Merge tag 'drm-fixes-2018-07-16-1' of git://anongit.freedesktop.org/drm/drm

Pull drm fixes from Dave Airlie:

 - two AGP fixes in here

 - a bunch of mostly amdgpu fixes

 - sun4i build fix

 - two armada fixes

 - some tegra fixes

 - one i915 core and one i915 gvt fix

* tag 'drm-fixes-2018-07-16-1' of git://anongit.freedesktop.org/drm/drm:
  drm/amdgpu/pp/smu7: use a local variable for toc indexing
  amd/dc/dce100: On dce100, set clocks to 0 on suspend
  drm/amd/display: Convert 10kHz clks from PPLib into kHz for Vega
  drm/amdgpu: Verify root PD is mapped into kernel address space (v4)
  drm/amd/display: fix invalid function table override
  drm/amdgpu: Reserve VM root shared fence slot for command submission (v3)
  Revert "drm/amd/display: Don't return ddc result and read_bytes in same return value"
  char: amd64-agp: Use 64-bit arithmetic instead of 32-bit
  char: agp: Change return type to vm_fault_t
  drm/i915: Fix hotplug irq ack on i965/g4x
  drm/armada: fix irq handling
  drm/armada: fix colorkey mode property
  drm/tegra: Fix comparison operator for buffer size
  gpu: host1x: Check whether size of unpin isn't 0
  gpu: host1x: Skip IOMMU initialization if firewall is enabled
  drm/sun4i: link in front-end code if needed
  drm/i915/gvt: update vreg on inhibit context lri command

24 files changed, 233 insertions, 60 deletions

diff --git a/drivers/char/agp/alpha-agp.c b/drivers/char/agp/alpha-agp.c
index 53fe633df1e8..c9bf2c219841 100644
--- a/drivers/char/agp/alpha-agp.c
+++ b/drivers/char/agp/alpha-agp.c
@@ -11,7 +11,7 @@
 
 #include "agp.h"
 
-static int alpha_core_agp_vm_fault(struct vm_fault *vmf)
+static vm_fault_t alpha_core_agp_vm_fault(struct vm_fault *vmf)
 {
 	alpha_agp_info *agp = agp_bridge->dev_private_data;
 	dma_addr_t dma_addr;
diff --git a/drivers/char/agp/amd64-agp.c b/drivers/char/agp/amd64-agp.c
index e50c29c97ca7..c69e39fdd02b 100644
--- a/drivers/char/agp/amd64-agp.c
+++ b/drivers/char/agp/amd64-agp.c
@@ -156,7 +156,7 @@ static u64 amd64_configure(struct pci_dev *hammer, u64 gatt_table)
 
 	/* Address to map to */
 	pci_read_config_dword(hammer, AMD64_GARTAPERTUREBASE, &tmp);
-	aperturebase = tmp << 25;
+	aperturebase = (u64)tmp << 25;
 	aper_base = (aperturebase & PCI_BASE_ADDRESS_MEM_MASK);
 
 	enable_gart_translation(hammer, gatt_table);
@@ -277,7 +277,7 @@ static int fix_northbridge(struct pci_dev *nb, struct pci_dev *agp, u16 cap)
 	pci_read_config_dword(nb, AMD64_GARTAPERTURECTL, &nb_order);
 	nb_order = (nb_order >> 1) & 7;
 	pci_read_config_dword(nb, AMD64_GARTAPERTUREBASE, &nb_base);
-	nb_aper = nb_base << 25;
+	nb_aper = (u64)nb_base << 25;
 
 	/* Northbridge seems to contain crap. Try the AGP bridge. */
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
index 82312a7bc6ad..9c85a90be293 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
@@ -927,6 +927,10 @@ static int amdgpu_cs_ib_vm_chunk(struct amdgpu_device *adev,
 		r = amdgpu_bo_vm_update_pte(p);
 		if (r)
 			return r;
+
+		r = reservation_object_reserve_shared(vm->root.base.bo->tbo.resv);
+		if (r)
+			return r;
 	}
 
 	return amdgpu_cs_sync_rings(p);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
index edf16b2b957a..fdcb498f6d19 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
@@ -107,6 +107,9 @@ static void amdgpu_vm_bo_base_init(struct amdgpu_vm_bo_base *base,
 		return;
 	list_add_tail(&base->bo_list, &bo->va);
 
