Merge tag 'v4.14.69' into release/qcomlt-4.14

This is the 4.14.69 stable release

48 files changed, 453 insertions, 204 deletions

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
index 4d08957d2108f..f08624f2f2094 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
@@ -683,8 +683,12 @@ int amdgpu_bo_pin_restricted(struct amdgpu_bo *bo, u32 domain,
 		return -EINVAL;
 
 	/* A shared bo cannot be migrated to VRAM */
-	if (bo->prime_shared_count && (domain == AMDGPU_GEM_DOMAIN_VRAM))
-		return -EINVAL;
+	if (bo->prime_shared_count) {
+		if (domain & AMDGPU_GEM_DOMAIN_GTT)
+			domain = AMDGPU_GEM_DOMAIN_GTT;
+		else
+			return -EINVAL;
+	}
 
 	if (bo->pin_count) {
 		uint32_t mem_type = bo->tbo.mem.mem_type;
@@ -747,8 +751,7 @@ int amdgpu_bo_pin_restricted(struct amdgpu_bo *bo, u32 domain,
 	}
 	if (domain == AMDGPU_GEM_DOMAIN_VRAM) {
 		adev->vram_pin_size += amdgpu_bo_size(bo);
-		if (bo->flags & AMDGPU_GEM_CREATE_NO_CPU_ACCESS)
-			adev->invisible_pin_size += amdgpu_bo_size(bo);
+		adev->invisible_pin_size += amdgpu_vram_mgr_bo_invisible_size(bo);
 	} else if (domain == AMDGPU_GEM_DOMAIN_GTT) {
 		adev->gart_pin_size += amdgpu_bo_size(bo);
 	}
@@ -786,8 +789,7 @@ int amdgpu_bo_unpin(struct amdgpu_bo *bo)
 
 	if (bo->tbo.mem.mem_type == TTM_PL_VRAM) {
 		adev->vram_pin_size -= amdgpu_bo_size(bo);
-		if (bo->flags & AMDGPU_GEM_CREATE_NO_CPU_ACCESS)
-			adev->invisible_pin_size -= amdgpu_bo_size(bo);
+		adev->invisible_pin_size -= amdgpu_vram_mgr_bo_invisible_size(bo);
 	} else if (bo->tbo.mem.mem_type == TTM_PL_TT) {
 		adev->gart_pin_size -= amdgpu_bo_size(bo);
 	}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h
index 43093bffa2cfa..557829a847784 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h
@@ -64,6 +64,7 @@ extern const struct ttm_mem_type_manager_func amdgpu_vram_mgr_func;
 bool amdgpu_gtt_mgr_is_allocated(struct ttm_mem_reg *mem);
 uint64_t amdgpu_gtt_mgr_usage(struct ttm_mem_type_manager *man);
 
+u64 amdgpu_vram_mgr_bo_invisible_size(struct amdgpu_bo *bo);
 uint64_t amdgpu_vram_mgr_usage(struct ttm_mem_type_manager *man);
 uint64_t amdgpu_vram_mgr_vis_usage(struct ttm_mem_type_manager *man);
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
index 041e0121590c9..308a9755eae39 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
@@ -85,6 +85,7 @@ int amdgpu_vcn_sw_init(struct amdgpu_device *adev)
 	}
 
 	hdr = (const struct common_firmware_header *)adev->vcn.fw->data;
+	adev->vcn.fw_version = le32_to_cpu(hdr->ucode_version);
 	family_id = le32_to_cpu(hdr->ucode_version) & 0xff;
 	version_major = (le32_to_cpu(hdr->ucode_version) >> 24) & 0xff;
 	version_minor = (le32_to_cpu(hdr->ucode_version) >> 8) & 0xff;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
index 26e9006279717..86d8a961518ed 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
@@ -102,6 +102,22 @@ static u64 amdgpu_vram_mgr_vis_size(struct amdgpu_device *adev,
 }
 
 /**
+ * amdgpu_vram_mgr_bo_invisible_size - CPU invisible BO size
+ *
+ * @bo: &amdgpu_bo buffer object (must be in VRAM)
+ *
+ * Returns:
+ * How much of the given &amdgpu_bo buffer object lies in CPU invisible VRAM.
+ */
+u64 amdgpu_vram_mgr_bo_invisible_size(struct amdgpu_bo *bo)
+{
+	if (bo->flags & AMDGPU_GEM_CREATE_NO_CPU_ACCESS)
+		return amdgpu_bo_size(bo);
+
+	return 0;
+}
+
+/**
  * amdgpu_vram_mgr_new - allocate new ranges
  *
  * @man: TTM memory type manager
@@ -140,7 +156,8 @@ static int amdgpu_vram_mgr_new(struct ttm_mem_type_manager *man,
 		num_nodes = DIV_ROUND_UP(mem->num_pages, pages_per_node);
 	}
 
-	nodes = kcalloc(num_nodes, sizeof(*nodes), GFP_KERNEL);
+	nodes = kvmalloc_array(num_nodes, sizeof(*nodes),
+			       GFP_KERNEL | __GFP_ZERO);
 	if (!nodes)
 		return -ENOMEM;
 
@@ -195,7 +212,7 @@ error:
 		drm_mm_remove_node(&nodes[i]);
 	spin_unlock(&mgr->lock);
 
-	kfree(nodes);
+	kvfree(nodes);
 	return r == -ENOSPC ? 0 : r;
 }
 
@@ -234,7 +251,7 @@ static void amdgpu_vram_mgr_del(struct ttm_mem_type_manager *man,
 	atomic64_sub(usage, &mgr->usage);
 	atomic64_sub(vis_usage, &mgr->vis_usage);
 
-	kfree(mem->mm_node);
+	kvfree(mem->mm_node);
 	mem->mm_node = NULL;
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c b/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c
index cf81065e3c5ab..242dfb1433d26 100644
--- a/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c
@@ -467,8 +467,8 @@ static int vce_v3_0_hw_init(void *handle)
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
 	vce_v3_0_override_vce_clock_gating(adev, true);
-	if (!(adev->flags & AMD_IS_APU))
-		amdgpu_asic_set_vce_clocks(adev, 10000, 10000);
+
+	amdgpu_asic_set_vce_clocks(adev, 10000, 10000);
 
 	for (i = 0; i < adev->vce.num_rings; i++)
 		adev->vce.ring[i].ready = false;
@@ -899,7 +899,7 @@ static const struct amdgpu_ring_funcs vce_v3_0_ring_phys_funcs = {
 	.emit_frame_size =
 		4 + /* vce_v3_0_emit_pipeline_sync */
 		6, /* amdgpu_vce_ring_emit_fence x1 no user fence */
-	.emit_ib_size = 5, /* vce_v3_0_ring_emit_ib */
+	.emit_ib_size = 4, /* amdgpu_vce_ring_emit_ib */
 	.emit_ib = amdgpu_vce_ring_emit_ib,
 	.emit_fence = amdgpu_vce_ring_emit_fence,
 	.test_ring = amdgpu_vce_ring_test_ring,
@@ -923,7 +923,7 @@ static const struct amdgpu_ring_funcs vce_v3_0_ring_vm_funcs = {
 		6 + /* vce_v3_0_emit_vm_flush */
 		4 + /* vce_v3_0_emit_pipeline_sync */
 		6 + 6, /* amdgpu_vce_ring_emit_fence x2 vm fence */
-	.emit_ib_size = 4, /* amdgpu_vce_ring_emit_ib */
+	.emit_ib_size = 5, /* vce_v3_0_ring_emit_ib */
 	.emit_ib = vce_v3_0_ring_emit_ib,
 	.emit_vm_flush = vce_v3_0_emit_vm_flush,
 	.emit_pipeline_sync = vce_v3_0_emit_pipeline_sync,
diff --git a/drivers/gpu/drm/amd/amdgpu/vi.c b/drivers/gpu/drm/amd/amdgpu/vi.c
index 4968b6bb9466d..0327e0a6802b7 100644
--- a/drivers/gpu/drm/amd/amdgpu/vi.c
+++ b/drivers/gpu/drm/amd/amdgpu/vi.c
@@ -729,33 +729,59 @@ static int vi_set_uvd_clock(struct amdgpu_device *adev, u32 clock,
 		return r;
 
 	tmp = RREG32_SMC(cntl_reg);
-	tmp &= ~(CG_DCLK_CNTL__DCLK_DIR_CNTL_EN_MASK |
-		CG_DCLK_CNTL__DCLK_DIVIDER_MASK);
+
+	if (adev->flags & AMD_IS_APU)
+		tmp &= ~CG_DCLK_CNTL__DCLK_DIVIDER_MASK;
+	else
+		tmp &= ~(CG_DCLK_CNTL__DCLK_DIR_CNTL_EN_MASK |
+				CG_DCLK_CNTL__DCLK_DIVIDER_MASK);
 	tmp |= dividers.post_divider;
 	WREG32_SMC(cntl_reg, tmp);
 
 	for (i = 0; i < 100; i++) {
-		if (RREG32_SMC(status_reg) & CG_DCLK_STATUS__DCLK_STATUS_MASK)
-			break;
+		tmp = RREG32_SMC(status_reg);
+		if (adev->flags & AMD_IS_APU) {
+			if (tmp & 0x10000)
+				break;
+		} else {
+			if (tmp & CG_DCLK_STATUS__DCLK_STATUS_MASK)
+				break;
+		}
 		mdelay(10);
 	}
 	if (i == 100)
 		return -ETIMEDOUT;
-
 	return 0;
 }
 
+#define ixGNB_CLK1_DFS_CNTL 0xD82200F0
+#define ixGNB_CLK1_STATUS   0xD822010C
+#define ixGNB_CLK2_DFS_CNTL 0xD8220110
+#define ixGNB_CLK2_STATUS   0xD822012C
+#define ixGNB_CLK3_DFS_CNTL 0xD8220130
+#define ixGNB_CLK3_STATUS   0xD822014C
+
 static int vi_set_uvd_clocks(struct amdgpu_device *adev, u32 vclk, u32 dclk)
 {
 	int r;
 
-	r = vi_set_uvd_clock(adev, vclk, ixCG_VCLK_CNTL, ixCG_VCLK_STATUS);
-	if (r)
-		return r;
+	if (adev->flags & AMD_IS_APU) {
+		r = vi_set_uvd_clock(adev, vclk, ixGNB_CLK2_DFS_CNTL, ixGNB_CLK2_STATUS);
+		if (r)
+			return r;
 
