Merge drm-upstream/drm-next into drm-intel-next-queued

Catch up with upstream while it's easy.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>

398 files changed, 8711 insertions, 6553 deletions

diff --git a/drivers/gpu/drm/Makefile b/drivers/gpu/drm/Makefile
index 24a066e1841c..a8acc197dec3 100644
--- a/drivers/gpu/drm/Makefile
+++ b/drivers/gpu/drm/Makefile
@@ -33,7 +33,7 @@ drm_kms_helper-y := drm_crtc_helper.o drm_dp_helper.o drm_probe_helper.o \
 		drm_plane_helper.o drm_dp_mst_topology.o drm_atomic_helper.o \
 		drm_kms_helper_common.o drm_dp_dual_mode_helper.o \
 		drm_simple_kms_helper.o drm_modeset_helper.o \
-		drm_scdc_helper.o
+		drm_scdc_helper.o drm_gem_framebuffer_helper.o
 
 drm_kms_helper-$(CONFIG_DRM_PANEL_BRIDGE) += bridge/panel.o
 drm_kms_helper-$(CONFIG_DRM_LOAD_EDID_FIRMWARE) += drm_edid_load.o
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index 51d1364cf185..12e71bbfd222 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -96,6 +96,7 @@ extern int amdgpu_bapm;
 extern int amdgpu_deep_color;
 extern int amdgpu_vm_size;
 extern int amdgpu_vm_block_size;
+extern int amdgpu_vm_fragment_size;
 extern int amdgpu_vm_fault_stop;
 extern int amdgpu_vm_debug;
 extern int amdgpu_vm_update_mode;
@@ -373,78 +374,10 @@ struct amdgpu_clock {
 };
 
 /*
- * BO.
+ * GEM.
  */
-struct amdgpu_bo_list_entry {
-	struct amdgpu_bo		*robj;
-	struct ttm_validate_buffer	tv;
-	struct amdgpu_bo_va		*bo_va;
-	uint32_t			priority;
-	struct page			**user_pages;
-	int				user_invalidated;
-};
-
-struct amdgpu_bo_va_mapping {
-	struct list_head		list;
-	struct rb_node			rb;
-	uint64_t			start;
-	uint64_t			last;
-	uint64_t			__subtree_last;
-	uint64_t			offset;
-	uint64_t			flags;
-};
-
-/* bo virtual addresses in a specific vm */
-struct amdgpu_bo_va {
-	/* protected by bo being reserved */
-	struct list_head		bo_list;
-	struct dma_fence	        *last_pt_update;
-	unsigned			ref_count;
-
-	/* protected by vm mutex and spinlock */
-	struct list_head		vm_status;
-
-	/* mappings for this bo_va */
-	struct list_head		invalids;
-	struct list_head		valids;
-
-	/* constant after initialization */
-	struct amdgpu_vm		*vm;
-	struct amdgpu_bo		*bo;
-};
 
 #define AMDGPU_GEM_DOMAIN_MAX		0x3
-
-struct amdgpu_bo {
-	/* Protected by tbo.reserved */
-	u32				prefered_domains;
-	u32				allowed_domains;
-	struct ttm_place		placements[AMDGPU_GEM_DOMAIN_MAX + 1];
-	struct ttm_placement		placement;
-	struct ttm_buffer_object	tbo;
-	struct ttm_bo_kmap_obj		kmap;
-	u64				flags;
-	unsigned			pin_count;
-	void				*kptr;
-	u64				tiling_flags;
-	u64				metadata_flags;
-	void				*metadata;
-	u32				metadata_size;
-	unsigned			prime_shared_count;
-	/* list of all virtual address to which this bo
-	 * is associated to
-	 */
-	struct list_head		va;
-	/* Constant after initialization */
-	struct drm_gem_object		gem_base;
-	struct amdgpu_bo		*parent;
-	struct amdgpu_bo		*shadow;
-
-	struct ttm_bo_kmap_obj		dma_buf_vmap;
-	struct amdgpu_mn		*mn;
-	struct list_head		mn_list;
-	struct list_head		shadow_list;
-};
 #define gem_to_amdgpu_bo(gobj) container_of((gobj), struct amdgpu_bo, gem_base)
 
 void amdgpu_gem_object_free(struct drm_gem_object *obj);
@@ -678,15 +611,15 @@ typedef enum _AMDGPU_DOORBELL64_ASSIGNMENT
 	/* overlap the doorbell assignment with VCN as they are  mutually exclusive
 	 * VCE engine's doorbell is 32 bit and two VCE ring share one QWORD
 	 */
-	AMDGPU_DOORBELL64_RING0_1                 = 0xF8,
-	AMDGPU_DOORBELL64_RING2_3                 = 0xF9,
-	AMDGPU_DOORBELL64_RING4_5                 = 0xFA,
-	AMDGPU_DOORBELL64_RING6_7                 = 0xFB,
+	AMDGPU_DOORBELL64_UVD_RING0_1             = 0xF8,
+	AMDGPU_DOORBELL64_UVD_RING2_3             = 0xF9,
+	AMDGPU_DOORBELL64_UVD_RING4_5             = 0xFA,
+	AMDGPU_DOORBELL64_UVD_RING6_7             = 0xFB,
 
-	AMDGPU_DOORBELL64_UVD_RING0_1             = 0xFC,
-	AMDGPU_DOORBELL64_UVD_RING2_3             = 0xFD,
-	AMDGPU_DOORBELL64_UVD_RING4_5             = 0xFE,
-	AMDGPU_DOORBELL64_UVD_RING6_7             = 0xFF,
+	AMDGPU_DOORBELL64_VCE_RING0_1             = 0xFC,
+	AMDGPU_DOORBELL64_VCE_RING2_3             = 0xFD,
+	AMDGPU_DOORBELL64_VCE_RING4_5             = 0xFE,
+	AMDGPU_DOORBELL64_VCE_RING6_7             = 0xFF,
 
 	AMDGPU_DOORBELL64_MAX_ASSIGNMENT          = 0xFF,
 	AMDGPU_DOORBELL64_INVALID                 = 0xFFFF
@@ -816,6 +749,7 @@ void amdgpu_ctx_mgr_fini(struct amdgpu_ctx_mgr *mgr);
 struct amdgpu_fpriv {
 	struct amdgpu_vm	vm;
 	struct amdgpu_bo_va	*prt_va;
+	struct amdgpu_bo_va	*csa_va;
 	struct mutex		bo_list_lock;
 	struct idr		bo_list_handles;
 	struct amdgpu_ctx_mgr	ctx_mgr;
@@ -825,6 +759,14 @@ struct amdgpu_fpriv {
 /*
  * residency list
  */
+struct amdgpu_bo_list_entry {
+	struct amdgpu_bo		*robj;
+	struct ttm_validate_buffer	tv;
+	struct amdgpu_bo_va		*bo_va;
+	uint32_t			priority;
+	struct page			**user_pages;
+	int				user_invalidated;
+};
 
 struct amdgpu_bo_list {
 	struct mutex lock;
@@ -1191,10 +1133,6 @@ struct amdgpu_wb {
 
 int amdgpu_wb_get(struct amdgpu_device *adev, u32 *wb);
 void amdgpu_wb_free(struct amdgpu_device *adev, u32 wb);
-int amdgpu_wb_get_64bit(struct amdgpu_device *adev, u32 *wb);
-int amdgpu_wb_get_256Bit(struct amdgpu_device *adev, u32 *wb);
-void amdgpu_wb_free_64bit(struct amdgpu_device *adev, u32 wb);
-void amdgpu_wb_free_256bit(struct amdgpu_device *adev, u32 wb);
 
 void amdgpu_get_pcie_info(struct amdgpu_device *adev);
 
@@ -1488,7 +1426,7 @@ struct amdgpu_device {
 	bool				is_atom_fw;
 	uint8_t				*bios;
 	uint32_t			bios_size;
-	struct amdgpu_bo		*stollen_vga_memory;
+	struct amdgpu_bo		*stolen_vga_memory;
 	uint32_t			bios_scratch_reg_offset;
 	uint32_t			bios_scratch[AMDGPU_BIOS_NUM_SCRATCH];
 