+	if (bo->tbo.type == ttm_bo_type_kernel)
+		list_move(&base->vm_status, &vm->relocated);
+
 	if (bo->tbo.resv != vm->root.base.bo->tbo.resv)
 		return;
 
@@ -468,7 +471,6 @@ static int amdgpu_vm_alloc_levels(struct amdgpu_device *adev,
 			pt->parent = amdgpu_bo_ref(parent->base.bo);
 
 			amdgpu_vm_bo_base_init(&entry->base, vm, pt);
-			list_move(&entry->base.vm_status, &vm->relocated);
 		}
 
 		if (level < AMDGPU_VM_PTB) {
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
index 4304d9e408b8..ace9ad578ca0 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
@@ -83,22 +83,21 @@ static ssize_t dm_dp_aux_transfer(struct drm_dp_aux *aux,
 	enum i2c_mot_mode mot = (msg->request & DP_AUX_I2C_MOT) ?
 		I2C_MOT_TRUE : I2C_MOT_FALSE;
 	enum ddc_result res;
-	uint32_t read_bytes = msg->size;
+	ssize_t read_bytes;
 
 	if (WARN_ON(msg->size > 16))
 		return -E2BIG;
 
 	switch (msg->request & ~DP_AUX_I2C_MOT) {
 	case DP_AUX_NATIVE_READ:
-		res = dal_ddc_service_read_dpcd_data(
+		read_bytes = dal_ddc_service_read_dpcd_data(
 				TO_DM_AUX(aux)->ddc_service,
 				false,
 				I2C_MOT_UNDEF,
 				msg->address,
 				msg->buffer,
-				msg->size,
-				&read_bytes);
-		break;
+				msg->size);
+		return read_bytes;
 	case DP_AUX_NATIVE_WRITE:
 		res = dal_ddc_service_write_dpcd_data(
 				TO_DM_AUX(aux)->ddc_service,
@@ -109,15 +108,14 @@ static ssize_t dm_dp_aux_transfer(struct drm_dp_aux *aux,
 				msg->size);
 		break;
 	case DP_AUX_I2C_READ:
-		res = dal_ddc_service_read_dpcd_data(
+		read_bytes = dal_ddc_service_read_dpcd_data(
 				TO_DM_AUX(aux)->ddc_service,
 				true,
 				mot,
 				msg->address,
 				msg->buffer,
-				msg->size,
-				&read_bytes);
-		break;
+				msg->size);
+		return read_bytes;
 	case DP_AUX_I2C_WRITE:
 		res = dal_ddc_service_write_dpcd_data(
 				TO_DM_AUX(aux)->ddc_service,
@@ -139,9 +137,7 @@ static ssize_t dm_dp_aux_transfer(struct drm_dp_aux *aux,
 		 r == DDC_RESULT_SUCESSFULL);
 #endif
 
-	if (res != DDC_RESULT_SUCESSFULL)
-		return -EIO;
-	return read_bytes;
+	return msg->size;
 }
 
 static enum drm_connector_status
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_services.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_services.c
index 5a3346124a01..5a2e952c5bea 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_services.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_services.c
@@ -255,8 +255,9 @@ static void pp_to_dc_clock_levels_with_latency(
 			DC_DECODE_PP_CLOCK_TYPE(dc_clk_type));
 
 	for (i = 0; i < clk_level_info->num_levels; i++) {
-		DRM_DEBUG("DM_PPLIB:\t %d\n", pp_clks->data[i].clocks_in_khz);
-		clk_level_info->data[i].clocks_in_khz = pp_clks->data[i].clocks_in_khz;
+		DRM_DEBUG("DM_PPLIB:\t %d in 10kHz\n", pp_clks->data[i].clocks_in_khz);
+		/* translate 10kHz to kHz */
+		clk_level_info->data[i].clocks_in_khz = pp_clks->data[i].clocks_in_khz * 10;
 		clk_level_info->data[i].latency_in_us = pp_clks->data[i].latency_in_us;
 	}
 }
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link_ddc.c b/drivers/gpu/drm/amd/display/dc/core/dc_link_ddc.c
index ae48d603ebd6..49c2face1e7a 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_link_ddc.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_link_ddc.c
@@ -629,14 +629,13 @@ bool dal_ddc_service_query_ddc_data(
 	return ret;
 }
 