-	r = vi_set_uvd_clock(adev, dclk, ixCG_DCLK_CNTL, ixCG_DCLK_STATUS);
-	if (r)
-		return r;
+		r = vi_set_uvd_clock(adev, dclk, ixGNB_CLK1_DFS_CNTL, ixGNB_CLK1_STATUS);
+		if (r)
+			return r;
+	} else {
+		r = vi_set_uvd_clock(adev, vclk, ixCG_VCLK_CNTL, ixCG_VCLK_STATUS);
+		if (r)
+			return r;
+
+		r = vi_set_uvd_clock(adev, dclk, ixCG_DCLK_CNTL, ixCG_DCLK_STATUS);
+		if (r)
+			return r;
+	}
 
 	return 0;
 }
@@ -765,6 +791,22 @@ static int vi_set_vce_clocks(struct amdgpu_device *adev, u32 evclk, u32 ecclk)
 	int r, i;
 	struct atom_clock_dividers dividers;
 	u32 tmp;
+	u32 reg_ctrl;
+	u32 reg_status;
+	u32 status_mask;
+	u32 reg_mask;
+
+	if (adev->flags & AMD_IS_APU) {
+		reg_ctrl = ixGNB_CLK3_DFS_CNTL;
+		reg_status = ixGNB_CLK3_STATUS;
+		status_mask = 0x00010000;
+		reg_mask = CG_ECLK_CNTL__ECLK_DIVIDER_MASK;
+	} else {
+		reg_ctrl = ixCG_ECLK_CNTL;
+		reg_status = ixCG_ECLK_STATUS;
+		status_mask = CG_ECLK_STATUS__ECLK_STATUS_MASK;
+		reg_mask = CG_ECLK_CNTL__ECLK_DIR_CNTL_EN_MASK | CG_ECLK_CNTL__ECLK_DIVIDER_MASK;
+	}
 
 	r = amdgpu_atombios_get_clock_dividers(adev,
 					       COMPUTE_GPUCLK_INPUT_FLAG_DEFAULT_GPUCLK,
@@ -773,24 +815,25 @@ static int vi_set_vce_clocks(struct amdgpu_device *adev, u32 evclk, u32 ecclk)
 		return r;
 
 	for (i = 0; i < 100; i++) {
-		if (RREG32_SMC(ixCG_ECLK_STATUS) & CG_ECLK_STATUS__ECLK_STATUS_MASK)
+		if (RREG32_SMC(reg_status) & status_mask)
 			break;
 		mdelay(10);
 	}
+
 	if (i == 100)
 		return -ETIMEDOUT;
 
-	tmp = RREG32_SMC(ixCG_ECLK_CNTL);
-	tmp &= ~(CG_ECLK_CNTL__ECLK_DIR_CNTL_EN_MASK |
-		CG_ECLK_CNTL__ECLK_DIVIDER_MASK);
+	tmp = RREG32_SMC(reg_ctrl);
+	tmp &= ~reg_mask;
 	tmp |= dividers.post_divider;
-	WREG32_SMC(ixCG_ECLK_CNTL, tmp);
+	WREG32_SMC(reg_ctrl, tmp);
 
 	for (i = 0; i < 100; i++) {
-		if (RREG32_SMC(ixCG_ECLK_STATUS) & CG_ECLK_STATUS__ECLK_STATUS_MASK)
+		if (RREG32_SMC(reg_status) & status_mask)
 			break;
 		mdelay(10);
 	}
+
 	if (i == 100)
 		return -ETIMEDOUT;
 
diff --git a/drivers/gpu/drm/arm/malidp_hw.c b/drivers/gpu/drm/arm/malidp_hw.c
index 17bca99e8ac82..7e2c341dfe5f5 100644
--- a/drivers/gpu/drm/arm/malidp_hw.c
+++ b/drivers/gpu/drm/arm/malidp_hw.c
@@ -634,7 +634,8 @@ const struct malidp_hw_device malidp_device[MALIDP_MAX_DEVICES] = {
 				.vsync_irq = MALIDP500_DE_IRQ_VSYNC,
 			},
 			.se_irq_map = {
-				.irq_mask = MALIDP500_SE_IRQ_CONF_MODE,
+				.irq_mask = MALIDP500_SE_IRQ_CONF_MODE |
+					    MALIDP500_SE_IRQ_GLOBAL,
 				.vsync_irq = 0,
 			},
 			.dc_irq_map = {
diff --git a/drivers/gpu/drm/arm/malidp_planes.c b/drivers/gpu/drm/arm/malidp_planes.c
index 94e7e3fa3408c..16b8b310ae5c7 100644
--- a/drivers/gpu/drm/arm/malidp_planes.c
+++ b/drivers/gpu/drm/arm/malidp_planes.c
@@ -23,6 +23,7 @@
 
 /* Layer specific register offsets */
 #define MALIDP_LAYER_FORMAT		0x000
+#define   LAYER_FORMAT_MASK		0x3f
 #define MALIDP_LAYER_CONTROL		0x004
 #define   LAYER_ENABLE			(1 << 0)
 #define   LAYER_FLOWCFG_MASK		7
@@ -278,7 +279,9 @@ static void malidp_de_plane_update(struct drm_plane *plane,
 	dest_w = plane->state->crtc_w;
 	dest_h = plane->state->crtc_h;
 
-	malidp_hw_write(mp->hwdev, ms->format, mp->layer->base);
+	val = malidp_hw_read(mp->hwdev, mp->layer->base);
+	val = (val & ~LAYER_FORMAT_MASK) | ms->format;
+	malidp_hw_write(mp->hwdev, val, mp->layer->base);
 
 	for (i = 0; i < ms->n_planes; i++) {
 		/* calculate the offset for the layer's plane registers */
diff --git a/drivers/gpu/drm/armada/armada_crtc.c b/drivers/gpu/drm/armada/armada_crtc.c
index 79ce877bf45f2..3039936f8f3fc 100644
--- a/drivers/gpu/drm/armada/armada_crtc.c
+++ b/drivers/gpu/drm/armada/armada_crtc.c
@@ -483,8 +483,9 @@ static irqreturn_t armada_drm_irq(int irq, void *arg)
 	u32 v, stat = readl_relaxed(dcrtc->base + LCD_SPU_IRQ_ISR);
 
 	/*
-	 * This is rediculous - rather than writing bits to clear, we
-	 * have to set the actual status register value.  This is racy.
+	 * Reading the ISR appears to clear bits provided CLEAN_SPU_IRQ_ISR
+	 * is set.  Writing has some other effect to acknowledge the IRQ -
+	 * without this, we only get a single IRQ.
 	 */
 	writel_relaxed(0, dcrtc->base + LCD_SPU_IRQ_ISR);
 
@@ -1104,16 +1105,22 @@ armada_drm_crtc_set_property(struct drm_crtc *crtc,
 static int armada_drm_crtc_enable_vblank(struct drm_crtc *crtc)
 {
 	struct armada_crtc *dcrtc = drm_to_armada_crtc(crtc);
+	unsigned long flags;
 
+	spin_lock_irqsave(&dcrtc->irq_lock, flags);
 	armada_drm_crtc_enable_irq(dcrtc, VSYNC_IRQ_ENA);
+	spin_unlock_irqrestore(&dcrtc->irq_lock, flags);
 	return 0;
 }
 
 static void armada_drm_crtc_disable_vblank(struct drm_crtc *crtc)
 {
 	struct armada_crtc *dcrtc = drm_to_armada_crtc(crtc);
+	unsigned long flags;
 
+	spin_lock_irqsave(&dcrtc->irq_lock, flags);
 	armada_drm_crtc_disable_irq(dcrtc, VSYNC_IRQ_ENA);
+	spin_unlock_irqrestore(&dcrtc->irq_lock, flags);
 }
 
 static const struct drm_crtc_funcs armada_crtc_funcs = {
@@ -1221,6 +1228,7 @@ static int armada_drm_crtc_create(struct drm_device *drm, struct device *dev,
 		       CFG_PDWN64x66, dcrtc->base + LCD_SPU_SRAM_PARA1);
 	writel_relaxed(0x2032ff81, dcrtc->base + LCD_SPU_DMA_CTRL1);
 	writel_relaxed(dcrtc->irq_ena, dcrtc->base + LCD_SPU_IRQ_ENA);
+	readl_relaxed(dcrtc->base + LCD_SPU_IRQ_ISR);
 	writel_relaxed(0, dcrtc->base + LCD_SPU_IRQ_ISR);
 
 	ret = devm_request_irq(dev, irq, armada_drm_irq, 0, "armada_drm_crtc",
diff --git a/drivers/gpu/drm/armada/armada_hw.h b/drivers/gpu/drm/armada/armada_hw.h
index 27319a8335e25..345dc4d0851ef 100644
--- a/drivers/gpu/drm/armada/armada_hw.h
+++ b/drivers/gpu/drm/armada/armada_hw.h
@@ -160,6 +160,7 @@ enum {
 	CFG_ALPHAM_GRA		= 0x1 << 16,
 	CFG_ALPHAM_CFG		= 0x2 << 16,
 	CFG_ALPHA_MASK		= 0xff << 8,
+#define CFG_ALPHA(x)		((x) << 8)
 	CFG_PIXCMD_MASK		= 0xff,
 };
 
diff --git a/drivers/gpu/drm/armada/armada_overlay.c b/drivers/gpu/drm/armada/armada_overlay.c
index edc44910d79fc..2076346b09ee9 100644
--- a/drivers/gpu/drm/armada/armada_overlay.c
+++ b/drivers/gpu/drm/armada/armada_overlay.c
@@ -28,6 +28,7 @@ struct armada_ovl_plane_properties {
 	uint16_t contrast;
 	uint16_t saturation;
 	uint32_t colorkey_mode;
+	uint32_t colorkey_enable;
 };
 
 struct armada_ovl_plane {
@@ -59,11 +60,13 @@ armada_ovl_update_attr(struct armada_ovl_plane_properties *prop,
 	writel_relaxed(0x00002000, dcrtc->base + LCD_SPU_CBSH_HUE);
 
 	spin_lock_irq(&dcrtc->irq_lock);
-	armada_updatel(prop->colorkey_mode | CFG_ALPHAM_GRA,
-		     CFG_CKMODE_MASK | CFG_ALPHAM_MASK | CFG_ALPHA_MASK,
-		     dcrtc->base + LCD_SPU_DMA_CTRL1);
-
-	armada_updatel(ADV_GRACOLORKEY, 0, dcrtc->base + LCD_SPU_ADV_REG);
+	armada_updatel(prop->colorkey_mode,
+		       CFG_CKMODE_MASK | CFG_ALPHAM_MASK | CFG_ALPHA_MASK,
+		       dcrtc->base + LCD_SPU_DMA_CTRL1);
+	if (dcrtc->variant->has_spu_adv_reg)
+		armada_updatel(prop->colorkey_enable,
+			       ADV_GRACOLORKEY | ADV_VIDCOLORKEY,
+			       dcrtc->base + LCD_SPU_ADV_REG);
 	spin_unlock_irq(&dcrtc->irq_lock);
 }
 