@@ -1546,9 +1484,6 @@ struct amdgpu_device {
 	struct amdgpu_mman		mman;
 	struct amdgpu_vram_scratch	vram_scratch;
 	struct amdgpu_wb		wb;
-	atomic64_t			vram_usage;
-	atomic64_t			vram_vis_usage;
-	atomic64_t			gtt_usage;
 	atomic64_t			num_bytes_moved;
 	atomic64_t			num_evictions;
 	atomic64_t			num_vram_cpu_page_faults;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_acp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_acp.c
index 06879d1dcabd..a52795d9b458 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_acp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_acp.c
@@ -285,19 +285,20 @@ static int acp_hw_init(void *handle)
 		return 0;
 	else if (r)
 		return r;
+	if (adev->asic_type != CHIP_STONEY) {
+		adev->acp.acp_genpd = kzalloc(sizeof(struct acp_pm_domain), GFP_KERNEL);
+		if (adev->acp.acp_genpd == NULL)
+			return -ENOMEM;
 
-	adev->acp.acp_genpd = kzalloc(sizeof(struct acp_pm_domain), GFP_KERNEL);
-	if (adev->acp.acp_genpd == NULL)
-		return -ENOMEM;
-
-	adev->acp.acp_genpd->gpd.name = "ACP_AUDIO";
-	adev->acp.acp_genpd->gpd.power_off = acp_poweroff;
-	adev->acp.acp_genpd->gpd.power_on = acp_poweron;
+		adev->acp.acp_genpd->gpd.name = "ACP_AUDIO";
+		adev->acp.acp_genpd->gpd.power_off = acp_poweroff;
+		adev->acp.acp_genpd->gpd.power_on = acp_poweron;
 
 
-	adev->acp.acp_genpd->cgs_dev = adev->acp.cgs_device;
+		adev->acp.acp_genpd->cgs_dev = adev->acp.cgs_device;
 
-	pm_genpd_init(&adev->acp.acp_genpd->gpd, NULL, false);
+		pm_genpd_init(&adev->acp.acp_genpd->gpd, NULL, false);
+	}
 
 	adev->acp.acp_cell = kzalloc(sizeof(struct mfd_cell) * ACP_DEVS,
 							GFP_KERNEL);
@@ -319,14 +320,29 @@ static int acp_hw_init(void *handle)
 		return -ENOMEM;
 	}
 
-	i2s_pdata[0].quirks = DW_I2S_QUIRK_COMP_REG_OFFSET;
+	switch (adev->asic_type) {
+	case CHIP_STONEY:
+		i2s_pdata[0].quirks = DW_I2S_QUIRK_COMP_REG_OFFSET |
+			DW_I2S_QUIRK_16BIT_IDX_OVERRIDE;
+		break;
+	default:
+		i2s_pdata[0].quirks = DW_I2S_QUIRK_COMP_REG_OFFSET;
+	}
 	i2s_pdata[0].cap = DWC_I2S_PLAY;
 	i2s_pdata[0].snd_rates = SNDRV_PCM_RATE_8000_96000;
 	i2s_pdata[0].i2s_reg_comp1 = ACP_I2S_COMP1_PLAY_REG_OFFSET;
 	i2s_pdata[0].i2s_reg_comp2 = ACP_I2S_COMP2_PLAY_REG_OFFSET;
+	switch (adev->asic_type) {
+	case CHIP_STONEY:
+		i2s_pdata[1].quirks = DW_I2S_QUIRK_COMP_REG_OFFSET |
+			DW_I2S_QUIRK_COMP_PARAM1 |
+			DW_I2S_QUIRK_16BIT_IDX_OVERRIDE;
+		break;
+	default:
+		i2s_pdata[1].quirks = DW_I2S_QUIRK_COMP_REG_OFFSET |
+			DW_I2S_QUIRK_COMP_PARAM1;
+	}
 
-	i2s_pdata[1].quirks = DW_I2S_QUIRK_COMP_REG_OFFSET |
-				DW_I2S_QUIRK_COMP_PARAM1;
 	i2s_pdata[1].cap = DWC_I2S_RECORD;
 	i2s_pdata[1].snd_rates = SNDRV_PCM_RATE_8000_96000;
 	i2s_pdata[1].i2s_reg_comp1 = ACP_I2S_COMP1_CAP_REG_OFFSET;
@@ -373,12 +389,14 @@ static int acp_hw_init(void *handle)
 	if (r)
 		return r;
 
-	for (i = 0; i < ACP_DEVS ; i++) {
-		dev = get_mfd_cell_dev(adev->acp.acp_cell[i].name, i);
-		r = pm_genpd_add_device(&adev->acp.acp_genpd->gpd, dev);
-		if (r) {
-			dev_err(dev, "Failed to add dev to genpd\n");
-			return r;
+	if (adev->asic_type != CHIP_STONEY) {
+		for (i = 0; i < ACP_DEVS ; i++) {
+			dev = get_mfd_cell_dev(adev->acp.acp_cell[i].name, i);
+			r = pm_genpd_add_device(&adev->acp.acp_genpd->gpd, dev);
+			if (r) {
+				dev_err(dev, "Failed to add dev to genpd\n");
+				return r;
+			}
 		}
 	}
 
@@ -398,20 +416,22 @@ static int acp_hw_fini(void *handle)
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
 	/* return early if no ACP */
-	if (!adev->acp.acp_genpd)
+	if (!adev->acp.acp_cell)
 		return 0;
 
-	for (i = 0; i < ACP_DEVS ; i++) {
-		dev = get_mfd_cell_dev(adev->acp.acp_cell[i].name, i);
-		ret = pm_genpd_remove_device(&adev->acp.acp_genpd->gpd, dev);
-		/* If removal fails, dont giveup and try rest */
-		if (ret)
-			dev_err(dev, "remove dev from genpd failed\n");
+	if (adev->acp.acp_genpd) {
+		for (i = 0; i < ACP_DEVS ; i++) {
+			dev = get_mfd_cell_dev(adev->acp.acp_cell[i].name, i);
+			ret = pm_genpd_remove_device(&adev->acp.acp_genpd->gpd, dev);
+			/* If removal fails, dont giveup and try rest */
+			if (ret)
+				dev_err(dev, "remove dev from genpd failed\n");
+		}
+		kfree(adev->acp.acp_genpd);
 	}
 
 	mfd_remove_devices(adev->acp.parent);
 	kfree(adev->acp.acp_res);
-	kfree(adev->acp.acp_genpd);
 	kfree(adev->acp.acp_cell);
 
 	return 0;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
index ef79551b4cb7..57afad79f55d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
@@ -30,10 +30,10 @@
 #include <drm/drmP.h>
 #include <drm/drm_crtc_helper.h>
 #include "amdgpu.h"
+#include "amdgpu_pm.h"
 #include "amd_acpi.h"
 #include "atom.h"
 
-extern void amdgpu_pm_acpi_event_handler(struct amdgpu_device *adev);
 /* Call the ATIF method
  */
 /**
@@ -289,7 +289,7 @@ out:
  * handles it.
  * Returns NOTIFY code
  */
-int amdgpu_atif_handler(struct amdgpu_device *adev,
+static int amdgpu_atif_handler(struct amdgpu_device *adev,
 			struct acpi_bus_event *event)
 {
 	struct amdgpu_atif *atif = &adev->atif;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
index 37971d9402e3..5432af39a674 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
@@ -27,16 +27,15 @@
 #include "amdgpu_gfx.h"
 #include <linux/module.h>
 
-const struct kfd2kgd_calls *kfd2kgd;
 const struct kgd2kfd_calls *kgd2kfd;
-bool (*kgd2kfd_init_p)(unsigned, const struct kgd2kfd_calls**);
+bool (*kgd2kfd_init_p)(unsigned int, const struct kgd2kfd_calls**);
 
 int amdgpu_amdkfd_init(void)
 {
 	int ret;
 
 #if defined(CONFIG_HSA_AMD_MODULE)
-	int (*kgd2kfd_init_p)(unsigned, const struct kgd2kfd_calls**);
+	int (*kgd2kfd_init_p)(unsigned int, const struct kgd2kfd_calls**);
 
 	kgd2kfd_init_p = symbol_request(kgd2kfd_init);
 