-enum ddc_result dal_ddc_service_read_dpcd_data(
+ssize_t dal_ddc_service_read_dpcd_data(
 	struct ddc_service *ddc,
 	bool i2c,
 	enum i2c_mot_mode mot,
 	uint32_t address,
 	uint8_t *data,
-	uint32_t len,
-	uint32_t *read)
+	uint32_t len)
 {
 	struct aux_payload read_payload = {
 		.i2c_over_aux = i2c,
@@ -653,8 +652,6 @@ enum ddc_result dal_ddc_service_read_dpcd_data(
 		.mot = mot
 	};
 
-	*read = 0;
-
 	if (len > DEFAULT_AUX_MAX_DATA_SIZE) {
 		BREAK_TO_DEBUGGER();
 		return DDC_RESULT_FAILED_INVALID_OPERATION;
@@ -664,8 +661,7 @@ enum ddc_result dal_ddc_service_read_dpcd_data(
 		ddc->ctx->i2caux,
 		ddc->ddc_pin,
 		&command)) {
-		*read = command.payloads->length;
-		return DDC_RESULT_SUCESSFULL;
+		return (ssize_t)command.payloads->length;
 	}
 
 	return DDC_RESULT_FAILED_OPERATION;
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_mem_input.c b/drivers/gpu/drm/amd/display/dc/dce/dce_mem_input.c
index b235a75355b8..bae752332a9f 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_mem_input.c
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_mem_input.c
@@ -741,6 +741,29 @@ static struct mem_input_funcs dce_mi_funcs = {
 	.mem_input_is_flip_pending = dce_mi_is_flip_pending
 };
 
+static struct mem_input_funcs dce112_mi_funcs = {
+	.mem_input_program_display_marks = dce112_mi_program_display_marks,
+	.allocate_mem_input = dce_mi_allocate_dmif,
+	.free_mem_input = dce_mi_free_dmif,
+	.mem_input_program_surface_flip_and_addr =
+			dce_mi_program_surface_flip_and_addr,
+	.mem_input_program_pte_vm = dce_mi_program_pte_vm,
+	.mem_input_program_surface_config =
+			dce_mi_program_surface_config,
+	.mem_input_is_flip_pending = dce_mi_is_flip_pending
+};
+
+static struct mem_input_funcs dce120_mi_funcs = {
+	.mem_input_program_display_marks = dce120_mi_program_display_marks,
+	.allocate_mem_input = dce_mi_allocate_dmif,
+	.free_mem_input = dce_mi_free_dmif,
+	.mem_input_program_surface_flip_and_addr =
+			dce_mi_program_surface_flip_and_addr,
+	.mem_input_program_pte_vm = dce_mi_program_pte_vm,
+	.mem_input_program_surface_config =
+			dce_mi_program_surface_config,
+	.mem_input_is_flip_pending = dce_mi_is_flip_pending
+};
 
 void dce_mem_input_construct(
 	struct dce_mem_input *dce_mi,
@@ -769,7 +792,7 @@ void dce112_mem_input_construct(
 	const struct dce_mem_input_mask *mi_mask)
 {
 	dce_mem_input_construct(dce_mi, ctx, inst, regs, mi_shift, mi_mask);
-	dce_mi->base.funcs->mem_input_program_display_marks = dce112_mi_program_display_marks;
+	dce_mi->base.funcs = &dce112_mi_funcs;
 }
 
 void dce120_mem_input_construct(
@@ -781,5 +804,5 @@ void dce120_mem_input_construct(
 	const struct dce_mem_input_mask *mi_mask)
 {
 	dce_mem_input_construct(dce_mi, ctx, inst, regs, mi_shift, mi_mask);
-	dce_mi->base.funcs->mem_input_program_display_marks = dce120_mi_program_display_marks;
+	dce_mi->base.funcs = &dce120_mi_funcs;
 }
diff --git a/drivers/gpu/drm/amd/display/dc/dce100/dce100_resource.c b/drivers/gpu/drm/amd/display/dc/dce100/dce100_resource.c
index 38ec0d609297..344dd2e69e7c 100644
--- a/drivers/gpu/drm/amd/display/dc/dce100/dce100_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dce100/dce100_resource.c
@@ -678,9 +678,22 @@ bool dce100_validate_bandwidth(
 	struct dc  *dc,
 	struct dc_state *context)
 {
-	/* TODO implement when needed but for now hardcode max value*/
-	context->bw.dce.dispclk_khz = 681000;
-	context->bw.dce.yclk_khz = 250000 * MEMORY_TYPE_MULTIPLIER;
+	int i;
+	bool at_least_one_pipe = false;
+
+	for (i = 0; i < dc->res_pool->pipe_count; i++) {
+		if (context->res_ctx.pipe_ctx[i].stream)
+			at_least_one_pipe = true;
+	}
+
+	if (at_least_one_pipe) {
+		/* TODO implement when needed but for now hardcode max value*/
+		context->bw.dce.dispclk_khz = 681000;
+		context->bw.dce.yclk_khz = 250000 * MEMORY_TYPE_MULTIPLIER;
+	} else {
+		context->bw.dce.dispclk_khz = 0;
+		context->bw.dce.yclk_khz = 0;
+	}
 