@@ -339,8 +342,17 @@ static int armada_ovl_plane_set_property(struct drm_plane *plane,
 		dplane->prop.colorkey_vb |= K2B(val);
 		update_attr = true;
 	} else if (property == priv->colorkey_mode_prop) {
-		dplane->prop.colorkey_mode &= ~CFG_CKMODE_MASK;
-		dplane->prop.colorkey_mode |= CFG_CKMODE(val);
+		if (val == CKMODE_DISABLE) {
+			dplane->prop.colorkey_mode =
+				CFG_CKMODE(CKMODE_DISABLE) |
+				CFG_ALPHAM_CFG | CFG_ALPHA(255);
+			dplane->prop.colorkey_enable = 0;
+		} else {
+			dplane->prop.colorkey_mode =
+				CFG_CKMODE(val) |
+				CFG_ALPHAM_GRA | CFG_ALPHA(0);
+			dplane->prop.colorkey_enable = ADV_GRACOLORKEY;
+		}
 		update_attr = true;
 	} else if (property == priv->brightness_prop) {
 		dplane->prop.brightness = val - 256;
@@ -470,7 +482,9 @@ int armada_overlay_plane_create(struct drm_device *dev, unsigned long crtcs)
 	dplane->prop.colorkey_yr = 0xfefefe00;
 	dplane->prop.colorkey_ug = 0x01010100;
 	dplane->prop.colorkey_vb = 0x01010100;
-	dplane->prop.colorkey_mode = CFG_CKMODE(CKMODE_RGB);
+	dplane->prop.colorkey_mode = CFG_CKMODE(CKMODE_RGB) |
+				     CFG_ALPHAM_GRA | CFG_ALPHA(0);
+	dplane->prop.colorkey_enable = ADV_GRACOLORKEY;
 	dplane->prop.brightness = 0;
 	dplane->prop.contrast = 0x4000;
 	dplane->prop.saturation = 0x4000;
diff --git a/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_plane.c b/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_plane.c
index 703c2d13603fc..eb7c4cf19bf6e 100644
--- a/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_plane.c
+++ b/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_plane.c
@@ -889,7 +889,7 @@ static int atmel_hlcdc_plane_init_properties(struct atmel_hlcdc_plane *plane,
 		drm_object_attach_property(&plane->base.base,
 					   props->alpha, 255);
 
-	if (desc->layout.xstride && desc->layout.pstride) {
+	if (desc->layout.xstride[0] && desc->layout.pstride[0]) {
 		int ret;
 
 		ret = drm_plane_create_rotation_property(&plane->base,
diff --git a/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c b/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c
index 558845e826b87..c0a623cf4ae17 100644
--- a/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c
+++ b/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c
@@ -424,6 +424,18 @@ static void adv7511_hpd_work(struct work_struct *work)
 	else
 		status = connector_status_disconnected;
 
+	/*
+	 * The bridge resets its registers on unplug. So when we get a plug
+	 * event and we're already supposed to be powered, cycle the bridge to
+	 * restore its state.
+	 */
+	if (status == connector_status_connected &&
+	    adv7511->connector.status == connector_status_disconnected &&
+	    adv7511->powered) {
+		regcache_mark_dirty(adv7511->regmap);
+		adv7511_power_on(adv7511);
+	}
+
 	if (adv7511->connector.status != status) {
 		adv7511->connector.status = status;
 		drm_kms_helper_hotplug_event(adv7511->connector.dev);
diff --git a/drivers/gpu/drm/bridge/sil-sii8620.c b/drivers/gpu/drm/bridge/sil-sii8620.c
index 5131bfb94f065..0cb69ee94ac16 100644
--- a/drivers/gpu/drm/bridge/sil-sii8620.c
+++ b/drivers/gpu/drm/bridge/sil-sii8620.c
@@ -788,6 +788,7 @@ static void sii8620_burst_rx_all(struct sii8620 *ctx)
 static void sii8620_fetch_edid(struct sii8620 *ctx)
 {
 	u8 lm_ddc, ddc_cmd, int3, cbus;
+	unsigned long timeout;
 	int fetched, i;
 	int edid_len = EDID_LENGTH;
 	u8 *edid;
@@ -837,23 +838,31 @@ static void sii8620_fetch_edid(struct sii8620 *ctx)
 			REG_DDC_CMD, ddc_cmd | VAL_DDC_CMD_ENH_DDC_READ_NO_ACK
 		);
 
-		do {
-			int3 = sii8620_readb(ctx, REG_INTR3);
+		int3 = 0;
+		timeout = jiffies + msecs_to_jiffies(200);
+		for (;;) {
 			cbus = sii8620_readb(ctx, REG_CBUS_STATUS);
-
-			if (int3 & BIT_DDC_CMD_DONE)
-				break;
-
-			if (!(cbus & BIT_CBUS_STATUS_CBUS_CONNECTED)) {
+			if (~cbus & BIT_CBUS_STATUS_CBUS_CONNECTED) {
+				kfree(edid);
+				edid = NULL;
+				goto end;
+			}
+			if (int3 & BIT_DDC_CMD_DONE) {
+				if (sii8620_readb(ctx, REG_DDC_DOUT_CNT)
+				    >= FETCH_SIZE)
+					break;
+			} else {
+				int3 = sii8620_readb(ctx, REG_INTR3);
+			}
+			if (time_is_before_jiffies(timeout)) {
+				ctx->error = -ETIMEDOUT;
+				dev_err(ctx->dev, "timeout during EDID read\n");
 				kfree(edid);
 				edid = NULL;
 				goto end;
 			}
-		} while (1);
-
-		sii8620_readb(ctx, REG_DDC_STATUS);
-		while (sii8620_readb(ctx, REG_DDC_DOUT_CNT) < FETCH_SIZE)
 			usleep_range(10, 20);
+		}
 
 		sii8620_read_buf(ctx, REG_DDC_DATA, edid + fetched, FETCH_SIZE);
 		if (fetched + FETCH_SIZE == EDID_LENGTH) {
@@ -1036,23 +1045,23 @@ static void sii8620_set_format(struct sii8620 *ctx)
 				BIT_M3_P0CTRL_MHL3_P0_PIXEL_MODE_PACKED,
 				ctx->use_packed_pixel ? ~0 : 0);
 	} else {
-		if (ctx->use_packed_pixel)
+		if (ctx->use_packed_pixel) {
 			sii8620_write_seq_static(ctx,
 				REG_VID_MODE, BIT_VID_MODE_M1080P,
 				REG_MHL_TOP_CTL, BIT_MHL_TOP_CTL_MHL_PP_SEL | 1,
 				REG_MHLTX_CTL6, 0x60
 			);
-		else
+		} else {
 			sii8620_write_seq_static(ctx,
 				REG_VID_MODE, 0,
 				REG_MHL_TOP_CTL, 1,
 				REG_MHLTX_CTL6, 0xa0
 			);
+		}
 	}
 
 	if (ctx->use_packed_pixel)
-		out_fmt = VAL_TPI_FORMAT(YCBCR422, FULL) |
-			BIT_TPI_OUTPUT_CSCMODE709;
+		out_fmt = VAL_TPI_FORMAT(YCBCR422, FULL);
 	else
 		out_fmt = VAL_TPI_FORMAT(RGB, FULL);
 
@@ -1187,7 +1196,7 @@ static void sii8620_start_hdmi(struct sii8620 *ctx)
 		int clk = ctx->pixel_clock * (ctx->use_packed_pixel ? 2 : 3);
 		int i;
 
-		for (i = 0; i < ARRAY_SIZE(clk_spec); ++i)
+		for (i = 0; i < ARRAY_SIZE(clk_spec) - 1; ++i)
 			if (clk < clk_spec[i].max_clk)
 				break;
 
diff --git a/drivers/gpu/drm/drm_atomic.c b/drivers/gpu/drm/drm_atomic.c
index 0d8a417e2cd67..bb5cc15fa0b92 100644
--- a/drivers/gpu/drm/drm_atomic.c
+++ b/drivers/gpu/drm/drm_atomic.c
@@ -1355,7 +1355,9 @@ drm_atomic_set_crtc_for_plane(struct drm_plane_state *plane_state,
 {
 	struct drm_plane *plane = plane_state->plane;
 	struct drm_crtc_state *crtc_state;
-
+	/* Nothing to do for same crtc*/
+	if (plane_state->crtc == crtc)
+		return 0;
 	if (plane_state->crtc) {
 		crtc_state = drm_atomic_get_crtc_state(plane_state->state,
 						       plane_state->crtc);
diff --git a/drivers/gpu/drm/drm_atomic_helper.c b/drivers/gpu/drm/drm_atomic_helper.c
index 0028591f3f959..1f08d597b87af 100644
--- a/drivers/gpu/drm/drm_atomic_helper.c
+++ b/drivers/gpu/drm/drm_atomic_helper.c
@@ -2683,31 +2683,9 @@ commit:
 	return 0;
 }
 
-/**
- * drm_atomic_helper_disable_all - disable all currently active outputs
- * @dev: DRM device
- * @ctx: lock acquisition context
- *
- * Loops through all connectors, finding those that aren't turned off and then
- * turns them off by setting their DPMS mode to OFF and deactivating the CRTC
- * that they are connected to.
- *
- * This is used for example in suspend/resume to disable all currently active
- * functions when suspending. If you just want to shut down everything at e.g.
- * driver unload, look at drm_atomic_helper_shutdown().
- *
- * Note that if callers haven't already acquired all modeset locks this might
- * return -EDEADLK, which must be handled by calling drm_modeset_backoff().
- *
- * Returns:
- * 0 on success or a negative error code on failure.
- *
- * See also:
- * drm_atomic_helper_suspend(), drm_atomic_helper_resume() and
- * drm_atomic_helper_shutdown().
- */
-int drm_atomic_helper_disable_all(struct drm_device *dev,
-				  struct drm_modeset_acquire_ctx *ctx)
+static int __drm_atomic_helper_disable_all(struct drm_device *dev,
+					   struct drm_modeset_acquire_ctx *ctx,
+					   bool clean_old_fbs)
 {
 	struct drm_atomic_state *state;
 	struct drm_connector_state *conn_state;
@@ -2759,8 +2737,11 @@ int drm_atomic_helper_disable_all(struct drm_device *dev,
 			goto free;
 
 		drm_atomic_set_fb_for_plane(plane_state, NULL);
-		plane_mask |= BIT(drm_plane_index(plane));
-		plane->old_fb = plane->fb;
+
+		if (clean_old_fbs) {
+			plane->old_fb = plane->fb;
+			plane_mask |= BIT(drm_plane_index(plane));
+		}
 	}
 
 	ret = drm_atomic_commit(state);
@@ -2771,6 +2752,34 @@ free:
 	return ret;
 }
 
+/**
+ * drm_atomic_helper_disable_all - disable all currently active outputs
+ * @dev: DRM device
+ * @ctx: lock acquisition context
+ *
+ * Loops through all connectors, finding those that aren't turned off and then
+ * turns them off by setting their DPMS mode to OFF and deactivating the CRTC
+ * that they are connected to.
+ *
+ * This is used for example in suspend/resume to disable all currently active
+ * functions when suspending. If you just want to shut down everything at e.g.
+ * driver unload, look at drm_atomic_helper_shutdown().
+ *
+ * Note that if callers haven't already acquired all modeset locks this might
+ * return -EDEADLK, which must be handled by calling drm_modeset_backoff().
+ *
+ * Returns:
+ * 0 on success or a negative error code on failure.
+ *
+ * See also:
+ * drm_atomic_helper_suspend(), drm_atomic_helper_resume() and
+ * drm_atomic_helper_shutdown().
+ */
+int drm_atomic_helper_disable_all(struct drm_device *dev,
+				  struct drm_modeset_acquire_ctx *ctx)
+{
+	return __drm_atomic_helper_disable_all(dev, ctx, false);
+}
 EXPORT_SYMBOL(drm_atomic_helper_disable_all);
 