@@ -61,8 +60,21 @@ int amdgpu_amdkfd_init(void)
 	return ret;
 }
 
-bool amdgpu_amdkfd_load_interface(struct amdgpu_device *adev)
+void amdgpu_amdkfd_fini(void)
+{
+	if (kgd2kfd) {
+		kgd2kfd->exit();
+		symbol_put(kgd2kfd_init);
+	}
+}
+
+void amdgpu_amdkfd_device_probe(struct amdgpu_device *adev)
 {
+	const struct kfd2kgd_calls *kfd2kgd;
+
+	if (!kgd2kfd)
+		return;
+
 	switch (adev->asic_type) {
 #ifdef CONFIG_DRM_AMDGPU_CIK
 	case CHIP_KAVERI:
@@ -73,25 +85,12 @@ bool amdgpu_amdkfd_load_interface(struct amdgpu_device *adev)
 		kfd2kgd = amdgpu_amdkfd_gfx_8_0_get_functions();
 		break;
 	default:
-		return false;
+		dev_info(adev->dev, "kfd not supported on this ASIC\n");
+		return;
 	}
 
-	return true;
-}
-
-void amdgpu_amdkfd_fini(void)
-{
-	if (kgd2kfd) {
-		kgd2kfd->exit();
-		symbol_put(kgd2kfd_init);
-	}
-}
-
-void amdgpu_amdkfd_device_probe(struct amdgpu_device *adev)
-{
-	if (kgd2kfd)
-		adev->kfd = kgd2kfd->probe((struct kgd_dev *)adev,
-					adev->pdev, kfd2kgd);
+	adev->kfd = kgd2kfd->probe((struct kgd_dev *)adev,
+				   adev->pdev, kfd2kgd);
 }
 
 void amdgpu_amdkfd_device_init(struct amdgpu_device *adev)
@@ -184,7 +183,8 @@ int alloc_gtt_mem(struct kgd_dev *kgd, size_t size,
 		return -ENOMEM;
 
 	r = amdgpu_bo_create(adev, size, PAGE_SIZE, true, AMDGPU_GEM_DOMAIN_GTT,
-			     AMDGPU_GEM_CREATE_CPU_GTT_USWC, NULL, NULL, &(*mem)->bo);
+			     AMDGPU_GEM_CREATE_CPU_GTT_USWC, NULL, NULL, 0,
+			     &(*mem)->bo);
 	if (r) {
 		dev_err(adev->dev,
 			"failed to allocate BO for amdkfd (%d)\n", r);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
index 73f83a10ae14..8d689ab7e429 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
@@ -26,6 +26,7 @@
 #define AMDGPU_AMDKFD_H_INCLUDED
 
 #include <linux/types.h>
+#include <linux/mmu_context.h>
 #include <kgd_kfd_interface.h>
 
 struct amdgpu_device;
@@ -39,8 +40,6 @@ struct kgd_mem {
 int amdgpu_amdkfd_init(void);
 void amdgpu_amdkfd_fini(void);
 
-bool amdgpu_amdkfd_load_interface(struct amdgpu_device *adev);
-
 void amdgpu_amdkfd_suspend(struct amdgpu_device *adev);
 int amdgpu_amdkfd_resume(struct amdgpu_device *adev);
 void amdgpu_amdkfd_interrupt(struct amdgpu_device *adev,
@@ -62,4 +61,19 @@ uint64_t get_gpu_clock_counter(struct kgd_dev *kgd);
 
 uint32_t get_max_engine_clock_in_mhz(struct kgd_dev *kgd);
 
+#define read_user_wptr(mmptr, wptr, dst)				\
+	({								\
+		bool valid = false;					\
+		if ((mmptr) && (wptr)) {				\
+			if ((mmptr) == current->mm) {			\
+				valid = !get_user((dst), (wptr));	\
+			} else if (current->mm == NULL) {		\
+				use_mm(mmptr);				\
+				valid = !get_user((dst), (wptr));	\
+				unuse_mm(mmptr);			\
+			}						\
+		}							\
+		valid;							\
+	})
+
 #endif /* AMDGPU_AMDKFD_H_INCLUDED */
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c
index 5254562fd0f9..b9dbbf9cb8b0 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c
@@ -39,6 +39,12 @@
 #include "gmc/gmc_7_1_sh_mask.h"
 #include "cik_structs.h"
 
+enum hqd_dequeue_request_type {
+	NO_ACTION = 0,
+	DRAIN_PIPE,
+	RESET_WAVES
+};
+
 enum {
 	MAX_TRAPID = 8,		/* 3 bits in the bitfield. */
 	MAX_WATCH_ADDRESSES = 4
@@ -96,12 +102,15 @@ static int kgd_init_pipeline(struct kgd_dev *kgd, uint32_t pipe_id,
 				uint32_t hpd_size, uint64_t hpd_gpu_addr);
 static int kgd_init_interrupts(struct kgd_dev *kgd, uint32_t pipe_id);
 static int kgd_hqd_load(struct kgd_dev *kgd, void *mqd, uint32_t pipe_id,
-			uint32_t queue_id, uint32_t __user *wptr);
+			uint32_t queue_id, uint32_t __user *wptr,
+			uint32_t wptr_shift, uint32_t wptr_mask,
+			struct mm_struct *mm);
 static int kgd_hqd_sdma_load(struct kgd_dev *kgd, void *mqd);
 static bool kgd_hqd_is_occupied(struct kgd_dev *kgd, uint64_t queue_address,
 				uint32_t pipe_id, uint32_t queue_id);
 
-static int kgd_hqd_destroy(struct kgd_dev *kgd, uint32_t reset_type,
+static int kgd_hqd_destroy(struct kgd_dev *kgd, void *mqd,
+				enum kfd_preempt_type reset_type,
 				unsigned int utimeout, uint32_t pipe_id,
 				uint32_t queue_id);
 static bool kgd_hqd_sdma_is_occupied(struct kgd_dev *kgd, void *mqd);
@@ -126,6 +135,33 @@ static uint16_t get_atc_vmid_pasid_mapping_pasid(struct kgd_dev *kgd,
 static void write_vmid_invalidate_request(struct kgd_dev *kgd, uint8_t vmid);
 
 static uint16_t get_fw_version(struct kgd_dev *kgd, enum kgd_engine_type type);
+static void set_scratch_backing_va(struct kgd_dev *kgd,
+					uint64_t va, uint32_t vmid);
+
+/* Because of REG_GET_FIELD() being used, we put this function in the
+ * asic specific file.
+ */
+static int get_tile_config(struct kgd_dev *kgd,
+		struct tile_config *config)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)kgd;
+
+	config->gb_addr_config = adev->gfx.config.gb_addr_config;
+	config->num_banks = REG_GET_FIELD(adev->gfx.config.mc_arb_ramcfg,
+				MC_ARB_RAMCFG, NOOFBANK);
+	config->num_ranks = REG_GET_FIELD(adev->gfx.config.mc_arb_ramcfg,
+				MC_ARB_RAMCFG, NOOFRANKS);
+
+	config->tile_config_ptr = adev->gfx.config.tile_mode_array;
+	config->num_tile_configs =
+			ARRAY_SIZE(adev->gfx.config.tile_mode_array);
+	config->macro_tile_config_ptr =
+			adev->gfx.config.macrotile_mode_array;
+	config->num_macro_tile_configs =
+			ARRAY_SIZE(adev->gfx.config.macrotile_mode_array);
+
+	return 0;
+}
 
 static const struct kfd2kgd_calls kfd2kgd = {
 	.init_gtt_mem_allocation = alloc_gtt_mem,
@@ -150,7 +186,9 @@ static const struct kfd2kgd_calls kfd2kgd = {
 	.get_atc_vmid_pasid_mapping_pasid = get_atc_vmid_pasid_mapping_pasid,
 	.get_atc_vmid_pasid_mapping_valid = get_atc_vmid_pasid_mapping_valid,
 	.write_vmid_invalidate_request = write_vmid_invalidate_request,
-	.get_fw_version = get_fw_version
+	.get_fw_version = get_fw_version,
+	.set_scratch_backing_va = set_scratch_backing_va,
+	.get_tile_config = get_tile_config,
 };
 
 struct kfd2kgd_calls *amdgpu_amdkfd_gfx_7_get_functions(void)
@@ -186,7 +224,7 @@ static void acquire_queue(struct kgd_dev *kgd, uint32_t pipe_id,
 {
 	struct amdgpu_device *adev = get_amdgpu_device(kgd);
 