 	return true;
 }
diff --git a/drivers/gpu/drm/amd/display/dc/inc/dc_link_ddc.h b/drivers/gpu/drm/amd/display/dc/inc/dc_link_ddc.h
index 30b3a08b91be..090b7a8dd67b 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/dc_link_ddc.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/dc_link_ddc.h
@@ -102,14 +102,13 @@ bool dal_ddc_service_query_ddc_data(
 		uint8_t *read_buf,
 		uint32_t read_size);
 
-enum ddc_result dal_ddc_service_read_dpcd_data(
+ssize_t dal_ddc_service_read_dpcd_data(
 		struct ddc_service *ddc,
 		bool i2c,
 		enum i2c_mot_mode mot,
 		uint32_t address,
 		uint8_t *data,
-		uint32_t len,
-		uint32_t *read);
+		uint32_t len);
 
 enum ddc_result dal_ddc_service_write_dpcd_data(
 		struct ddc_service *ddc,
diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/smu7_smumgr.c b/drivers/gpu/drm/amd/powerplay/smumgr/smu7_smumgr.c
index d644a9bb9078..9f407c48d4f0 100644
--- a/drivers/gpu/drm/amd/powerplay/smumgr/smu7_smumgr.c
+++ b/drivers/gpu/drm/amd/powerplay/smumgr/smu7_smumgr.c
@@ -381,6 +381,7 @@ int smu7_request_smu_load_fw(struct pp_hwmgr *hwmgr)
 	uint32_t fw_to_load;
 	int result = 0;
 	struct SMU_DRAMData_TOC *toc;
+	uint32_t num_entries = 0;
 
 	if (!hwmgr->reload_fw) {
 		pr_info("skip reloading...\n");
@@ -422,41 +423,41 @@ int smu7_request_smu_load_fw(struct pp_hwmgr *hwmgr)
 	}
 
 	toc = (struct SMU_DRAMData_TOC *)smu_data->header;
-	toc->num_entries = 0;
 	toc->structure_version = 1;
 