 /**
@@ -2793,7 +2802,7 @@ void drm_atomic_helper_shutdown(struct drm_device *dev)
 	while (1) {
 		ret = drm_modeset_lock_all_ctx(dev, &ctx);
 		if (!ret)
-			ret = drm_atomic_helper_disable_all(dev, &ctx);
+			ret = __drm_atomic_helper_disable_all(dev, &ctx, true);
 
 		if (ret != -EDEADLK)
 			break;
@@ -2897,16 +2906,11 @@ int drm_atomic_helper_commit_duplicated_state(struct drm_atomic_state *state,
 	struct drm_connector_state *new_conn_state;
 	struct drm_crtc *crtc;
 	struct drm_crtc_state *new_crtc_state;
-	unsigned plane_mask = 0;
-	struct drm_device *dev = state->dev;
-	int ret;
 
 	state->acquire_ctx = ctx;
 
-	for_each_new_plane_in_state(state, plane, new_plane_state, i) {
-		plane_mask |= BIT(drm_plane_index(plane));
+	for_each_new_plane_in_state(state, plane, new_plane_state, i)
 		state->planes[i].old_state = plane->state;
-	}
 
 	for_each_new_crtc_in_state(state, crtc, new_crtc_state, i)
 		state->crtcs[i].old_state = crtc->state;
@@ -2914,11 +2918,7 @@ int drm_atomic_helper_commit_duplicated_state(struct drm_atomic_state *state,
 	for_each_new_connector_in_state(state, connector, new_conn_state, i)
 		state->connectors[i].old_state = connector->state;
 
-	ret = drm_atomic_commit(state);
-	if (plane_mask)
-		drm_atomic_clean_old_fb(dev, plane_mask, ret);
-
-	return ret;
+	return drm_atomic_commit(state);
 }
 EXPORT_SYMBOL(drm_atomic_helper_commit_duplicated_state);
 
diff --git a/drivers/gpu/drm/drm_dp_mst_topology.c b/drivers/gpu/drm/drm_dp_mst_topology.c
index 41b492f99955f..c022ab6e84bdd 100644
--- a/drivers/gpu/drm/drm_dp_mst_topology.c
+++ b/drivers/gpu/drm/drm_dp_mst_topology.c
@@ -2862,12 +2862,14 @@ static void drm_dp_mst_dump_mstb(struct seq_file *m,
 	}
 }
 
+#define DP_PAYLOAD_TABLE_SIZE		64
+
 static bool dump_dp_payload_table(struct drm_dp_mst_topology_mgr *mgr,
 				  char *buf)
 {
 	int i;
 
-	for (i = 0; i < 64; i += 16) {
+	for (i = 0; i < DP_PAYLOAD_TABLE_SIZE; i += 16) {
 		if (drm_dp_dpcd_read(mgr->aux,
 				     DP_PAYLOAD_TABLE_UPDATE_STATUS + i,
 				     &buf[i], 16) != 16)
@@ -2936,7 +2938,7 @@ void drm_dp_mst_dump_topology(struct seq_file *m,
 
 	mutex_lock(&mgr->lock);
 	if (mgr->mst_primary) {
-		u8 buf[64];
+		u8 buf[DP_PAYLOAD_TABLE_SIZE];
 		int ret;
 
 		ret = drm_dp_dpcd_read(mgr->aux, DP_DPCD_REV, buf, DP_RECEIVER_CAP_SIZE);
@@ -2954,8 +2956,7 @@ void drm_dp_mst_dump_topology(struct seq_file *m,
 		seq_printf(m, " revision: hw: %x.%x sw: %x.%x\n",
 			   buf[0x9] >> 4, buf[0x9] & 0xf, buf[0xa], buf[0xb]);
 		if (dump_dp_payload_table(mgr, buf))
-			seq_printf(m, "payload table: %*ph\n", 63, buf);
-
+			seq_printf(m, "payload table: %*ph\n", DP_PAYLOAD_TABLE_SIZE, buf);
 	}
 
 	mutex_unlock(&mgr->lock);
diff --git a/drivers/gpu/drm/drm_property.c b/drivers/gpu/drm/drm_property.c
index bc5128203056d..78e6307712149 100644
--- a/drivers/gpu/drm/drm_property.c
+++ b/drivers/gpu/drm/drm_property.c
@@ -516,7 +516,7 @@ static void drm_property_free_blob(struct kref *kref)
 
 	drm_mode_object_unregister(blob->dev, &blob->base);
 
-	kfree(blob);
+	kvfree(blob);
 }
 
 /**
@@ -543,7 +543,7 @@ drm_property_create_blob(struct drm_device *dev, size_t length,
 	if (!length || length > ULONG_MAX - sizeof(struct drm_property_blob))
 		return ERR_PTR(-EINVAL);
 
-	blob = kzalloc(sizeof(struct drm_property_blob)+length, GFP_KERNEL);
+	blob = kvzalloc(sizeof(struct drm_property_blob)+length, GFP_KERNEL);
 	if (!blob)
 		return ERR_PTR(-ENOMEM);
 
@@ -559,7 +559,7 @@ drm_property_create_blob(struct drm_device *dev, size_t length,
 	ret = __drm_mode_object_add(dev, &blob->base, DRM_MODE_OBJECT_BLOB,
 				    true, drm_property_free_blob);
 	if (ret) {
-		kfree(blob);
+		kvfree(blob);
 		return ERR_PTR(-EINVAL);
 	}
 
diff --git a/drivers/gpu/drm/exynos/exynos5433_drm_decon.c b/drivers/gpu/drm/exynos/exynos5433_drm_decon.c
index 6be5b53c3b279..f905c214fdd0e 100644
--- a/drivers/gpu/drm/exynos/exynos5433_drm_decon.c
+++ b/drivers/gpu/drm/exynos/exynos5433_drm_decon.c
@@ -261,7 +261,7 @@ static void decon_win_set_pixfmt(struct decon_context *ctx, unsigned int win,
 	unsigned long val;
 
 	val = readl(ctx->addr + DECON_WINCONx(win));
-	val &= ~WINCONx_BPPMODE_MASK;
+	val &= WINCONx_ENWIN_F;
 
 	switch (fb->format->format) {
 	case DRM_FORMAT_XRGB1555:
@@ -352,8 +352,8 @@ static void decon_update_plane(struct exynos_drm_crtc *crtc,
 		writel(val, ctx->addr + DECON_VIDOSDxB(win));
 	}
 
-	val = VIDOSD_Wx_ALPHA_R_F(0x0) | VIDOSD_Wx_ALPHA_G_F(0x0) |
-		VIDOSD_Wx_ALPHA_B_F(0x0);
+	val = VIDOSD_Wx_ALPHA_R_F(0xff) | VIDOSD_Wx_ALPHA_G_F(0xff) |
+		VIDOSD_Wx_ALPHA_B_F(0xff);
 	writel(val, ctx->addr + DECON_VIDOSDxC(win));
 
 	val = VIDOSD_Wx_ALPHA_R_F(0x0) | VIDOSD_Wx_ALPHA_G_F(0x0) |
diff --git a/drivers/gpu/drm/exynos/exynos_drm_gsc.c b/drivers/gpu/drm/exynos/exynos_drm_gsc.c
index 0506b2b17ac1c..48f913d8208c4 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_gsc.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_gsc.c
@@ -532,21 +532,25 @@ static int gsc_src_set_fmt(struct device *dev, u32 fmt)
 			GSC_IN_CHROMA_ORDER_CRCB);
 		break;
 	case DRM_FORMAT_NV21:
+		cfg |= (GSC_IN_CHROMA_ORDER_CRCB | GSC_IN_YUV420_2P);
+		break;
 	case DRM_FORMAT_NV61:
-		cfg |= (GSC_IN_CHROMA_ORDER_CRCB |
-			GSC_IN_YUV420_2P);
+		cfg |= (GSC_IN_CHROMA_ORDER_CRCB | GSC_IN_YUV422_2P);
 		break;
 	case DRM_FORMAT_YUV422:
 		cfg |= GSC_IN_YUV422_3P;
 		break;
 	case DRM_FORMAT_YUV420:
+		cfg |= (GSC_IN_CHROMA_ORDER_CBCR | GSC_IN_YUV420_3P);
+		break;
 	case DRM_FORMAT_YVU420:
-		cfg |= GSC_IN_YUV420_3P;
+		cfg |= (GSC_IN_CHROMA_ORDER_CRCB | GSC_IN_YUV420_3P);
 		break;
 	case DRM_FORMAT_NV12:
+		cfg |= (GSC_IN_CHROMA_ORDER_CBCR | GSC_IN_YUV420_2P);
+		break;
 	case DRM_FORMAT_NV16:
-		cfg |= (GSC_IN_CHROMA_ORDER_CBCR |
-			GSC_IN_YUV420_2P);
+		cfg |= (GSC_IN_CHROMA_ORDER_CBCR | GSC_IN_YUV422_2P);
 		break;
 	default:
 		dev_err(ippdrv->dev, "invalid target yuv order 0x%x.\n", fmt);
@@ -806,18 +810,25 @@ static int gsc_dst_set_fmt(struct device *dev, u32 fmt)
 			GSC_OUT_CHROMA_ORDER_CRCB);
 		break;
 	case DRM_FORMAT_NV21:
-	case DRM_FORMAT_NV61:
 		cfg |= (GSC_OUT_CHROMA_ORDER_CRCB | GSC_OUT_YUV420_2P);
 		break;
+	case DRM_FORMAT_NV61:
+		cfg |= (GSC_OUT_CHROMA_ORDER_CRCB | GSC_OUT_YUV422_2P);
+		break;
 	case DRM_FORMAT_YUV422:
+		cfg |= GSC_OUT_YUV422_3P;
+		break;
 	case DRM_FORMAT_YUV420:
+		cfg |= (GSC_OUT_CHROMA_ORDER_CBCR | GSC_OUT_YUV420_3P);
+		break;
 	case DRM_FORMAT_YVU420:
-		cfg |= GSC_OUT_YUV420_3P;
+		cfg |= (GSC_OUT_CHROMA_ORDER_CRCB | GSC_OUT_YUV420_3P);
 		break;
 	case DRM_FORMAT_NV12:
+		cfg |= (GSC_OUT_CHROMA_ORDER_CBCR | GSC_OUT_YUV420_2P);
+		break;
 	case DRM_FORMAT_NV16:
-		cfg |= (GSC_OUT_CHROMA_ORDER_CBCR |
-			GSC_OUT_YUV420_2P);
+		cfg |= (GSC_OUT_CHROMA_ORDER_CBCR | GSC_OUT_YUV422_2P);
 		break;
 	default:
 		dev_err(ippdrv->dev, "invalid target yuv order 0x%x.\n", fmt);
diff --git a/drivers/gpu/drm/exynos/regs-gsc.h b/drivers/gpu/drm/exynos/regs-gsc.h
index 4704a993cbb7f..16b39734115c9 100644
--- a/drivers/gpu/drm/exynos/regs-gsc.h
+++ b/drivers/gpu/drm/exynos/regs-gsc.h
@@ -138,6 +138,7 @@
 #define GSC_OUT_YUV420_3P		(3 << 4)
 #define GSC_OUT_YUV422_1P		(4 << 4)
 #define GSC_OUT_YUV422_2P		(5 << 4)
+#define GSC_OUT_YUV422_3P		(6 << 4)
 #define GSC_OUT_YUV444			(7 << 4)
 #define GSC_OUT_TILE_TYPE_MASK		(1 << 2)
 #define GSC_OUT_TILE_C_16x8		(0 << 2)
diff --git a/drivers/gpu/drm/gma500/psb_intel_drv.h b/drivers/gpu/drm/gma500/psb_intel_drv.h
index e8e4ea14b12ba..e05e5399af2db 100644
--- a/drivers/gpu/drm/gma500/psb_intel_drv.h
+++ b/drivers/gpu/drm/gma500/psb_intel_drv.h
@@ -255,7 +255,7 @@ extern int intelfb_remove(struct drm_device *dev,
 extern bool psb_intel_lvds_mode_fixup(struct drm_encoder *encoder,
 				      const struct drm_display_mode *mode,
 				      struct drm_display_mode *adjusted_mode);
-extern int psb_intel_lvds_mode_valid(struct drm_connector *connector,
+extern enum drm_mode_status psb_intel_lvds_mode_valid(struct drm_connector *connector,
 				     struct drm_display_mode *mode);
 extern int psb_intel_lvds_set_property(struct drm_connector *connector,
 					struct drm_property *property,
diff --git a/drivers/gpu/drm/gma500/psb_intel_lvds.c b/drivers/gpu/drm/gma500/psb_intel_lvds.c
index be3eefec5152a..8baf6325c6e46 100644
--- a/drivers/gpu/drm/gma500/psb_intel_lvds.c
+++ b/drivers/gpu/drm/gma500/psb_intel_lvds.c
@@ -343,7 +343,7 @@ static void psb_intel_lvds_restore(struct drm_connector *connector)
 	}
 }
 