-	uint32_t mec = (++pipe_id / adev->gfx.mec.num_pipe_per_mec) + 1;
+	uint32_t mec = (pipe_id / adev->gfx.mec.num_pipe_per_mec) + 1;
 	uint32_t pipe = (pipe_id % adev->gfx.mec.num_pipe_per_mec);
 
 	lock_srbm(kgd, mec, pipe, queue_id, 0);
@@ -290,20 +328,38 @@ static inline struct cik_sdma_rlc_registers *get_sdma_mqd(void *mqd)
 }
 
 static int kgd_hqd_load(struct kgd_dev *kgd, void *mqd, uint32_t pipe_id,
-			uint32_t queue_id, uint32_t __user *wptr)
+			uint32_t queue_id, uint32_t __user *wptr,
+			uint32_t wptr_shift, uint32_t wptr_mask,
+			struct mm_struct *mm)
 {
 	struct amdgpu_device *adev = get_amdgpu_device(kgd);
-	uint32_t wptr_shadow, is_wptr_shadow_valid;
 	struct cik_mqd *m;
+	uint32_t *mqd_hqd;
+	uint32_t reg, wptr_val, data;
 
 	m = get_mqd(mqd);
 
-	is_wptr_shadow_valid = !get_user(wptr_shadow, wptr);
-	if (is_wptr_shadow_valid)
-		m->cp_hqd_pq_wptr = wptr_shadow;
-
 	acquire_queue(kgd, pipe_id, queue_id);
-	gfx_v7_0_mqd_commit(adev, m);
+
+	/* HQD registers extend from CP_MQD_BASE_ADDR to CP_MQD_CONTROL. */
+	mqd_hqd = &m->cp_mqd_base_addr_lo;
+
+	for (reg = mmCP_MQD_BASE_ADDR; reg <= mmCP_MQD_CONTROL; reg++)
+		WREG32(reg, mqd_hqd[reg - mmCP_MQD_BASE_ADDR]);
+
+	/* Copy userspace write pointer value to register.
+	 * Activate doorbell logic to monitor subsequent changes.
+	 */
+	data = REG_SET_FIELD(m->cp_hqd_pq_doorbell_control,
+			     CP_HQD_PQ_DOORBELL_CONTROL, DOORBELL_EN, 1);
+	WREG32(mmCP_HQD_PQ_DOORBELL_CONTROL, data);
+
+	if (read_user_wptr(mm, wptr, wptr_val))
+		WREG32(mmCP_HQD_PQ_WPTR, (wptr_val << wptr_shift) & wptr_mask);
+
+	data = REG_SET_FIELD(m->cp_hqd_active, CP_HQD_ACTIVE, ACTIVE, 1);
+	WREG32(mmCP_HQD_ACTIVE, data);
+
 	release_queue(kgd);
 
 	return 0;
@@ -382,30 +438,99 @@ static bool kgd_hqd_sdma_is_occupied(struct kgd_dev *kgd, void *mqd)
 	return false;
 }
 
-static int kgd_hqd_destroy(struct kgd_dev *kgd, uint32_t reset_type,
+static int kgd_hqd_destroy(struct kgd_dev *kgd, void *mqd,
+				enum kfd_preempt_type reset_type,
 				unsigned int utimeout, uint32_t pipe_id,
 				uint32_t queue_id)
 {
 	struct amdgpu_device *adev = get_amdgpu_device(kgd);
 	uint32_t temp;
-	int timeout = utimeout;
+	enum hqd_dequeue_request_type type;
+	unsigned long flags, end_jiffies;
+	int retry;
 
 	acquire_queue(kgd, pipe_id, queue_id);
 	WREG32(mmCP_HQD_PQ_DOORBELL_CONTROL, 0);
 
-	WREG32(mmCP_HQD_DEQUEUE_REQUEST, reset_type);
+	switch (reset_type) {
+	case KFD_PREEMPT_TYPE_WAVEFRONT_DRAIN:
+		type = DRAIN_PIPE;
+		break;
+	case KFD_PREEMPT_TYPE_WAVEFRONT_RESET:
+		type = RESET_WAVES;
+		break;
+	default:
+		type = DRAIN_PIPE;
+		break;
+	}
+
+	/* Workaround: If IQ timer is active and the wait time is close to or
+	 * equal to 0, dequeueing is not safe. Wait until either the wait time
+	 * is larger or timer is cleared. Also, ensure that IQ_REQ_PEND is
+	 * cleared before continuing. Also, ensure wait times are set to at
+	 * least 0x3.
+	 */
+	local_irq_save(flags);
+	preempt_disable();
+	retry = 5000; /* wait for 500 usecs at maximum */
+	while (true) {
+		temp = RREG32(mmCP_HQD_IQ_TIMER);
+		if (REG_GET_FIELD(temp, CP_HQD_IQ_TIMER, PROCESSING_IQ)) {
+			pr_debug("HW is processing IQ\n");
+			goto loop;
+		}
+		if (REG_GET_FIELD(temp, CP_HQD_IQ_TIMER, ACTIVE)) {
+			if (REG_GET_FIELD(temp, CP_HQD_IQ_TIMER, RETRY_TYPE)
+					== 3) /* SEM-rearm is safe */
+				break;
+			/* Wait time 3 is safe for CP, but our MMIO read/write
+			 * time is close to 1 microsecond, so check for 10 to
+			 * leave more buffer room
+			 */
+			if (REG_GET_FIELD(temp, CP_HQD_IQ_TIMER, WAIT_TIME)
+					>= 10)
+				break;
+			pr_debug("IQ timer is active\n");
+		} else
+			break;
+loop:
+		if (!retry) {
+			pr_err("CP HQD IQ timer status time out\n");
+			break;
+		}
+		ndelay(100);
+		--retry;
+	}
+	retry = 1000;
+	while (true) {
+		temp = RREG32(mmCP_HQD_DEQUEUE_REQUEST);
+		if (!(temp & CP_HQD_DEQUEUE_REQUEST__IQ_REQ_PEND_MASK))
+			break;
+		pr_debug("Dequeue request is pending\n");
 
+		if (!retry) {
+			pr_err("CP HQD dequeue request time out\n");
+			break;
+		}
+		ndelay(100);
+		--retry;
+	}
+	local_irq_restore(flags);
+	preempt_enable();
+
+	WREG32(mmCP_HQD_DEQUEUE_REQUEST, type);
+
+	end_jiffies = (utimeout * HZ / 1000) + jiffies;
 	while (true) {
 		temp = RREG32(mmCP_HQD_ACTIVE);
-		if (temp & CP_HQD_ACTIVE__ACTIVE_MASK)
+		if (!(temp & CP_HQD_ACTIVE__ACTIVE_MASK))
 			break;
-		if (timeout <= 0) {
-			pr_err("kfd: cp queue preemption time out.\n");
+		if (time_after(jiffies, end_jiffies)) {
+			pr_err("cp queue preemption time out\n");
 			release_queue(kgd);
 			return -ETIME;
 		}
-		msleep(20);
-		timeout -= 20;
+		usleep_range(500, 1000);
 	}
 
 	release_queue(kgd);
@@ -556,6 +681,16 @@ static void write_vmid_invalidate_request(struct kgd_dev *kgd, uint8_t vmid)
 	WREG32(mmVM_INVALIDATE_REQUEST, 1 << vmid);
 }
 