 	PP_ASSERT_WITH_CODE(0 == smu7_populate_single_firmware_entry(hwmgr,
-				UCODE_ID_RLC_G, &toc->entry[toc->num_entries++]),
+				UCODE_ID_RLC_G, &toc->entry[num_entries++]),
 				"Failed to Get Firmware Entry.", return -EINVAL);
 	PP_ASSERT_WITH_CODE(0 == smu7_populate_single_firmware_entry(hwmgr,
-				UCODE_ID_CP_CE, &toc->entry[toc->num_entries++]),
+				UCODE_ID_CP_CE, &toc->entry[num_entries++]),
 				"Failed to Get Firmware Entry.", return -EINVAL);
 	PP_ASSERT_WITH_CODE(0 == smu7_populate_single_firmware_entry(hwmgr,
-				UCODE_ID_CP_PFP, &toc->entry[toc->num_entries++]),
+				UCODE_ID_CP_PFP, &toc->entry[num_entries++]),
 				"Failed to Get Firmware Entry.", return -EINVAL);
 	PP_ASSERT_WITH_CODE(0 == smu7_populate_single_firmware_entry(hwmgr,
-				UCODE_ID_CP_ME, &toc->entry[toc->num_entries++]),
+				UCODE_ID_CP_ME, &toc->entry[num_entries++]),
 				"Failed to Get Firmware Entry.", return -EINVAL);
 	PP_ASSERT_WITH_CODE(0 == smu7_populate_single_firmware_entry(hwmgr,
-				UCODE_ID_CP_MEC, &toc->entry[toc->num_entries++]),
+				UCODE_ID_CP_MEC, &toc->entry[num_entries++]),
 				"Failed to Get Firmware Entry.", return -EINVAL);
 	PP_ASSERT_WITH_CODE(0 == smu7_populate_single_firmware_entry(hwmgr,
-				UCODE_ID_CP_MEC_JT1, &toc->entry[toc->num_entries++]),
+				UCODE_ID_CP_MEC_JT1, &toc->entry[num_entries++]),
 				"Failed to Get Firmware Entry.", return -EINVAL);
 	PP_ASSERT_WITH_CODE(0 == smu7_populate_single_firmware_entry(hwmgr,
-				UCODE_ID_CP_MEC_JT2, &toc->entry[toc->num_entries++]),
+				UCODE_ID_CP_MEC_JT2, &toc->entry[num_entries++]),
 				"Failed to Get Firmware Entry.", return -EINVAL);
 	PP_ASSERT_WITH_CODE(0 == smu7_populate_single_firmware_entry(hwmgr,
-				UCODE_ID_SDMA0, &toc->entry[toc->num_entries++]),
+				UCODE_ID_SDMA0, &toc->entry[num_entries++]),
 				"Failed to Get Firmware Entry.", return -EINVAL);
 	PP_ASSERT_WITH_CODE(0 == smu7_populate_single_firmware_entry(hwmgr,
-				UCODE_ID_SDMA1, &toc->entry[toc->num_entries++]),
+				UCODE_ID_SDMA1, &toc->entry[num_entries++]),
 				"Failed to Get Firmware Entry.", return -EINVAL);
 	if (!hwmgr->not_vf)
 		PP_ASSERT_WITH_CODE(0 == smu7_populate_single_firmware_entry(hwmgr,
-				UCODE_ID_MEC_STORAGE, &toc->entry[toc->num_entries++]),
+				UCODE_ID_MEC_STORAGE, &toc->entry[num_entries++]),
 				"Failed to Get Firmware Entry.", return -EINVAL);
 
+	toc->num_entries = num_entries;
 	smu7_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_DRV_DRAM_ADDR_HI, upper_32_bits(smu_data->header_buffer.mc_addr));
 	smu7_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_DRV_DRAM_ADDR_LO, lower_32_bits(smu_data->header_buffer.mc_addr));
 
diff --git a/drivers/gpu/drm/armada/armada_crtc.c b/drivers/gpu/drm/armada/armada_crtc.c
index 03eeee11dd5b..42a40daff132 100644
--- a/drivers/gpu/drm/armada/armada_crtc.c
+++ b/drivers/gpu/drm/armada/armada_crtc.c
@@ -519,8 +519,9 @@ 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);
 
@@ -1116,16 +1117,22 @@ 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 = {
@@ -1415,6 +1422,7 @@ 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 27319a8335e2..345dc4d0851e 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 c391955009d6..afa7ded3ae31 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 {
@@ -54,11 +55,13 @@ 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);
 }
 
@@ -321,8 +324,17 @@ 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;
@@ -453,7 +465,9 @@ 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/i915/gvt/cmd_parser.c b/drivers/gpu/drm/i915/gvt/cmd_parser.c
index b51c05d03f14..7f562410f9cf 100644
--- a/drivers/gpu/drm/i915/gvt/cmd_parser.c
+++ b/drivers/gpu/drm/i915/gvt/cmd_parser.c
@@ -862,6 +862,7 @@ static int cmd_reg_handler(struct parser_exec_state *s,
 {
 	struct intel_vgpu *vgpu = s->vgpu;
 	struct intel_gvt *gvt = vgpu->gvt;
+	u32 ctx_sr_ctl;
 
 	if (offset + 4 > gvt->device_info.mmio_size) {
 		gvt_vgpu_err("%s access to (%x) outside of MMIO range\n",
@@ -894,6 +895,28 @@ static int cmd_reg_handler(struct parser_exec_state *s,
 		patch_value(s, cmd_ptr(s, index), VGT_PVINFO_PAGE);
 	}
 
