANE_FLIP_DONE_INT_STATUS_VLV |
 			PIPE_CRC_DONE_INTERRUPT_STATUS;
···
 
 	enable_mask = I915_DISPLAY_PORT_INTERRUPT |
 		I915_DISPLAY_PIPE_A_EVENT_INTERRUPT |
-		I915_DISPLAY_PIPE_B_EVENT_INTERRUPT;
+		I915_DISPLAY_PIPE_B_EVENT_INTERRUPT |
+		I915_LPE_PIPE_A_INTERRUPT |
+		I915_LPE_PIPE_B_INTERRUPT;
+
 	if (IS_CHERRYVIEW(dev_priv))
-		enable_mask |= I915_DISPLAY_PIPE_C_EVENT_INTERRUPT;
+		enable_mask |= I915_DISPLAY_PIPE_C_EVENT_INTERRUPT |
+			I915_LPE_PIPE_C_INTERRUPT;
 
 	WARN_ON(dev_priv->irq_mask != ~0);
-
-	val = (I915_LPE_PIPE_A_INTERRUPT |
-		I915_LPE_PIPE_B_INTERRUPT |
-		I915_LPE_PIPE_C_INTERRUPT);
-
-	enable_mask |= val;
 
 	dev_priv->irq_mask = ~enable_mask;
 

drivers/gpu/drm/i915/i915_reg.h

···
 
 /* MIPI DSI registers */
 
-#define _MIPI_PORT(port, a, c)	((port) ? c : a)	/* ports A and C only */
+#define _MIPI_PORT(port, a, c)	(((port) == PORT_A) ? a : c)	/* ports A and C only */
 #define _MMIO_MIPI(port, a, c)	_MMIO(_MIPI_PORT(port, a, c))
 
 #define MIPIO_TXESC_CLK_DIV1			_MMIO(0x160004)

drivers/gpu/drm/i915/intel_lpe_audio.c

···
 
 static void lpe_audio_irq_unmask(struct irq_data *d)
 {
-	struct drm_i915_private *dev_priv = d->chip_data;
-	unsigned long irqflags;
-	u32 val = (I915_LPE_PIPE_A_INTERRUPT |
-		I915_LPE_PIPE_B_INTERRUPT);
-
-	if (IS_CHERRYVIEW(dev_priv))
-		val |= I915_LPE_PIPE_C_INTERRUPT;
-
-	spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
-
-	dev_priv->irq_mask &= ~val;
-	I915_WRITE(VLV_IIR, val);
-	I915_WRITE(VLV_IIR, val);
-	I915_WRITE(VLV_IMR, dev_priv->irq_mask);
-	POSTING_READ(VLV_IMR);
-
-	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
 }
 
 static void lpe_audio_irq_mask(struct irq_data *d)
 {
-	struct drm_i915_private *dev_priv = d->chip_data;
-	unsigned long irqflags;
-	u32 val = (I915_LPE_PIPE_A_INTERRUPT |
-		I915_LPE_PIPE_B_INTERRUPT);
-
-	if (IS_CHERRYVIEW(dev_priv))
-		val |= I915_LPE_PIPE_C_INTERRUPT;
-
-	spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
-
-	dev_priv->irq_mask |= val;
-	I915_WRITE(VLV_IMR, dev_priv->irq_mask);
-	I915_WRITE(VLV_IIR, val);
-	I915_WRITE(VLV_IIR, val);
-	POSTING_READ(VLV_IIR);
-
-	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
 }
 
 static struct irq_chip lpe_audio_irqchip = {
···
 		return;
 
 	desc = irq_to_desc(dev_priv->lpe_audio.irq);
-
-	lpe_audio_irq_mask(&desc->irq_data);
 
 	lpe_audio_platdev_destroy(dev_priv);
 

drivers/gpu/drm/i915/intel_lrc.c

···
 
 	ce->ring = ring;
 	ce->state = vma;
-	ce->initialised = engine->init_context == NULL;
+	ce->initialised |= engine->init_context == NULL;
 