+static void set_scratch_backing_va(struct kgd_dev *kgd,
+					uint64_t va, uint32_t vmid)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *) kgd;
+
+	lock_srbm(kgd, 0, 0, 0, vmid);
+	WREG32(mmSH_HIDDEN_PRIVATE_BASE_VMID, va);
+	unlock_srbm(kgd);
+}
+
 static uint16_t get_fw_version(struct kgd_dev *kgd, enum kgd_engine_type type)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *) kgd;
@@ -566,42 +701,42 @@ static uint16_t get_fw_version(struct kgd_dev *kgd, enum kgd_engine_type type)
 	switch (type) {
 	case KGD_ENGINE_PFP:
 		hdr = (const union amdgpu_firmware_header *)
-							adev->gfx.pfp_fw->data;
+						adev->gfx.pfp_fw->data;
 		break;
 
 	case KGD_ENGINE_ME:
 		hdr = (const union amdgpu_firmware_header *)
-							adev->gfx.me_fw->data;
+						adev->gfx.me_fw->data;
 		break;
 
 	case KGD_ENGINE_CE:
 		hdr = (const union amdgpu_firmware_header *)
-							adev->gfx.ce_fw->data;
+						adev->gfx.ce_fw->data;
 		break;
 
 	case KGD_ENGINE_MEC1:
 		hdr = (const union amdgpu_firmware_header *)
-							adev->gfx.mec_fw->data;
+						adev->gfx.mec_fw->data;
 		break;
 
 	case KGD_ENGINE_MEC2:
 		hdr = (const union amdgpu_firmware_header *)
-							adev->gfx.mec2_fw->data;
+						adev->gfx.mec2_fw->data;
 		break;
 
 	case KGD_ENGINE_RLC:
 		hdr = (const union amdgpu_firmware_header *)
-							adev->gfx.rlc_fw->data;
+						adev->gfx.rlc_fw->data;
 		break;
 
 	case KGD_ENGINE_SDMA1:
 		hdr = (const union amdgpu_firmware_header *)
-							adev->sdma.instance[0].fw->data;
+						adev->sdma.instance[0].fw->data;
 		break;
 
 	case KGD_ENGINE_SDMA2:
 		hdr = (const union amdgpu_firmware_header *)
-							adev->sdma.instance[1].fw->data;
+						adev->sdma.instance[1].fw->data;
 		break;
 
 	default:
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c
index 133d06671e46..fb6e5dbd5a03 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c
@@ -39,6 +39,12 @@
 #include "vi_structs.h"
 #include "vid.h"
 
+enum hqd_dequeue_request_type {
+	NO_ACTION = 0,
+	DRAIN_PIPE,
+	RESET_WAVES
+};
+
 struct cik_sdma_rlc_registers;
 
 /*
@@ -55,12 +61,15 @@ static int kgd_init_pipeline(struct kgd_dev *kgd, uint32_t pipe_id,
 		uint32_t hpd_size, uint64_t hpd_gpu_addr);
 static int kgd_init_interrupts(struct kgd_dev *kgd, uint32_t pipe_id);
 static int kgd_hqd_load(struct kgd_dev *kgd, void *mqd, uint32_t pipe_id,
-		uint32_t queue_id, uint32_t __user *wptr);
+			uint32_t queue_id, uint32_t __user *wptr,
+			uint32_t wptr_shift, uint32_t wptr_mask,
+			struct mm_struct *mm);
 static int kgd_hqd_sdma_load(struct kgd_dev *kgd, void *mqd);
 static bool kgd_hqd_is_occupied(struct kgd_dev *kgd, uint64_t queue_address,
 		uint32_t pipe_id, uint32_t queue_id);
 static bool kgd_hqd_sdma_is_occupied(struct kgd_dev *kgd, void *mqd);
-static int kgd_hqd_destroy(struct kgd_dev *kgd, uint32_t reset_type,
+static int kgd_hqd_destroy(struct kgd_dev *kgd, void *mqd,
+				enum kfd_preempt_type reset_type,
 				unsigned int utimeout, uint32_t pipe_id,
 				uint32_t queue_id);
 static int kgd_hqd_sdma_destroy(struct kgd_dev *kgd, void *mqd,
@@ -85,6 +94,33 @@ static uint16_t get_atc_vmid_pasid_mapping_pasid(struct kgd_dev *kgd,
 		uint8_t vmid);
 static void write_vmid_invalidate_request(struct kgd_dev *kgd, uint8_t vmid);
 static uint16_t get_fw_version(struct kgd_dev *kgd, enum kgd_engine_type type);
+static void set_scratch_backing_va(struct kgd_dev *kgd,
+					uint64_t va, uint32_t vmid);
+
+/* Because of REG_GET_FIELD() being used, we put this function in the
+ * asic specific file.
+ */
+static int get_tile_config(struct kgd_dev *kgd,
+		struct tile_config *config)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)kgd;
+
+	config->gb_addr_config = adev->gfx.config.gb_addr_config;
+	config->num_banks = REG_GET_FIELD(adev->gfx.config.mc_arb_ramcfg,
+				MC_ARB_RAMCFG, NOOFBANK);
+	config->num_ranks = REG_GET_FIELD(adev->gfx.config.mc_arb_ramcfg,
+				MC_ARB_RAMCFG, NOOFRANKS);
+
+	config->tile_config_ptr = adev->gfx.config.tile_mode_array;
+	config->num_tile_configs =
+			ARRAY_SIZE(adev->gfx.config.tile_mode_array);
+	config->macro_tile_config_ptr =
+			adev->gfx.config.macrotile_mode_array;
+	config->num_macro_tile_configs =
+			ARRAY_SIZE(adev->gfx.config.macrotile_mode_array);
+
+	return 0;
+}
 
 static const struct kfd2kgd_calls kfd2kgd = {
 	.init_gtt_mem_allocation = alloc_gtt_mem,
@@ -111,12 +147,15 @@ static const struct kfd2kgd_calls kfd2kgd = {
 	.get_atc_vmid_pasid_mapping_valid =
 			get_atc_vmid_pasid_mapping_valid,
 	.write_vmid_invalidate_request = write_vmid_invalidate_request,
-	.get_fw_version = get_fw_version
+	.get_fw_version = get_fw_version,
+	.set_scratch_backing_va = set_scratch_backing_va,
+	.get_tile_config = get_tile_config,
 };
 
 struct kfd2kgd_calls *amdgpu_amdkfd_gfx_8_0_get_functions(void)
 {
 	return (struct kfd2kgd_calls *)&kfd2kgd;
+	return (struct kfd2kgd_calls *)&kfd2kgd;
 }
 
 static inline struct amdgpu_device *get_amdgpu_device(struct kgd_dev *kgd)
@@ -147,7 +186,7 @@ static void acquire_queue(struct kgd_dev *kgd, uint32_t pipe_id,
 {
 	struct amdgpu_device *adev = get_amdgpu_device(kgd);
 
-	uint32_t mec = (++pipe_id / adev->gfx.mec.num_pipe_per_mec) + 1;
+	uint32_t mec = (pipe_id / adev->gfx.mec.num_pipe_per_mec) + 1;
 	uint32_t pipe = (pipe_id % adev->gfx.mec.num_pipe_per_mec);
 
 	lock_srbm(kgd, mec, pipe, queue_id, 0);
@@ -216,7 +255,7 @@ static int kgd_init_interrupts(struct kgd_dev *kgd, uint32_t pipe_id)
 	uint32_t mec;
 	uint32_t pipe;
 
-	mec = (++pipe_id / adev->gfx.mec.num_pipe_per_mec) + 1;
+	mec = (pipe_id / adev->gfx.mec.num_pipe_per_mec) + 1;
 	pipe = (pipe_id % adev->gfx.mec.num_pipe_per_mec);
 
 	lock_srbm(kgd, mec, pipe, 0, 0);
@@ -244,20 +283,67 @@ static inline struct cik_sdma_rlc_registers *get_sdma_mqd(void *mqd)
 }
 
 static int kgd_hqd_load(struct kgd_dev *kgd, void *mqd, uint32_t pipe_id,
-			uint32_t queue_id, uint32_t __user *wptr)
+			uint32_t queue_id, uint32_t __user *wptr,
+			uint32_t wptr_shift, uint32_t wptr_mask,
+			struct mm_struct *mm)
 {
-	struct vi_mqd *m;
-	uint32_t shadow_wptr, valid_wptr;
 	struct amdgpu_device *adev = get_amdgpu_device(kgd);
+	struct vi_mqd *m;
+	uint32_t *mqd_hqd;
+	uint32_t reg, wptr_val, data;
 
 	m = get_mqd(mqd);
 