+	/* TODO
+	 * Right now only scan LRI command on KBL and in inhibit context.
+	 * It's good enough to support initializing mmio by lri command in
+	 * vgpu inhibit context on KBL.
+	 */
+	if (IS_KABYLAKE(s->vgpu->gvt->dev_priv) &&
+			intel_gvt_mmio_is_in_ctx(gvt, offset) &&
+			!strncmp(cmd, "lri", 3)) {
+		intel_gvt_hypervisor_read_gpa(s->vgpu,
+			s->workload->ring_context_gpa + 12, &ctx_sr_ctl, 4);
+		/* check inhibit context */
+		if (ctx_sr_ctl & 1) {
+			u32 data = cmd_val(s, index + 1);
+
+			if (intel_gvt_mmio_has_mode_mask(s->vgpu->gvt, offset))
+				intel_vgpu_mask_mmio_write(vgpu,
+							offset, &data, 4);
+			else
+				vgpu_vreg(vgpu, offset) = data;
+		}
+	}
+
 	/* TODO: Update the global mask if this MMIO is a masked-MMIO */
 	intel_gvt_mmio_set_cmd_accessed(gvt, offset);
 	return 0;
diff --git a/drivers/gpu/drm/i915/gvt/gvt.h b/drivers/gpu/drm/i915/gvt/gvt.h
index 05d15a095310..858967daf04b 100644
--- a/drivers/gpu/drm/i915/gvt/gvt.h
+++ b/drivers/gpu/drm/i915/gvt/gvt.h
@@ -268,6 +268,8 @@ struct intel_gvt_mmio {
 #define F_CMD_ACCESSED	(1 << 5)
 /* This reg could be accessed by unaligned address */
 #define F_UNALIGN	(1 << 6)
+/* This reg is saved/restored in context */
+#define F_IN_CTX	(1 << 7)
 
 	struct gvt_mmio_block *mmio_block;
 	unsigned int num_mmio_block;
@@ -639,6 +641,33 @@ static inline bool intel_gvt_mmio_has_mode_mask(
 	return gvt->mmio.mmio_attribute[offset >> 2] & F_MODE_MASK;
 }
 
+/**
+ * intel_gvt_mmio_is_in_ctx - check if a MMIO has in-ctx mask
+ * @gvt: a GVT device
+ * @offset: register offset
+ *
+ * Returns:
+ * True if a MMIO has a in-context mask, false if it isn't.
+ *
+ */
+static inline bool intel_gvt_mmio_is_in_ctx(
+			struct intel_gvt *gvt, unsigned int offset)
+{
+	return gvt->mmio.mmio_attribute[offset >> 2] & F_IN_CTX;
+}
+
+/**
+ * intel_gvt_mmio_set_in_ctx - mask a MMIO in logical context
+ * @gvt: a GVT device
+ * @offset: register offset
+ *
+ */
+static inline void intel_gvt_mmio_set_in_ctx(
+			struct intel_gvt *gvt, unsigned int offset)
+{
+	gvt->mmio.mmio_attribute[offset >> 2] |= F_IN_CTX;
+}
+
 int intel_gvt_debugfs_add_vgpu(struct intel_vgpu *vgpu);
 void intel_gvt_debugfs_remove_vgpu(struct intel_vgpu *vgpu);
 int intel_gvt_debugfs_init(struct intel_gvt *gvt);
diff --git a/drivers/gpu/drm/i915/gvt/handlers.c b/drivers/gpu/drm/i915/gvt/handlers.c
index bcbc47a88a70..8f1caacdc78a 100644
--- a/drivers/gpu/drm/i915/gvt/handlers.c
+++ b/drivers/gpu/drm/i915/gvt/handlers.c
@@ -3046,6 +3046,30 @@ int intel_vgpu_default_mmio_write(struct intel_vgpu *vgpu, unsigned int offset,
 }
 
 /**
+ * intel_vgpu_mask_mmio_write - write mask register
+ * @vgpu: a vGPU
+ * @offset: access offset
+ * @p_data: write data buffer
+ * @bytes: access data length
+ *
+ * Returns:
+ * Zero on success, negative error code if failed.
+ */
+int intel_vgpu_mask_mmio_write(struct intel_vgpu *vgpu, unsigned int offset,
+		void *p_data, unsigned int bytes)
+{
+	u32 mask, old_vreg;
+
+	old_vreg = vgpu_vreg(vgpu, offset);
+	write_vreg(vgpu, offset, p_data, bytes);
+	mask = vgpu_vreg(vgpu, offset) >> 16;
+	vgpu_vreg(vgpu, offset) = (old_vreg & ~mask) |
+				(vgpu_vreg(vgpu, offset) & mask);
+
+	return 0;
+}
+
+/**
  * intel_gvt_in_force_nonpriv_whitelist - if a mmio is in whitelist to be
  * force-nopriv register
  *
diff --git a/drivers/gpu/drm/i915/gvt/mmio.h b/drivers/gpu/drm/i915/gvt/mmio.h
index 71b620875943..dac8c6401e26 100644
--- a/drivers/gpu/drm/i915/gvt/mmio.h
+++ b/drivers/gpu/drm/i915/gvt/mmio.h
@@ -98,4 +98,6 @@ bool intel_gvt_in_force_nonpriv_whitelist(struct intel_gvt *gvt,
 int intel_vgpu_mmio_reg_rw(struct intel_vgpu *vgpu, unsigned int offset,
 			   void *pdata, unsigned int bytes, bool is_read);
 