 	return 0;
 

drivers/gpu/drm/i915/selftests/i915_gem_context.c

···
 static int igt_ctx_exec(void *arg)
 {
 	struct drm_i915_private *i915 = arg;
-	struct drm_i915_gem_object *obj;
+	struct drm_i915_gem_object *obj = NULL;
 	struct drm_file *file;
 	IGT_TIMEOUT(end_time);
 	LIST_HEAD(objects);
···
 		}
 
 		for_each_engine(engine, i915, id) {
-			if (dw == 0) {
+			if (!obj) {
 				obj = create_test_object(ctx, file, &objects);
 				if (IS_ERR(obj)) {
 					err = PTR_ERR(obj);
···
 				goto out_unlock;
 			}
 
-			if (++dw == max_dwords(obj))
+			if (++dw == max_dwords(obj)) {
+				obj = NULL;
 				dw = 0;
+			}
 			ndwords++;
 		}
 		ncontexts++;

drivers/gpu/drm/msm/Kconfig

···
 	select QCOM_SCM
 	select SND_SOC_HDMI_CODEC if SND_SOC
 	select SYNC_FILE
+	select PM_OPP
 	default y
 	help
 	  DRM/KMS driver for MSM/snapdragon.

drivers/gpu/drm/msm/mdp/mdp5/mdp5_mdss.c

···
 	return 0;
 }
 
-static struct irq_domain_ops mdss_hw_irqdomain_ops = {
+static const struct irq_domain_ops mdss_hw_irqdomain_ops = {
 	.map = mdss_hw_irqdomain_map,
 	.xlate = irq_domain_xlate_onecell,
 };

drivers/gpu/drm/msm/mdp/mdp5/mdp5_plane.c

···
 
 	mdp5_state = kmemdup(to_mdp5_plane_state(plane->state),
 			sizeof(*mdp5_state), GFP_KERNEL);
+	if (!mdp5_state)
+		return NULL;
 
-	if (mdp5_state && mdp5_state->base.fb)
-		drm_framebuffer_reference(mdp5_state->base.fb);
+	__drm_atomic_helper_plane_duplicate_state(plane, &mdp5_state->base);
 
 	return &mdp5_state->base;
 }
···
 			mdp5_pipe_release(state->state, old_hwpipe);
 			mdp5_pipe_release(state->state, old_right_hwpipe);
 		}
+	} else {
+		mdp5_pipe_release(state->state, mdp5_state->hwpipe);
+		mdp5_pipe_release(state->state, mdp5_state->r_hwpipe);
+		mdp5_state->hwpipe = mdp5_state->r_hwpipe = NULL;
 	}
 
 	return 0;

drivers/gpu/drm/msm/msm_drv.c

···
 	.prime_fd_to_handle = drm_gem_prime_fd_to_handle,
 	.gem_prime_export   = drm_gem_prime_export,
 	.gem_prime_import   = drm_gem_prime_import,
+	.gem_prime_res_obj  = msm_gem_prime_res_obj,
 	.gem_prime_pin      = msm_gem_prime_pin,
 	.gem_prime_unpin    = msm_gem_prime_unpin,
 	.gem_prime_get_sg_table = msm_gem_prime_get_sg_table,

drivers/gpu/drm/msm/msm_drv.h

···
 void *msm_gem_prime_vmap(struct drm_gem_object *obj);
 void msm_gem_prime_vunmap(struct drm_gem_object *obj, void *vaddr);
 int msm_gem_prime_mmap(struct drm_gem_object *obj, struct vm_area_struct *vma);
+struct reservation_object *msm_gem_prime_res_obj(struct drm_gem_object *obj);
 struct drm_gem_object *msm_gem_prime_import_sg_table(struct drm_device *dev,
 		struct dma_buf_attachment *attach, struct sg_table *sg);
 int msm_gem_prime_pin(struct drm_gem_object *obj);

drivers/gpu/drm/msm/msm_fence.c

···
 }
 
 struct msm_fence {
-	struct msm_fence_context *fctx;
 	struct dma_fence base;
+	struct msm_fence_context *fctx;
 };
 
 static inline struct msm_fence *to_msm_fence(struct dma_fence *fence)
···
 	return fence_completed(f->fctx, f->base.seqno);
 }
 