-int psb_intel_lvds_mode_valid(struct drm_connector *connector,
+enum drm_mode_status psb_intel_lvds_mode_valid(struct drm_connector *connector,
 				 struct drm_display_mode *mode)
 {
 	struct drm_psb_private *dev_priv = connector->dev->dev_private;
diff --git a/drivers/gpu/drm/i915/gvt/kvmgt.c b/drivers/gpu/drm/i915/gvt/kvmgt.c
index 9bf4045cd679a..73c672fc17c4a 100644
--- a/drivers/gpu/drm/i915/gvt/kvmgt.c
+++ b/drivers/gpu/drm/i915/gvt/kvmgt.c
@@ -42,6 +42,8 @@
 #include <linux/vfio.h>
 #include <linux/mdev.h>
 
+#include <linux/nospec.h>
+
 #include "i915_drv.h"
 #include "gvt.h"
 
@@ -953,7 +955,8 @@ static long intel_vgpu_ioctl(struct mdev_device *mdev, unsigned int cmd,
 	} else if (cmd == VFIO_DEVICE_GET_REGION_INFO) {
 		struct vfio_region_info info;
 		struct vfio_info_cap caps = { .buf = NULL, .size = 0 };
-		int i, ret;
+		unsigned int i;
+		int ret;
 		struct vfio_region_info_cap_sparse_mmap *sparse = NULL;
 		size_t size;
 		int nr_areas = 1;
@@ -1030,6 +1033,10 @@ static long intel_vgpu_ioctl(struct mdev_device *mdev, unsigned int cmd,
 				if (info.index >= VFIO_PCI_NUM_REGIONS +
 						vgpu->vdev.num_regions)
 					return -EINVAL;
+				info.index =
+					array_index_nospec(info.index,
+							VFIO_PCI_NUM_REGIONS +
+							vgpu->vdev.num_regions);
 
 				i = info.index - VFIO_PCI_NUM_REGIONS;
 
diff --git a/drivers/gpu/drm/i915/i915_gem_userptr.c b/drivers/gpu/drm/i915/i915_gem_userptr.c
index 709efe2357eac..05ae8c4a8a1b6 100644
--- a/drivers/gpu/drm/i915/i915_gem_userptr.c
+++ b/drivers/gpu/drm/i915/i915_gem_userptr.c
@@ -782,6 +782,9 @@ i915_gem_userptr_ioctl(struct drm_device *dev, void *data, struct drm_file *file
 			    I915_USERPTR_UNSYNCHRONIZED))
 		return -EINVAL;
 
+	if (!args->user_size)
+		return -EINVAL;
+
 	if (offset_in_page(args->user_ptr | args->user_size))
 		return -EINVAL;
 
diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index b63893eeca73d..20a471ad0ad27 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -1786,10 +1786,38 @@ static void valleyview_pipestat_irq_handler(struct drm_i915_private *dev_priv,
 
 static u32 i9xx_hpd_irq_ack(struct drm_i915_private *dev_priv)
 {
-	u32 hotplug_status = I915_READ(PORT_HOTPLUG_STAT);
+	u32 hotplug_status = 0, hotplug_status_mask;
+	int i;
+
+	if (IS_G4X(dev_priv) ||
+	    IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv))
+		hotplug_status_mask = HOTPLUG_INT_STATUS_G4X |
+			DP_AUX_CHANNEL_MASK_INT_STATUS_G4X;
+	else
+		hotplug_status_mask = HOTPLUG_INT_STATUS_I915;
 
-	if (hotplug_status)
+	/*
+	 * We absolutely have to clear all the pending interrupt
+	 * bits in PORT_HOTPLUG_STAT. Otherwise the ISR port
+	 * interrupt bit won't have an edge, and the i965/g4x
+	 * edge triggered IIR will not notice that an interrupt
+	 * is still pending. We can't use PORT_HOTPLUG_EN to
+	 * guarantee the edge as the act of toggling the enable
+	 * bits can itself generate a new hotplug interrupt :(
+	 */
+	for (i = 0; i < 10; i++) {
+		u32 tmp = I915_READ(PORT_HOTPLUG_STAT) & hotplug_status_mask;
+
+		if (tmp == 0)
+			return hotplug_status;
+
+		hotplug_status |= tmp;
 		I915_WRITE(PORT_HOTPLUG_STAT, hotplug_status);
+	}
+
+	WARN_ONCE(1,
+		  "PORT_HOTPLUG_STAT did not clear (0x%08x)\n",
+		  I915_READ(PORT_HOTPLUG_STAT));
 
 	return hotplug_status;
 }
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 61a2203b75df9..be813b2738c15 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -2484,12 +2484,17 @@ enum i915_power_well_id {
 #define _3D_CHICKEN	_MMIO(0x2084)
 #define  _3D_CHICKEN_HIZ_PLANE_DISABLE_MSAA_4X_SNB	(1 << 10)
 #define _3D_CHICKEN2	_MMIO(0x208c)
+
+#define FF_SLICE_CHICKEN	_MMIO(0x2088)
+#define  FF_SLICE_CHICKEN_CL_PROVOKING_VERTEX_FIX	(1 << 1)
+
 /* Disables pipelining of read flushes past the SF-WIZ interface.
  * Required on all Ironlake steppings according to the B-Spec, but the
  * particular danger of not doing so is not specified.
  */
 # define _3D_CHICKEN2_WM_READ_PIPELINED			(1 << 14)
 #define _3D_CHICKEN3	_MMIO(0x2090)
+#define  _3D_CHICKEN_SF_PROVOKING_VERTEX_FIX		(1 << 12)
 #define  _3D_CHICKEN_SF_DISABLE_OBJEND_CULL		(1 << 10)
 #define  _3D_CHICKEN3_SF_DISABLE_FASTCLIP_CULL		(1 << 5)
 #define  _3D_CHICKEN_SDE_LIMIT_FIFO_POLY_DEPTH(x)	((x)<<1) /* gen8+ */
diff --git a/drivers/gpu/drm/i915/intel_lpe_audio.c b/drivers/gpu/drm/i915/intel_lpe_audio.c
index 3bf65288ffffd..2fdf302ebdad0 100644
--- a/drivers/gpu/drm/i915/intel_lpe_audio.c
+++ b/drivers/gpu/drm/i915/intel_lpe_audio.c
@@ -62,6 +62,7 @@
 
 #include <linux/acpi.h>
 #include <linux/device.h>
+#include <linux/irq.h>
 #include <linux/pci.h>
 #include <linux/pm_runtime.h>
 
diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c
index 6f972e6ec6639..d638b641b7601 100644
--- a/drivers/gpu/drm/i915/intel_lrc.c
+++ b/drivers/gpu/drm/i915/intel_lrc.c
@@ -1067,11 +1067,21 @@ static u32 *gen9_init_indirectctx_bb(struct intel_engine_cs *engine, u32 *batch)
 	/* WaFlushCoherentL3CacheLinesAtContextSwitch:skl,bxt,glk */
 	batch = gen8_emit_flush_coherentl3_wa(engine, batch);
 
+	*batch++ = MI_LOAD_REGISTER_IMM(3);
+
 	/* WaDisableGatherAtSetShaderCommonSlice:skl,bxt,kbl,glk */
-	*batch++ = MI_LOAD_REGISTER_IMM(1);
 	*batch++ = i915_mmio_reg_offset(COMMON_SLICE_CHICKEN2);
 	*batch++ = _MASKED_BIT_DISABLE(
 			GEN9_DISABLE_GATHER_AT_SET_SHADER_COMMON_SLICE);
+
+	/* BSpec: 11391 */
+	*batch++ = i915_mmio_reg_offset(FF_SLICE_CHICKEN);
+	*batch++ = _MASKED_BIT_ENABLE(FF_SLICE_CHICKEN_CL_PROVOKING_VERTEX_FIX);
+
+	/* BSpec: 11299 */
+	*batch++ = i915_mmio_reg_offset(_3D_CHICKEN3);
+	*batch++ = _MASKED_BIT_ENABLE(_3D_CHICKEN_SF_PROVOKING_VERTEX_FIX);
+
 	*batch++ = MI_NOOP;
 
 	/* WaClearSlmSpaceAtContextSwitch:kbl */
diff --git a/drivers/gpu/drm/imx/imx-ldb.c b/drivers/gpu/drm/imx/imx-ldb.c
index 56dd7a9a8e254..dd5312b02a8d2 100644
--- a/drivers/gpu/drm/imx/imx-ldb.c
+++ b/drivers/gpu/drm/imx/imx-ldb.c
@@ -612,6 +612,9 @@ static int imx_ldb_bind(struct device *dev, struct device *master, void *data)
 		return PTR_ERR(imx_ldb->regmap);
 	}
 
+	/* disable LDB by resetting the control register to POR default */
+	regmap_write(imx_ldb->regmap, IOMUXC_GPR2, 0);
+
 	imx_ldb->dev = dev;
 
 	if (of_id)
@@ -652,14 +655,14 @@ static int imx_ldb_bind(struct device *dev, struct device *master, void *data)
 		if (ret || i < 0 || i > 1)
 			return -EINVAL;
 
+		if (!of_device_is_available(child))
+			continue;
+
 		if (dual && i > 0) {
 			dev_warn(dev, "dual-channel mode, ignoring second output\n");
 			continue;
 		}
 
-		if (!of_device_is_available(child))
-			continue;
-
 		channel = &imx_ldb->channel[i];
 		channel->ldb = imx_ldb;
 		channel->chno = i;
diff --git a/drivers/gpu/drm/nouveau/dispnv04/disp.c b/drivers/gpu/drm/nouveau/dispnv04/disp.c
index 5b9d549aa791f..e7926da59214f 100644
--- a/drivers/gpu/drm/nouveau/dispnv04/disp.c
+++ b/drivers/gpu/drm/nouveau/dispnv04/disp.c
@@ -55,6 +55,9 @@ nv04_display_create(struct drm_device *dev)
 	nouveau_display(dev)->init = nv04_display_init;
 	nouveau_display(dev)->fini = nv04_display_fini;
 