+int intel_vgpu_mask_mmio_write(struct intel_vgpu *vgpu, unsigned int offset,
+				  void *p_data, unsigned int bytes);
 #endif
diff --git a/drivers/gpu/drm/i915/gvt/mmio_context.c b/drivers/gpu/drm/i915/gvt/mmio_context.c
index 0f949554d118..5ca9caf7552a 100644
--- a/drivers/gpu/drm/i915/gvt/mmio_context.c
+++ b/drivers/gpu/drm/i915/gvt/mmio_context.c
@@ -581,7 +581,9 @@ void intel_gvt_init_engine_mmio_context(struct intel_gvt *gvt)
 
 	for (mmio = gvt->engine_mmio_list.mmio;
 	     i915_mmio_reg_valid(mmio->reg); mmio++) {
-		if (mmio->in_context)
+		if (mmio->in_context) {
 			gvt->engine_mmio_list.ctx_mmio_count[mmio->ring_id]++;
+			intel_gvt_mmio_set_in_ctx(gvt, mmio->reg.reg);
+		}
 	}
 }
diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 4a02747ac658..c16cb025755e 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -1998,10 +1998,38 @@ 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/sun4i/Makefile b/drivers/gpu/drm/sun4i/Makefile
index 2589f4acd5ae..9c81301d0eed 100644
--- a/drivers/gpu/drm/sun4i/Makefile
+++ b/drivers/gpu/drm/sun4i/Makefile
@@ -32,7 +32,10 @@ obj-$(CONFIG_DRM_SUN4I)		+= sun4i-tcon.o
 obj-$(CONFIG_DRM_SUN4I)		+= sun4i_tv.o
 obj-$(CONFIG_DRM_SUN4I)		+= sun6i_drc.o
 
-obj-$(CONFIG_DRM_SUN4I_BACKEND)	+= sun4i-backend.o sun4i-frontend.o
+obj-$(CONFIG_DRM_SUN4I_BACKEND)	+= sun4i-backend.o
+ifdef CONFIG_DRM_SUN4I_BACKEND
+obj-$(CONFIG_DRM_SUN4I)		+= sun4i-frontend.o
+endif
 obj-$(CONFIG_DRM_SUN4I_HDMI)	+= sun4i-drm-hdmi.o
 obj-$(CONFIG_DRM_SUN6I_DSI)	+= sun6i-dsi.o
 obj-$(CONFIG_DRM_SUN8I_DW_HDMI)	+= sun8i-drm-hdmi.o
diff --git a/drivers/gpu/drm/tegra/drm.c b/drivers/gpu/drm/tegra/drm.c
index 776c1513e582..a2bd5876c633 100644
--- a/drivers/gpu/drm/tegra/drm.c
+++ b/drivers/gpu/drm/tegra/drm.c
@@ -398,7 +398,7 @@ 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/host1x/dev.c b/drivers/gpu/host1x/dev.c
index f1d5f76e9c33..d88073e7d22d 100644
--- a/drivers/gpu/host1x/dev.c
+++ b/drivers/gpu/host1x/dev.c
@@ -218,6 +218,9 @@ static int host1x_probe(struct platform_device *pdev)
 		return err;
 	}
 
+	if (IS_ENABLED(CONFIG_TEGRA_HOST1X_FIREWALL))
+		goto skip_iommu;
+
 	host->group = iommu_group_get(&pdev->dev);
 	if (host->group) {
 		struct iommu_domain_geometry *geometry;
diff --git a/drivers/gpu/host1x/job.c b/drivers/gpu/host1x/job.c
index e2f4a4d93d20..527a1cddb14f 100644
--- a/drivers/gpu/host1x/job.c
+++ b/drivers/gpu/host1x/job.c
@@ -569,7 +569,8 @@ 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,