-	valid_wptr = copy_from_user(&shadow_wptr, wptr, sizeof(shadow_wptr));
-	if (valid_wptr == 0)
-		m->cp_hqd_pq_wptr = shadow_wptr;
-
 	acquire_queue(kgd, pipe_id, queue_id);
-	gfx_v8_0_mqd_commit(adev, mqd);
+
+	/* HIQ is set during driver init period with vmid set to 0*/
+	if (m->cp_hqd_vmid == 0) {
+		uint32_t value, mec, pipe;
+
+		mec = (pipe_id / adev->gfx.mec.num_pipe_per_mec) + 1;
+		pipe = (pipe_id % adev->gfx.mec.num_pipe_per_mec);
+
+		pr_debug("kfd: set HIQ, mec:%d, pipe:%d, queue:%d.\n",
+			mec, pipe, queue_id);
+		value = RREG32(mmRLC_CP_SCHEDULERS);
+		value = REG_SET_FIELD(value, RLC_CP_SCHEDULERS, scheduler1,
+			((mec << 5) | (pipe << 3) | queue_id | 0x80));
+		WREG32(mmRLC_CP_SCHEDULERS, value);
+	}
+
+	/* HQD registers extend from CP_MQD_BASE_ADDR to CP_HQD_EOP_WPTR_MEM. */
+	mqd_hqd = &m->cp_mqd_base_addr_lo;
+
+	for (reg = mmCP_MQD_BASE_ADDR; reg <= mmCP_HQD_EOP_CONTROL; reg++)
+		WREG32(reg, mqd_hqd[reg - mmCP_MQD_BASE_ADDR]);
+
+	/* Tonga errata: EOP RPTR/WPTR should be left unmodified.
+	 * This is safe since EOP RPTR==WPTR for any inactive HQD
+	 * on ASICs that do not support context-save.
+	 * EOP writes/reads can start anywhere in the ring.
+	 */
+	if (get_amdgpu_device(kgd)->asic_type != CHIP_TONGA) {
+		WREG32(mmCP_HQD_EOP_RPTR, m->cp_hqd_eop_rptr);
+		WREG32(mmCP_HQD_EOP_WPTR, m->cp_hqd_eop_wptr);
+		WREG32(mmCP_HQD_EOP_WPTR_MEM, m->cp_hqd_eop_wptr_mem);
+	}
+
+	for (reg = mmCP_HQD_EOP_EVENTS; reg <= mmCP_HQD_ERROR; reg++)
+		WREG32(reg, mqd_hqd[reg - mmCP_MQD_BASE_ADDR]);
+
+	/* Copy userspace write pointer value to register.
+	 * Activate doorbell logic to monitor subsequent changes.
+	 */
+	data = REG_SET_FIELD(m->cp_hqd_pq_doorbell_control,
+			     CP_HQD_PQ_DOORBELL_CONTROL, DOORBELL_EN, 1);
+	WREG32(mmCP_HQD_PQ_DOORBELL_CONTROL, data);
+
+	if (read_user_wptr(mm, wptr, wptr_val))
+		WREG32(mmCP_HQD_PQ_WPTR, (wptr_val << wptr_shift) & wptr_mask);
+
+	data = REG_SET_FIELD(m->cp_hqd_active, CP_HQD_ACTIVE, ACTIVE, 1);
+	WREG32(mmCP_HQD_ACTIVE, data);
+
 	release_queue(kgd);
 
 	return 0;
@@ -308,29 +394,102 @@ static bool kgd_hqd_sdma_is_occupied(struct kgd_dev *kgd, void *mqd)
 	return false;
 }
 
-static int kgd_hqd_destroy(struct kgd_dev *kgd, uint32_t reset_type,
+static int kgd_hqd_destroy(struct kgd_dev *kgd, void *mqd,
+				enum kfd_preempt_type reset_type,
 				unsigned int utimeout, uint32_t pipe_id,
 				uint32_t queue_id)
 {
 	struct amdgpu_device *adev = get_amdgpu_device(kgd);
 	uint32_t temp;
-	int timeout = utimeout;
+	enum hqd_dequeue_request_type type;
+	unsigned long flags, end_jiffies;
+	int retry;
+	struct vi_mqd *m = get_mqd(mqd);
 
 	acquire_queue(kgd, pipe_id, queue_id);
 
-	WREG32(mmCP_HQD_DEQUEUE_REQUEST, reset_type);
+	if (m->cp_hqd_vmid == 0)
+		WREG32_FIELD(RLC_CP_SCHEDULERS, scheduler1, 0);
+
+	switch (reset_type) {
+	case KFD_PREEMPT_TYPE_WAVEFRONT_DRAIN:
+		type = DRAIN_PIPE;
+		break;
+	case KFD_PREEMPT_TYPE_WAVEFRONT_RESET:
+		type = RESET_WAVES;
+		break;
+	default:
+		type = DRAIN_PIPE;
+		break;
+	}
 
+	/* Workaround: If IQ timer is active and the wait time is close to or
+	 * equal to 0, dequeueing is not safe. Wait until either the wait time
+	 * is larger or timer is cleared. Also, ensure that IQ_REQ_PEND is
+	 * cleared before continuing. Also, ensure wait times are set to at
+	 * least 0x3.
+	 */
+	local_irq_save(flags);
+	preempt_disable();
+	retry = 5000; /* wait for 500 usecs at maximum */
+	while (true) {
+		temp = RREG32(mmCP_HQD_IQ_TIMER);
+		if (REG_GET_FIELD(temp, CP_HQD_IQ_TIMER, PROCESSING_IQ)) {
+			pr_debug("HW is processing IQ\n");
+			goto loop;
+		}
+		if (REG_GET_FIELD(temp, CP_HQD_IQ_TIMER, ACTIVE)) {
+			if (REG_GET_FIELD(temp, CP_HQD_IQ_TIMER, RETRY_TYPE)
+					== 3) /* SEM-rearm is safe */
+				break;
+			/* Wait time 3 is safe for CP, but our MMIO read/write
+			 * time is close to 1 microsecond, so check for 10 to
+			 * leave more buffer room
+			 */
+			if (REG_GET_FIELD(temp, CP_HQD_IQ_TIMER, WAIT_TIME)
+					>= 10)
+				break;
+			pr_debug("IQ timer is active\n");
+		} else
+			break;
+loop:
+		if (!retry) {
+			pr_err("CP HQD IQ timer status time out\n");
+			break;
+		}
+		ndelay(100);
+		--retry;
+	}
+	retry = 1000;
+	while (true) {
+		temp = RREG32(mmCP_HQD_DEQUEUE_REQUEST);
+		if (!(temp & CP_HQD_DEQUEUE_REQUEST__IQ_REQ_PEND_MASK))
+			break;
+		pr_debug("Dequeue request is pending\n");
+
+		if (!retry) {
+			pr_err("CP HQD dequeue request time out\n");
+			break;
+		}
+		ndelay(100);
+		--retry;
+	}
+	local_irq_restore(flags);
+	preempt_enable();
+
+	WREG32(mmCP_HQD_DEQUEUE_REQUEST, type);
+
+	end_jiffies = (utimeout * HZ / 1000) + jiffies;
 	while (true) {
 		temp = RREG32(mmCP_HQD_ACTIVE);
-		if (temp & CP_HQD_ACTIVE__ACTIVE_MASK)
+		if (!(temp & CP_HQD_ACTIVE__ACTIVE_MASK))
 			break;
-		if (timeout <= 0) {
-			pr_err("kfd: cp queue preemption time out.\n");
+		if (time_after(jiffies, end_jiffies)) {
+			pr_err("cp queue preemption time out.\n");
 			release_queue(kgd);
 			return -ETIME;
 		}
-		msleep(20);
-		timeout -= 20;
+		usleep_range(500, 1000);
 	}
 
 	release_queue(kgd);
@@ -444,6 +603,16 @@ static uint32_t kgd_address_watch_get_offset(struct kgd_dev *kgd,
 	return 0;
 }
 