-static void msm_fence_release(struct dma_fence *fence)
-{
-	struct msm_fence *f = to_msm_fence(fence);
-	kfree_rcu(f, base.rcu);
-}
-
 static const struct dma_fence_ops msm_fence_ops = {
 	.get_driver_name = msm_fence_get_driver_name,
 	.get_timeline_name = msm_fence_get_timeline_name,
 	.enable_signaling = msm_fence_enable_signaling,
 	.signaled = msm_fence_signaled,
 	.wait = dma_fence_default_wait,
-	.release = msm_fence_release,
+	.release = dma_fence_free,
 };
 
 struct dma_fence *

drivers/gpu/drm/msm/msm_gem.c

···
 	struct msm_gem_object *msm_obj;
 	bool use_vram = false;
 
+	WARN_ON(!mutex_is_locked(&dev->struct_mutex));
+
 	switch (flags & MSM_BO_CACHE_MASK) {
 	case MSM_BO_UNCACHED:
 	case MSM_BO_CACHED:
···
 
 	size = PAGE_ALIGN(dmabuf->size);
 
+	/* Take mutex so we can modify the inactive list in msm_gem_new_impl */
+	mutex_lock(&dev->struct_mutex);
 	ret = msm_gem_new_impl(dev, size, MSM_BO_WC, dmabuf->resv, &obj);
+	mutex_unlock(&dev->struct_mutex);
+
 	if (ret)
 		goto fail;
 

drivers/gpu/drm/msm/msm_gem_prime.c

···
 	if (!obj->import_attach)
 		msm_gem_put_pages(obj);
 }
+
+struct reservation_object *msm_gem_prime_res_obj(struct drm_gem_object *obj)
+{
+	struct msm_gem_object *msm_obj = to_msm_bo(obj);
+
+	return msm_obj->resv;
+}

drivers/gpu/drm/msm/msm_gem_submit.c

···
 		if (!in_fence)
 			return -EINVAL;
 
-		/* TODO if we get an array-fence due to userspace merging multiple
-		 * fences, we need a way to determine if all the backing fences
-		 * are from our own context..
+		/*
+		 * Wait if the fence is from a foreign context, or if the fence
+		 * array contains any fence from a foreign context.
 		 */
-
-		if (in_fence->context != gpu->fctx->context) {
+		if (!dma_fence_match_context(in_fence, gpu->fctx->context)) {
 			ret = dma_fence_wait(in_fence, true);
 			if (ret)
 				return ret;
···
 			goto out;
 		}
 
-		if ((submit_cmd.size + submit_cmd.submit_offset) >=
-				msm_obj->base.size) {
+		if (!submit_cmd.size ||
+			((submit_cmd.size + submit_cmd.submit_offset) >
+				msm_obj->base.size)) {
 			DRM_ERROR("invalid cmdstream size: %u\n", submit_cmd.size);
 			ret = -EINVAL;
 			goto out;

drivers/gpu/drm/msm/msm_gpu.c

···
 		gpu->grp_clks[i] = get_clock(dev, name);
 
 		/* Remember the key clocks that we need to control later */
-		if (!strcmp(name, "core"))
+		if (!strcmp(name, "core") || !strcmp(name, "core_clk"))
 			gpu->core_clk = gpu->grp_clks[i];
-		else if (!strcmp(name, "rbbmtimer"))
+		else if (!strcmp(name, "rbbmtimer") || !strcmp(name, "rbbmtimer_clk"))
 			gpu->rbbmtimer_clk = gpu->grp_clks[i];
 
 		++i;