+	/* Pre-nv50 doesn't support atomic, so don't expose the ioctls */
+	dev->driver->driver_features &= ~DRIVER_ATOMIC;
+
 	nouveau_hw_save_vga_fonts(dev, 1);
 
 	nv04_crtc_create(dev, 0);
diff --git a/drivers/gpu/drm/nouveau/nouveau_backlight.c b/drivers/gpu/drm/nouveau/nouveau_backlight.c
index debbbf0fd4bdd..408b955e5c39a 100644
--- a/drivers/gpu/drm/nouveau/nouveau_backlight.c
+++ b/drivers/gpu/drm/nouveau/nouveau_backlight.c
@@ -267,6 +267,7 @@ nouveau_backlight_init(struct drm_device *dev)
 	struct nouveau_drm *drm = nouveau_drm(dev);
 	struct nvif_device *device = &drm->client.device;
 	struct drm_connector *connector;
+	struct drm_connector_list_iter conn_iter;
 
 	INIT_LIST_HEAD(&drm->bl_connectors);
 
@@ -275,7 +276,8 @@ nouveau_backlight_init(struct drm_device *dev)
 		return 0;
 	}
 
-	list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
+	drm_connector_list_iter_begin(dev, &conn_iter);
+	drm_for_each_connector_iter(connector, &conn_iter) {
 		if (connector->connector_type != DRM_MODE_CONNECTOR_LVDS &&
 		    connector->connector_type != DRM_MODE_CONNECTOR_eDP)
 			continue;
@@ -292,7 +294,7 @@ nouveau_backlight_init(struct drm_device *dev)
 			break;
 		}
 	}
-
+	drm_connector_list_iter_end(&conn_iter);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/nouveau/nouveau_connector.c b/drivers/gpu/drm/nouveau/nouveau_connector.c
index c902a851eb512..430830d63a33d 100644
--- a/drivers/gpu/drm/nouveau/nouveau_connector.c
+++ b/drivers/gpu/drm/nouveau/nouveau_connector.c
@@ -1208,14 +1208,19 @@ nouveau_connector_create(struct drm_device *dev, int index)
 	struct nouveau_display *disp = nouveau_display(dev);
 	struct nouveau_connector *nv_connector = NULL;
 	struct drm_connector *connector;
+	struct drm_connector_list_iter conn_iter;
 	int type, ret = 0;
 	bool dummy;
 
-	list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
+	drm_connector_list_iter_begin(dev, &conn_iter);
+	nouveau_for_each_non_mst_connector_iter(connector, &conn_iter) {
 		nv_connector = nouveau_connector(connector);
-		if (nv_connector->index == index)
+		if (nv_connector->index == index) {
+			drm_connector_list_iter_end(&conn_iter);
 			return connector;
+		}
 	}
+	drm_connector_list_iter_end(&conn_iter);
 
 	nv_connector = kzalloc(sizeof(*nv_connector), GFP_KERNEL);
 	if (!nv_connector)
diff --git a/drivers/gpu/drm/nouveau/nouveau_connector.h b/drivers/gpu/drm/nouveau/nouveau_connector.h
index a4d1a059bd3d4..dc7454e7f19aa 100644
--- a/drivers/gpu/drm/nouveau/nouveau_connector.h
+++ b/drivers/gpu/drm/nouveau/nouveau_connector.h
@@ -33,6 +33,7 @@
 #include <drm/drm_encoder.h>
 #include <drm/drm_dp_helper.h>
 #include "nouveau_crtc.h"
+#include "nouveau_encoder.h"
 
 struct nvkm_i2c_port;
 
@@ -60,19 +61,46 @@ static inline struct nouveau_connector *nouveau_connector(
 	return container_of(con, struct nouveau_connector, base);
 }
 
+static inline bool
+nouveau_connector_is_mst(struct drm_connector *connector)
+{
+	const struct nouveau_encoder *nv_encoder;
+	const struct drm_encoder *encoder;
+
+	if (connector->connector_type != DRM_MODE_CONNECTOR_DisplayPort)
+		return false;
+
+	nv_encoder = find_encoder(connector, DCB_OUTPUT_ANY);
+	if (!nv_encoder)
+		return false;
+
+	encoder = &nv_encoder->base.base;
+	return encoder->encoder_type == DRM_MODE_ENCODER_DPMST;
+}
+
+#define nouveau_for_each_non_mst_connector_iter(connector, iter) \
+	drm_for_each_connector_iter(connector, iter) \
+		for_each_if(!nouveau_connector_is_mst(connector))
+
 static inline struct nouveau_connector *
 nouveau_crtc_connector_get(struct nouveau_crtc *nv_crtc)
 {
 	struct drm_device *dev = nv_crtc->base.dev;
 	struct drm_connector *connector;
+	struct drm_connector_list_iter conn_iter;
+	struct nouveau_connector *nv_connector = NULL;
 	struct drm_crtc *crtc = to_drm_crtc(nv_crtc);
 
-	list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
-		if (connector->encoder && connector->encoder->crtc == crtc)
-			return nouveau_connector(connector);
+	drm_connector_list_iter_begin(dev, &conn_iter);
+	nouveau_for_each_non_mst_connector_iter(connector, &conn_iter) {
+		if (connector->encoder && connector->encoder->crtc == crtc) {
+			nv_connector = nouveau_connector(connector);
+			break;
+		}
 	}
+	drm_connector_list_iter_end(&conn_iter);
 
-	return NULL;
+	return nv_connector;
 }
 
 struct drm_connector *
diff --git a/drivers/gpu/drm/nouveau/nouveau_display.c b/drivers/gpu/drm/nouveau/nouveau_display.c
index 2e7785f49e6d5..caf53503c0f7a 100644
--- a/drivers/gpu/drm/nouveau/nouveau_display.c
+++ b/drivers/gpu/drm/nouveau/nouveau_display.c
@@ -405,6 +405,7 @@ nouveau_display_init(struct drm_device *dev)
 	struct nouveau_display *disp = nouveau_display(dev);
 	struct nouveau_drm *drm = nouveau_drm(dev);
 	struct drm_connector *connector;
+	struct drm_connector_list_iter conn_iter;
 	int ret;
 
 	ret = disp->init(dev);
@@ -412,10 +413,12 @@ nouveau_display_init(struct drm_device *dev)
 		return ret;
 
 	/* enable hotplug interrupts */
-	list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
+	drm_connector_list_iter_begin(dev, &conn_iter);
+	nouveau_for_each_non_mst_connector_iter(connector, &conn_iter) {
 		struct nouveau_connector *conn = nouveau_connector(connector);
 		nvif_notify_get(&conn->hpd);
 	}
+	drm_connector_list_iter_end(&conn_iter);
 
 	/* enable flip completion events */
 	nvif_notify_get(&drm->flip);
@@ -428,6 +431,7 @@ nouveau_display_fini(struct drm_device *dev, bool suspend)
 	struct nouveau_display *disp = nouveau_display(dev);
 	struct nouveau_drm *drm = nouveau_drm(dev);
 	struct drm_connector *connector;
+	struct drm_connector_list_iter conn_iter;
 
 	if (!suspend) {
 		if (drm_drv_uses_atomic_modeset(dev))
@@ -440,10 +444,12 @@ nouveau_display_fini(struct drm_device *dev, bool suspend)
 	nvif_notify_put(&drm->flip);
 
 	/* disable hotplug interrupts */
-	list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
+	drm_connector_list_iter_begin(dev, &conn_iter);
+	nouveau_for_each_non_mst_connector_iter(connector, &conn_iter) {
 		struct nouveau_connector *conn = nouveau_connector(connector);
 		nvif_notify_put(&conn->hpd);
 	}
+	drm_connector_list_iter_end(&conn_iter);
 
 	drm_kms_helper_poll_disable(dev);
 	disp->fini(dev);
diff --git a/drivers/gpu/drm/nouveau/nouveau_drm.c b/drivers/gpu/drm/nouveau/nouveau_drm.c
index 595630d1fb9e2..362a34cb435db 100644
--- a/drivers/gpu/drm/nouveau/nouveau_drm.c
+++ b/drivers/gpu/drm/nouveau/nouveau_drm.c
@@ -79,6 +79,10 @@ MODULE_PARM_DESC(modeset, "enable driver (default: auto, "
 int nouveau_modeset = -1;
 module_param_named(modeset, nouveau_modeset, int, 0400);
 
+MODULE_PARM_DESC(atomic, "Expose atomic ioctl (default: disabled)");
+static int nouveau_atomic = 0;
+module_param_named(atomic, nouveau_atomic, int, 0400);
+
 MODULE_PARM_DESC(runpm, "disable (0), force enable (1), optimus only default (-1)");
 static int nouveau_runtime_pm = -1;
 module_param_named(runpm, nouveau_runtime_pm, int, 0400);
@@ -383,6 +387,9 @@ static int nouveau_drm_probe(struct pci_dev *pdev,
 
 	pci_set_master(pdev);
 
+	if (nouveau_atomic)
+		driver_pci.driver_features |= DRIVER_ATOMIC;
+
 	ret = drm_get_pci_dev(pdev, pent, &driver_pci);
 	if (ret) {
 		nvkm_device_del(&device);
diff --git a/drivers/gpu/drm/nouveau/nouveau_gem.c b/drivers/gpu/drm/nouveau/nouveau_gem.c
index 2170534101caf..60ffb70bb9089 100644
--- a/drivers/gpu/drm/nouveau/nouveau_gem.c
+++ b/drivers/gpu/drm/nouveau/nouveau_gem.c
@@ -599,7 +599,7 @@ nouveau_gem_pushbuf_reloc_apply(struct nouveau_cli *cli,
 		struct nouveau_bo *nvbo;
 		uint32_t data;
 
-		if (unlikely(r->bo_index > req->nr_buffers)) {
+		if (unlikely(r->bo_index >= req->nr_buffers)) {
 			NV_PRINTK(err, cli, "reloc bo index invalid\n");
 			ret = -EINVAL;
 			break;
@@ -609,7 +609,7 @@ nouveau_gem_pushbuf_reloc_apply(struct nouveau_cli *cli,
 		if (b->presumed.valid)
 			continue;
 
-		if (unlikely(r->reloc_bo_index > req->nr_buffers)) {
+		if (unlikely(r->reloc_bo_index >= req->nr_buffers)) {
 			NV_PRINTK(err, cli, "reloc container bo index invalid\n");
 			ret = -EINVAL;
 			break;
diff --git a/drivers/gpu/drm/nouveau/nv50_display.c b/drivers/gpu/drm/nouveau/nv50_display.c
index a29474528e850..926ec51ba5be1 100644
--- a/drivers/gpu/drm/nouveau/nv50_display.c
+++ b/drivers/gpu/drm/nouveau/nv50_display.c
@@ -4150,7 +4150,7 @@ nv50_disp_atomic_commit(struct drm_device *dev,
 		nv50_disp_atomic_commit_tail(state);
 
 	drm_for_each_crtc(crtc, dev) {
-		if (crtc->state->enable) {
+		if (crtc->state->active) {
 			if (!drm->have_disp_power_ref) {
 				drm->have_disp_power_ref = true;
 				return 0;
@@ -4398,10 +4398,6 @@ nv50_display_destroy(struct drm_device *dev)
 	kfree(disp);
 }
 