+static void set_scratch_backing_va(struct kgd_dev *kgd,
+					uint64_t va, uint32_t vmid)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *) kgd;
+
+	lock_srbm(kgd, 0, 0, 0, vmid);
+	WREG32(mmSH_HIDDEN_PRIVATE_BASE_VMID, va);
+	unlock_srbm(kgd);
+}
+
 static uint16_t get_fw_version(struct kgd_dev *kgd, enum kgd_engine_type type)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *) kgd;
@@ -454,42 +623,42 @@ static uint16_t get_fw_version(struct kgd_dev *kgd, enum kgd_engine_type type)
 	switch (type) {
 	case KGD_ENGINE_PFP:
 		hdr = (const union amdgpu_firmware_header *)
-							adev->gfx.pfp_fw->data;
+						adev->gfx.pfp_fw->data;
 		break;
 
 	case KGD_ENGINE_ME:
 		hdr = (const union amdgpu_firmware_header *)
-							adev->gfx.me_fw->data;
+						adev->gfx.me_fw->data;
 		break;
 
 	case KGD_ENGINE_CE:
 		hdr = (const union amdgpu_firmware_header *)
-							adev->gfx.ce_fw->data;
+						adev->gfx.ce_fw->data;
 		break;
 
 	case KGD_ENGINE_MEC1:
 		hdr = (const union amdgpu_firmware_header *)
-							adev->gfx.mec_fw->data;
+						adev->gfx.mec_fw->data;
 		break;
 
 	case KGD_ENGINE_MEC2:
 		hdr = (const union amdgpu_firmware_header *)
-							adev->gfx.mec2_fw->data;
+						adev->gfx.mec2_fw->data;
 		break;
 
 	case KGD_ENGINE_RLC:
 		hdr = (const union amdgpu_firmware_header *)
-							adev->gfx.rlc_fw->data;
+						adev->gfx.rlc_fw->data;
 		break;
 
 	case KGD_ENGINE_SDMA1:
 		hdr = (const union amdgpu_firmware_header *)
-							adev->sdma.instance[0].fw->data;
+						adev->sdma.instance[0].fw->data;
 		break;
 
 	case KGD_ENGINE_SDMA2:
 		hdr = (const union amdgpu_firmware_header *)
-							adev->sdma.instance[1].fw->data;
+						adev->sdma.instance[1].fw->data;
 		break;
 
 	default:
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_benchmark.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_benchmark.c
index 2fb299afc12b..63ec1e1bb6aa 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_benchmark.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_benchmark.c
@@ -81,7 +81,7 @@ static void amdgpu_benchmark_move(struct amdgpu_device *adev, unsigned size,
 
 	n = AMDGPU_BENCHMARK_ITERATIONS;
 	r = amdgpu_bo_create(adev, size, PAGE_SIZE, true, sdomain, 0, NULL,
-			     NULL, &sobj);
+			     NULL, 0, &sobj);
 	if (r) {
 		goto out_cleanup;
 	}
@@ -94,7 +94,7 @@ static void amdgpu_benchmark_move(struct amdgpu_device *adev, unsigned size,
 		goto out_cleanup;
 	}
 	r = amdgpu_bo_create(adev, size, PAGE_SIZE, true, ddomain, 0, NULL,
-			     NULL, &dobj);
+			     NULL, 0, &dobj);
 	if (r) {
 		goto out_cleanup;
 	}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c
index d324e1c24028..59089e027f4d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c
@@ -136,7 +136,7 @@ static int amdgpu_bo_list_set(struct amdgpu_device *adev,
 		}
 
 		bo = amdgpu_bo_ref(gem_to_amdgpu_bo(gobj));
-		drm_gem_object_unreference_unlocked(gobj);
+		drm_gem_object_put_unlocked(gobj);
 
 		usermm = amdgpu_ttm_tt_get_usermm(bo->tbo.ttm);
 		if (usermm) {
@@ -156,11 +156,11 @@ static int amdgpu_bo_list_set(struct amdgpu_device *adev,
 		entry->tv.bo = &entry->robj->tbo;
 		entry->tv.shared = !entry->robj->prime_shared_count;
 
-		if (entry->robj->prefered_domains == AMDGPU_GEM_DOMAIN_GDS)
+		if (entry->robj->preferred_domains == AMDGPU_GEM_DOMAIN_GDS)
 			gds_obj = entry->robj;
-		if (entry->robj->prefered_domains == AMDGPU_GEM_DOMAIN_GWS)
+		if (entry->robj->preferred_domains == AMDGPU_GEM_DOMAIN_GWS)
 			gws_obj = entry->robj;
-		if (entry->robj->prefered_domains == AMDGPU_GEM_DOMAIN_OA)
+		if (entry->robj->preferred_domains == AMDGPU_GEM_DOMAIN_OA)
 			oa_obj = entry->robj;
 
 		total_size += amdgpu_bo_size(entry->robj);
@@ -270,7 +270,7 @@ int amdgpu_bo_list_ioctl(struct drm_device *dev, void *data,
 	struct amdgpu_fpriv *fpriv = filp->driver_priv;
 	union drm_amdgpu_bo_list *args = data;
 	uint32_t handle = args->in.list_handle;
-	const void __user *uptr = (const void*)(uintptr_t)args->in.bo_info_ptr;
+	const void __user *uptr = u64_to_user_ptr(args->in.bo_info_ptr);
 
 	struct drm_amdgpu_bo_list_entry *info;
 	struct amdgpu_bo_list *list;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c
index a99e0bca6812..fd435a96481c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c
@@ -124,7 +124,7 @@ static int amdgpu_cgs_alloc_gpu_mem(struct cgs_device *cgs_device,
 	ret = amdgpu_bo_create_restricted(adev, size, PAGE_SIZE,
 					  true, domain, flags,
 					  NULL, &placement, NULL,
-					  &obj);
+					  0, &obj);
 	if (ret) {
 		DRM_ERROR("(%d) bo create failed\n", ret);
 		return ret;
@@ -166,7 +166,7 @@ static int amdgpu_cgs_gmap_gpu_mem(struct cgs_device *cgs_device, cgs_handle_t h
 	r = amdgpu_bo_reserve(obj, true);
 	if (unlikely(r != 0))
 		return r;
-	r = amdgpu_bo_pin_restricted(obj, obj->prefered_domains,
+	r = amdgpu_bo_pin_restricted(obj, obj->preferred_domains,
 				     min_offset, max_offset, mcaddr);
 	amdgpu_bo_unreserve(obj);
 	return r;
@@ -659,7 +659,7 @@ static int amdgpu_cgs_get_firmware_info(struct cgs_device *cgs_device,
 		info->version = (uint16_t)le32_to_cpu(header->header.ucode_version);
 
 		if (CGS_UCODE_ID_CP_MEC == type)
-			info->image_size = (header->jt_offset) << 2;
+			info->image_size = le32_to_cpu(header->jt_offset) << 2;
 
 		info->fw_version = amdgpu_get_firmware_version(cgs_device, type);
 		info->feature_version = (uint16_t)le32_to_cpu(header->ucode_feature_version);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
index 33789510e663..269b835571eb 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
@@ -54,7 +54,7 @@ static int amdgpu_cs_user_fence_chunk(struct amdgpu_cs_parser *p,
 
 	*offset = data->offset;
 
-	drm_gem_object_unreference_unlocked(gobj);
+	drm_gem_object_put_unlocked(gobj);
 
 	if (amdgpu_ttm_tt_get_usermm(p->uf_entry.robj->tbo.ttm)) {
 		amdgpu_bo_unref(&p->uf_entry.robj);
@@ -90,7 +90,7 @@ static int amdgpu_cs_parser_init(struct amdgpu_cs_parser *p, void *data)
 	}
 
 	/* get chunks */
-	chunk_array_user = (uint64_t __user *)(uintptr_t)(cs->in.chunks);
+	chunk_array_user = u64_to_user_ptr(cs->in.chunks);
 	if (copy_from_user(chunk_array, chunk_array_user,
 			   sizeof(uint64_t)*cs->in.num_chunks)) {
 		ret = -EFAULT;
@@ -110,7 +110,7 @@ static int amdgpu_cs_parser_init(struct amdgpu_cs_parser *p, void *data)
 		struct drm_amdgpu_cs_chunk user_chunk;
 		uint32_t __user *cdata;
 