-MODULE_PARM_DESC(atomic, "Expose atomic ioctl (default: disabled)");
-static int nouveau_atomic = 0;
-module_param_named(atomic, nouveau_atomic, int, 0400);
-
 int
 nv50_display_create(struct drm_device *dev)
 {
@@ -4426,8 +4422,6 @@ nv50_display_create(struct drm_device *dev)
 	disp->disp = &nouveau_display(dev)->disp;
 	dev->mode_config.funcs = &nv50_disp_func;
 	dev->driver->driver_features |= DRIVER_PREFER_XBGR_30BPP;
-	if (nouveau_atomic)
-		dev->driver->driver_features |= DRIVER_ATOMIC;
 
 	/* small shared memory area we use for notifiers and semaphores */
 	ret = nouveau_bo_new(&drm->client, 4096, 0x1000, TTM_PL_FLAG_VRAM,
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk104.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk104.c
index a7e55c422501c..0b632dc0cf7d4 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk104.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk104.c
@@ -155,10 +155,10 @@ gk104_fifo_runlist_commit(struct gk104_fifo *fifo, int runl)
 				    (target << 28));
 	nvkm_wr32(device, 0x002274, (runl << 20) | nr);
 
-	if (wait_event_timeout(fifo->runlist[runl].wait,
-			       !(nvkm_rd32(device, 0x002284 + (runl * 0x08))
-				       & 0x00100000),
-			       msecs_to_jiffies(2000)) == 0)
+	if (nvkm_msec(device, 2000,
+		if (!(nvkm_rd32(device, 0x002284 + (runl * 0x08)) & 0x00100000))
+			break;
+	) < 0)
 		nvkm_error(subdev, "runlist %d update timeout\n", runl);
 unlock:
 	mutex_unlock(&subdev->mutex);
diff --git a/drivers/gpu/drm/qxl/qxl_display.c b/drivers/gpu/drm/qxl/qxl_display.c
index 9a9214ae0fb50..573bab2221230 100644
--- a/drivers/gpu/drm/qxl/qxl_display.c
+++ b/drivers/gpu/drm/qxl/qxl_display.c
@@ -630,7 +630,7 @@ static void qxl_cursor_atomic_update(struct drm_plane *plane,
 	struct qxl_cursor_cmd *cmd;
 	struct qxl_cursor *cursor;
 	struct drm_gem_object *obj;
-	struct qxl_bo *cursor_bo = NULL, *user_bo = NULL;
+	struct qxl_bo *cursor_bo = NULL, *user_bo = NULL, *old_cursor_bo = NULL;
 	int ret;
 	void *user_ptr;
 	int size = 64*64*4;
@@ -684,7 +684,7 @@ static void qxl_cursor_atomic_update(struct drm_plane *plane,
 							   cursor_bo, 0);
 		cmd->type = QXL_CURSOR_SET;
 
-		qxl_bo_unref(&qcrtc->cursor_bo);
+		old_cursor_bo = qcrtc->cursor_bo;
 		qcrtc->cursor_bo = cursor_bo;
 		cursor_bo = NULL;
 	} else {
@@ -704,6 +704,9 @@ static void qxl_cursor_atomic_update(struct drm_plane *plane,
 	qxl_push_cursor_ring_release(qdev, release, QXL_CMD_CURSOR, false);
 	qxl_release_fence_buffer_objects(release);
 
+	if (old_cursor_bo)
+		qxl_bo_unref(&old_cursor_bo);
+
 	qxl_bo_unref(&cursor_bo);
 
 	return;
diff --git a/drivers/gpu/drm/radeon/radeon_connectors.c b/drivers/gpu/drm/radeon/radeon_connectors.c
index 424cd1b665759..337d3a1c2a409 100644
--- a/drivers/gpu/drm/radeon/radeon_connectors.c
+++ b/drivers/gpu/drm/radeon/radeon_connectors.c
@@ -853,7 +853,7 @@ static int radeon_lvds_get_modes(struct drm_connector *connector)
 	return ret;
 }
 
-static int radeon_lvds_mode_valid(struct drm_connector *connector,
+static enum drm_mode_status radeon_lvds_mode_valid(struct drm_connector *connector,
 				  struct drm_display_mode *mode)
 {
 	struct drm_encoder *encoder = radeon_best_single_encoder(connector);
@@ -1013,7 +1013,7 @@ static int radeon_vga_get_modes(struct drm_connector *connector)
 	return ret;
 }
 
-static int radeon_vga_mode_valid(struct drm_connector *connector,
+static enum drm_mode_status radeon_vga_mode_valid(struct drm_connector *connector,
 				  struct drm_display_mode *mode)
 {
 	struct drm_device *dev = connector->dev;
@@ -1157,7 +1157,7 @@ static int radeon_tv_get_modes(struct drm_connector *connector)
 	return 1;
 }
 
-static int radeon_tv_mode_valid(struct drm_connector *connector,
+static enum drm_mode_status radeon_tv_mode_valid(struct drm_connector *connector,
 				struct drm_display_mode *mode)
 {
 	if ((mode->hdisplay > 1024) || (mode->vdisplay > 768))
@@ -1499,7 +1499,7 @@ static void radeon_dvi_force(struct drm_connector *connector)
 		radeon_connector->use_digital = true;
 }
 
-static int radeon_dvi_mode_valid(struct drm_connector *connector,
+static enum drm_mode_status radeon_dvi_mode_valid(struct drm_connector *connector,
 				  struct drm_display_mode *mode)
 {
 	struct drm_device *dev = connector->dev;
@@ -1801,7 +1801,7 @@ out:
 	return ret;
 }
 
-static int radeon_dp_mode_valid(struct drm_connector *connector,
+static enum drm_mode_status radeon_dp_mode_valid(struct drm_connector *connector,
 				  struct drm_display_mode *mode)
 {
 	struct drm_device *dev = connector->dev;
diff --git a/drivers/gpu/drm/tegra/drm.c b/drivers/gpu/drm/tegra/drm.c
index 0598b4c18c253..75b1c8c03ce95 100644
--- a/drivers/gpu/drm/tegra/drm.c
+++ b/drivers/gpu/drm/tegra/drm.c
@@ -470,7 +470,7 @@ int tegra_drm_submit(struct tegra_drm_context *context,
 		 * unaligned offset is malformed and cause commands stream
 		 * corruption on the buffer address relocation.
 		 */
-		if (offset & 3 || offset >= obj->gem.size) {
+		if (offset & 3 || offset > obj->gem.size) {
 			err = -EINVAL;
 			goto fail;
 		}
diff --git a/drivers/gpu/drm/udl/udl_drv.h b/drivers/gpu/drm/udl/udl_drv.h
index 2a75ab80527a9..2c149b841cf1e 100644
--- a/drivers/gpu/drm/udl/udl_drv.h
+++ b/drivers/gpu/drm/udl/udl_drv.h
@@ -110,7 +110,7 @@ udl_fb_user_fb_create(struct drm_device *dev,
 		      struct drm_file *file,
 		      const struct drm_mode_fb_cmd2 *mode_cmd);
 
-int udl_render_hline(struct drm_device *dev, int bpp, struct urb **urb_ptr,
+int udl_render_hline(struct drm_device *dev, int log_bpp, struct urb **urb_ptr,
 		     const char *front, char **urb_buf_ptr,
 		     u32 byte_offset, u32 device_byte_offset, u32 byte_width,
 		     int *ident_ptr, int *sent_ptr);
diff --git a/drivers/gpu/drm/udl/udl_fb.c b/drivers/gpu/drm/udl/udl_fb.c
index 2ebdc6d5a76e6..8746eeeec44d9 100644
--- a/drivers/gpu/drm/udl/udl_fb.c
+++ b/drivers/gpu/drm/udl/udl_fb.c
@@ -90,7 +90,10 @@ int udl_handle_damage(struct udl_framebuffer *fb, int x, int y,
 	int bytes_identical = 0;
 	struct urb *urb;
 	int aligned_x;
-	int bpp = fb->base.format->cpp[0];
+	int log_bpp;
+
+	BUG_ON(!is_power_of_2(fb->base.format->cpp[0]));
+	log_bpp = __ffs(fb->base.format->cpp[0]);
 
 	if (!fb->active_16)
 		return 0;
@@ -125,19 +128,22 @@ int udl_handle_damage(struct udl_framebuffer *fb, int x, int y,
 
 	for (i = y; i < y + height ; i++) {
 		const int line_offset = fb->base.pitches[0] * i;
-		const int byte_offset = line_offset + (x * bpp);
-		const int dev_byte_offset = (fb->base.width * bpp * i) + (x * bpp);
-		if (udl_render_hline(dev, bpp, &urb,
+		const int byte_offset = line_offset + (x << log_bpp);
+		const int dev_byte_offset = (fb->base.width * i + x) << log_bpp;
+		if (udl_render_hline(dev, log_bpp, &urb,
 				     (char *) fb->obj->vmapping,
 				     &cmd, byte_offset, dev_byte_offset,
-				     width * bpp,
+				     width << log_bpp,
 				     &bytes_identical, &bytes_sent))
 			goto error;
 	}
 
 	if (cmd > (char *) urb->transfer_buffer) {
 		/* Send partial buffer remaining before exiting */
-		int len = cmd - (char *) urb->transfer_buffer;
+		int len;
+		if (cmd < (char *) urb->transfer_buffer + urb->transfer_buffer_length)
+			*cmd++ = 0xAF;
+		len = cmd - (char *) urb->transfer_buffer;
 		ret = udl_submit_urb(dev, urb, len);
 		bytes_sent += len;
 	} else
@@ -146,7 +152,7 @@ int udl_handle_damage(struct udl_framebuffer *fb, int x, int y,
 error:
 	atomic_add(bytes_sent, &udl->bytes_sent);
 	atomic_add(bytes_identical, &udl->bytes_identical);
-	atomic_add(width*height*bpp, &udl->bytes_rendered);
+	atomic_add((width * height) << log_bpp, &udl->bytes_rendered);
 	end_cycles = get_cycles();
 	atomic_add(((unsigned int) ((end_cycles - start_cycles)
 		    >> 10)), /* Kcycles */
@@ -218,7 +224,7 @@ static int udl_fb_open(struct fb_info *info, int user)
 
 		struct fb_deferred_io *fbdefio;
 
-		fbdefio = kmalloc(sizeof(struct fb_deferred_io), GFP_KERNEL);
+		fbdefio = kzalloc(sizeof(struct fb_deferred_io), GFP_KERNEL);
 
 		if (fbdefio) {
 			fbdefio->delay = DL_DEFIO_WRITE_DELAY;
diff --git a/drivers/gpu/drm/udl/udl_main.c b/drivers/gpu/drm/udl/udl_main.c
index 0328b2c7b210a..f8ea3c99b5232 100644
--- a/drivers/gpu/drm/udl/udl_main.c
+++ b/drivers/gpu/drm/udl/udl_main.c
@@ -169,18 +169,13 @@ static void udl_free_urb_list(struct drm_device *dev)
 	struct list_head *node;
 	struct urb_node *unode;
 	struct urb *urb;
-	int ret;
 	unsigned long flags;
 
 	DRM_DEBUG("Waiting for completes and freeing all render urbs\n");
 
 	/* keep waiting and freeing, until we've got 'em all */
 	while (count--) {
-
-		/* Getting interrupted means a leak, but ok at shutdown*/
-		ret = down_interruptible(&udl->urbs.limit_sem);
-		if (ret)
-			break;
+		down(&udl->urbs.limit_sem);
 
 		spin_lock_irqsave(&udl->urbs.lock, flags);
 
@@ -204,17 +199,22 @@ static void udl_free_urb_list(struct drm_device *dev)
 static int udl_alloc_urb_list(struct drm_device *dev, int count, size_t size)
 {
 	struct udl_device *udl = dev->dev_private;
-	int i = 0;
 	struct urb *urb;
 	struct urb_node *unode;
 	char *buf;
+	size_t wanted_size = count * size;
 
 	spin_lock_init(&udl->urbs.lock);
 
+retry:
 	udl->urbs.size = size;
 	INIT_LIST_HEAD(&udl->urbs.list);
 
-	while (i < count) {
+	sema_init(&udl->urbs.limit_sem, 0);
+	udl->urbs.count = 0;
+	udl->urbs.available = 0;
+
+	while (udl->urbs.count * size < wanted_size) {
 		unode = kzalloc(sizeof(struct urb_node), GFP_KERNEL);
 		if (!unode)
 			break;
@@ -230,11 +230,16 @@ static int udl_alloc_urb_list(struct drm_device *dev, int count, size_t size)
 		}
 		unode->urb = urb;
 
-		buf = usb_alloc_coherent(udl->udev, MAX_TRANSFER, GFP_KERNEL,
+		buf = usb_alloc_coherent(udl->udev, size, GFP_KERNEL,
 					 &urb->transfer_dma);
 		if (!buf) {
 			kfree(unode);
 			usb_free_urb(urb);
+			if (size > PAGE_SIZE) {
+				size /= 2;
+				udl_free_urb_list(dev);
+				goto retry;
+			}
 			break;
 		}
 
@@ -245,16 +250,14 @@ static int udl_alloc_urb_list(struct drm_device *dev, int count, size_t size)
 
 		list_add_tail(&unode->entry, &udl->urbs.list);
 
-		i++;
+		up(&udl->urbs.limit_sem);
+		udl->urbs.count++;
+		udl->urbs.available++;
 	}
 
-	sema_init(&udl->urbs.limit_sem, i);
-	udl->urbs.count = i;
-	udl->urbs.available = i;
-
-	DRM_DEBUG("allocated %d %d byte urbs\n", i, (int) size);
+	DRM_DEBUG("allocated %d %d byte urbs\n", udl->urbs.count, (int) size);
 
-	return i;
+	return udl->urbs.count;
 }
 
 struct urb *udl_get_urb(struct drm_device *dev)
diff --git a/drivers/gpu/drm/udl/udl_transfer.c b/drivers/gpu/drm/udl/udl_transfer.c
index 0c87b1ac6b68f..f3331d33547a1 100644
--- a/drivers/gpu/drm/udl/udl_transfer.c
+++ b/drivers/gpu/drm/udl/udl_transfer.c
@@ -83,12 +83,12 @@ static inline u16 pixel32_to_be16(const uint32_t pixel)
 		((pixel >> 8) & 0xf800));
 }
 
-static inline u16 get_pixel_val16(const uint8_t *pixel, int bpp)
+static inline u16 get_pixel_val16(const uint8_t *pixel, int log_bpp)
 {
-	u16 pixel_val16 = 0;
-	if (bpp == 2)
+	u16 pixel_val16;
+	if (log_bpp == 1)
 		pixel_val16 = *(const uint16_t *)pixel;
-	else if (bpp == 4)
+	else
 		pixel_val16 = pixel32_to_be16(*(const uint32_t *)pixel);
 	return pixel_val16;
 }
@@ -125,8 +125,9 @@ static void udl_compress_hline16(
 	const u8 *const pixel_end,
 	uint32_t *device_address_ptr,
 	uint8_t **command_buffer_ptr,
-	const uint8_t *const cmd_buffer_end, int bpp)
+	const uint8_t *const cmd_buffer_end, int log_bpp)
 {
+	const int bpp = 1 << log_bpp;
 	const u8 *pixel = *pixel_start_ptr;
 	uint32_t dev_addr  = *device_address_ptr;
 	uint8_t *cmd = *command_buffer_ptr;
@@ -153,12 +154,12 @@ static void udl_compress_hline16(
 		raw_pixels_count_byte = cmd++; /*  we'll know this later */
 		raw_pixel_start = pixel;
 
-		cmd_pixel_end = pixel + (min(MAX_CMD_PIXELS + 1,
-			min((int)(pixel_end - pixel) / bpp,
-			    (int)(cmd_buffer_end - cmd) / 2))) * bpp;
+		cmd_pixel_end = pixel + (min3(MAX_CMD_PIXELS + 1UL,
+					(unsigned long)(pixel_end - pixel) >> log_bpp,
+					(unsigned long)(cmd_buffer_end - 1 - cmd) / 2) << log_bpp);
 
-		prefetch_range((void *) pixel, (cmd_pixel_end - pixel) * bpp);
-		pixel_val16 = get_pixel_val16(pixel, bpp);
+		prefetch_range((void *) pixel, cmd_pixel_end - pixel);
+		pixel_val16 = get_pixel_val16(pixel, log_bpp);
 
 		while (pixel < cmd_pixel_end) {
 			const u8 *const start = pixel;
@@ -170,7 +171,7 @@ static void udl_compress_hline16(
 			pixel += bpp;
 
 			while (pixel < cmd_pixel_end) {
-				pixel_val16 = get_pixel_val16(pixel, bpp);
+				pixel_val16 = get_pixel_val16(pixel, log_bpp);
 				if (pixel_val16 != repeating_pixel_val16)
 					break;
 				pixel += bpp;
@@ -179,10 +180,10 @@ static void udl_compress_hline16(
 			if (unlikely(pixel > start + bpp)) {
 				/* go back and fill in raw pixel count */
 				*raw_pixels_count_byte = (((start -
-						raw_pixel_start) / bpp) + 1) & 0xFF;
+						raw_pixel_start) >> log_bpp) + 1) & 0xFF;
 
 				/* immediately after raw data is repeat byte */
-				*cmd++ = (((pixel - start) / bpp) - 1) & 0xFF;
+				*cmd++ = (((pixel - start) >> log_bpp) - 1) & 0xFF;
 
 				/* Then start another raw pixel span */
 				raw_pixel_start = pixel;
@@ -192,11 +193,14 @@ static void udl_compress_hline16(
 
 		if (pixel > raw_pixel_start) {
 			/* finalize last RAW span */
-			*raw_pixels_count_byte = ((pixel-raw_pixel_start) / bpp) & 0xFF;
+			*raw_pixels_count_byte = ((pixel - raw_pixel_start) >> log_bpp) & 0xFF;
+		} else {
+			/* undo unused byte */
+			cmd--;
 		}
 
-		*cmd_pixels_count_byte = ((pixel - cmd_pixel_start) / bpp) & 0xFF;
-		dev_addr += ((pixel - cmd_pixel_start) / bpp) * 2;
+		*cmd_pixels_count_byte = ((pixel - cmd_pixel_start) >> log_bpp) & 0xFF;
+		dev_addr += ((pixel - cmd_pixel_start) >> log_bpp) * 2;
 	}
 
 	if (cmd_buffer_end <= MIN_RLX_CMD_BYTES + cmd) {
@@ -219,19 +223,19 @@ static void udl_compress_hline16(
  * (that we can only write to, slowly, and can never read), and (optionally)
  * our shadow copy that tracks what's been sent to that hardware buffer.
  */
-int udl_render_hline(struct drm_device *dev, int bpp, struct urb **urb_ptr,
+int udl_render_hline(struct drm_device *dev, int log_bpp, struct urb **urb_ptr,
 		     const char *front, char **urb_buf_ptr,
 		     u32 byte_offset, u32 device_byte_offset,
 		     u32 byte_width,
 		     int *ident_ptr, int *sent_ptr)
 {
 	const u8 *line_start, *line_end, *next_pixel;
-	u32 base16 = 0 + (device_byte_offset / bpp) * 2;
+	u32 base16 = 0 + (device_byte_offset >> log_bpp) * 2;
 	struct urb *urb = *urb_ptr;
 	u8 *cmd = *urb_buf_ptr;
 	u8 *cmd_end = (u8 *) urb->transfer_buffer + urb->transfer_buffer_length;
 
-	BUG_ON(!(bpp == 2 || bpp == 4));
+	BUG_ON(!(log_bpp == 1 || log_bpp == 2));
 
 	line_start = (u8 *) (front + byte_offset);
 	next_pixel = line_start;
@@ -241,7 +245,7 @@ int udl_render_hline(struct drm_device *dev, int bpp, struct urb **urb_ptr,
 
 		udl_compress_hline16(&next_pixel,
 			     line_end, &base16,
-			     (u8 **) &cmd, (u8 *) cmd_end, bpp);
+			     (u8 **) &cmd, (u8 *) cmd_end, log_bpp);
 
 		if (cmd >= cmd_end) {
 			int len = cmd - (u8 *) urb->transfer_buffer;
diff --git a/drivers/gpu/drm/vc4/vc4_plane.c b/drivers/gpu/drm/vc4/vc4_plane.c
index 77c56264c05bb..17590cb2b80d9 100644
--- a/drivers/gpu/drm/vc4/vc4_plane.c
+++ b/drivers/gpu/drm/vc4/vc4_plane.c
@@ -352,6 +352,9 @@ static int vc4_plane_setup_clipping_and_scaling(struct drm_plane_state *state)
 			vc4_state->x_scaling[0] = VC4_SCALING_TPZ;
 		if (vc4_state->y_scaling[0] == VC4_SCALING_NONE)
 			vc4_state->y_scaling[0] = VC4_SCALING_TPZ;
+	} else {
+		vc4_state->x_scaling[1] = VC4_SCALING_NONE;
+		vc4_state->y_scaling[1] = VC4_SCALING_NONE;
 	}
 
 	vc4_state->is_unity = (vc4_state->x_scaling[0] == VC4_SCALING_NONE &&
diff --git a/drivers/gpu/host1x/job.c b/drivers/gpu/host1x/job.c
index db509ab8874e5..acd99783bbca1 100644
--- a/drivers/gpu/host1x/job.c
+++ b/drivers/gpu/host1x/job.c
@@ -686,7 +686,8 @@ void host1x_job_unpin(struct host1x_job *job)
 	for (i = 0; i < job->num_unpins; i++) {
 		struct host1x_job_unpin_data *unpin = &job->unpins[i];
 
-		if (!IS_ENABLED(CONFIG_TEGRA_HOST1X_FIREWALL) && host->domain) {
+		if (!IS_ENABLED(CONFIG_TEGRA_HOST1X_FIREWALL) &&
+		    unpin->size && host->domain) {
 			iommu_unmap(host->domain, job->addr_phys[i],
 				    unpin->size);
 			free_iova(&host->iova,