-		chunk_ptr = (void __user *)(uintptr_t)chunk_array[i];
+		chunk_ptr = u64_to_user_ptr(chunk_array[i]);
 		if (copy_from_user(&user_chunk, chunk_ptr,
 				       sizeof(struct drm_amdgpu_cs_chunk))) {
 			ret = -EFAULT;
@@ -121,7 +121,7 @@ static int amdgpu_cs_parser_init(struct amdgpu_cs_parser *p, void *data)
 		p->chunks[i].length_dw = user_chunk.length_dw;
 
 		size = p->chunks[i].length_dw;
-		cdata = (void __user *)(uintptr_t)user_chunk.chunk_data;
+		cdata = u64_to_user_ptr(user_chunk.chunk_data);
 
 		p->chunks[i].kdata = kvmalloc_array(size, sizeof(uint32_t), GFP_KERNEL);
 		if (p->chunks[i].kdata == NULL) {
@@ -246,7 +246,7 @@ static void amdgpu_cs_get_threshold_for_moves(struct amdgpu_device *adev,
 	}
 
 	total_vram = adev->mc.real_vram_size - adev->vram_pin_size;
-	used_vram = atomic64_read(&adev->vram_usage);
+	used_vram = amdgpu_vram_mgr_usage(&adev->mman.bdev.man[TTM_PL_VRAM]);
 	free_vram = used_vram >= total_vram ? 0 : total_vram - used_vram;
 
 	spin_lock(&adev->mm_stats.lock);
@@ -292,7 +292,8 @@ static void amdgpu_cs_get_threshold_for_moves(struct amdgpu_device *adev,
 	/* Do the same for visible VRAM if half of it is free */
 	if (adev->mc.visible_vram_size < adev->mc.real_vram_size) {
 		u64 total_vis_vram = adev->mc.visible_vram_size;
-		u64 used_vis_vram = atomic64_read(&adev->vram_vis_usage);
+		u64 used_vis_vram =
+			amdgpu_vram_mgr_vis_usage(&adev->mman.bdev.man[TTM_PL_VRAM]);
 
 		if (used_vis_vram < total_vis_vram) {
 			u64 free_vis_vram = total_vis_vram - used_vis_vram;
@@ -348,11 +349,11 @@ static int amdgpu_cs_bo_validate(struct amdgpu_cs_parser *p,
 			 * that.
 			 */
 			if (p->bytes_moved_vis < p->bytes_moved_vis_threshold)
-				domain = bo->prefered_domains;
+				domain = bo->preferred_domains;
 			else
 				domain = bo->allowed_domains;
 		} else {
-			domain = bo->prefered_domains;
+			domain = bo->preferred_domains;
 		}
 	} else {
 		domain = bo->allowed_domains;
@@ -673,10 +674,8 @@ static int amdgpu_cs_parser_bos(struct amdgpu_cs_parser *p,
 	}
 
 error_validate:
-	if (r) {
-		amdgpu_vm_move_pt_bos_in_lru(p->adev, &fpriv->vm);
+	if (r)
 		ttm_eu_backoff_reservation(&p->ticket, &p->validated);
-	}
 
 error_free_pages:
 
@@ -724,21 +723,18 @@ static int amdgpu_cs_sync_rings(struct amdgpu_cs_parser *p)
  * If error is set than unvalidate buffer, otherwise just free memory
  * used by parsing context.
  **/
-static void amdgpu_cs_parser_fini(struct amdgpu_cs_parser *parser, int error, bool backoff)
+static void amdgpu_cs_parser_fini(struct amdgpu_cs_parser *parser, int error,
+				  bool backoff)
 {
-	struct amdgpu_fpriv *fpriv = parser->filp->driver_priv;
 	unsigned i;
 
-	if (!error) {
-		amdgpu_vm_move_pt_bos_in_lru(parser->adev, &fpriv->vm);
-
+	if (!error)
 		ttm_eu_fence_buffer_objects(&parser->ticket,
 					    &parser->validated,
 					    parser->fence);
-	} else if (backoff) {
+	else if (backoff)
 		ttm_eu_backoff_reservation(&parser->ticket,
 					   &parser->validated);
-	}
 
 	for (i = 0; i < parser->num_post_dep_syncobjs; i++)
 		drm_syncobj_put(parser->post_dep_syncobjs[i]);
@@ -791,7 +787,8 @@ static int amdgpu_bo_vm_update_pte(struct amdgpu_cs_parser *p)
 
 	if (amdgpu_sriov_vf(adev)) {
 		struct dma_fence *f;
-		bo_va = vm->csa_bo_va;
+
+		bo_va = fpriv->csa_va;
 		BUG_ON(!bo_va);
 		r = amdgpu_vm_bo_update(adev, bo_va, false);
 		if (r)
@@ -828,7 +825,7 @@ static int amdgpu_bo_vm_update_pte(struct amdgpu_cs_parser *p)
 
 	}
 
-	r = amdgpu_vm_clear_invalids(adev, vm, &p->job->sync);
+	r = amdgpu_vm_clear_moved(adev, vm, &p->job->sync);
 
 	if (amdgpu_vm_debug && p->bo_list) {
 		/* Invalidate all BOs to test for userspace bugs */
@@ -1038,7 +1035,7 @@ static int amdgpu_syncobj_lookup_and_add_to_sync(struct amdgpu_cs_parser *p,
 {
 	int r;
 	struct dma_fence *fence;
-	r = drm_syncobj_fence_get(p->filp, handle, &fence);
+	r = drm_syncobj_find_fence(p->filp, handle, &fence);
 	if (r)
 		return r;
 
@@ -1437,7 +1434,7 @@ int amdgpu_cs_wait_fences_ioctl(struct drm_device *dev, void *data,
 	if (fences == NULL)
 		return -ENOMEM;
 
-	fences_user = (void __user *)(uintptr_t)(wait->in.fences);
+	fences_user = u64_to_user_ptr(wait->in.fences);
 	if (copy_from_user(fences, fences_user,
 		sizeof(struct drm_amdgpu_fence) * fence_count)) {
 		r = -EFAULT;
@@ -1490,7 +1487,7 @@ amdgpu_cs_find_mapping(struct amdgpu_cs_parser *parser,
 			    addr > mapping->last)
 				continue;
 
-			*bo = lobj->bo_va->bo;
+			*bo = lobj->bo_va->base.bo;
 			return mapping;
 		}
 
@@ -1499,7 +1496,7 @@ amdgpu_cs_find_mapping(struct amdgpu_cs_parser *parser,
 			    addr > mapping->last)
 				continue;
 
-			*bo = lobj->bo_va->bo;
+			*bo = lobj->bo_va->base.bo;
 			return mapping;
 		}
 	}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index 6279956e92a4..1a459ac63df4 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -336,51 +336,16 @@ static void amdgpu_block_invalid_wreg(struct amdgpu_device *adev,
 
 static int amdgpu_vram_scratch_init(struct amdgpu_device *adev)
 {
-	int r;
-
-	if (adev->vram_scratch.robj == NULL) {
-		r = amdgpu_bo_create(adev, AMDGPU_GPU_PAGE_SIZE,
-				     PAGE_SIZE, true, AMDGPU_GEM_DOMAIN_VRAM,
-				     AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED |
-				     AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS,
-				     NULL, NULL, &adev->vram_scratch.robj);
-		if (r) {
-			return r;
-		}
-	}
-
-	r = amdgpu_bo_reserve(adev->vram_scratch.robj, false);
-	if (unlikely(r != 0))
-		return r;
-	r = amdgpu_bo_pin(adev->vram_scratch.robj,
-			  AMDGPU_GEM_DOMAIN_VRAM, &adev->vram_scratch.gpu_addr);
-	if (r) {
-		amdgpu_bo_unreserve(adev->vram_scratch.robj);
-		return r;
-	}
-	r = amdgpu_bo_kmap(adev->vram_scratch.robj,
-				(void **)&adev->vram_scratch.ptr);
-	if (r)
-		amdgpu_bo_unpin(adev->vram_scratch.robj);
-	amdgpu_bo_unreserve(adev->vram_scratch.robj);
-
-	return r;
+	return amdgpu_bo_create_kernel(adev, AMDGPU_GPU_PAGE_SIZE,
+				       PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM,
+				       &adev->vram_scratch.robj,
+				       &adev->vram_scratch.gpu_addr,
+				       (void **)&adev->vram_scratch.ptr);
 }
 
 static void amdgpu_vram_scratch_fini(struct amdgpu_device *adev)
 {
-	int r;
-
-	if (adev->vram_scratch.robj == NULL) {
-		return;
-	}
-	r = amdgpu_bo_reserve(adev->vram_scratch.robj, true);
-	if (likely(r == 0)) {
-		amdgpu_bo_kunmap(adev->vram_scratch.robj);
-		amdgpu_bo_unpin(adev->vram_scratch.robj);
-		amdgpu_bo_unreserve(adev->vram_scratch.robj);
-	}
-	amdgpu_bo_unref(&adev->vram_scratch.robj);
+	amdgpu_bo_free_kernel(&adev->vram_scratch.robj, NULL, NULL);
 }
 
 /**
@@ -539,7 +504,8 @@ static int amdgpu_wb_init(struct amdgpu_device *adev)
 	int r;
 
 	if (adev->wb.wb_obj == NULL) {
-		r = amdgpu_bo_create_kernel(adev, AMDGPU_MAX_WB * sizeof(uint32_t),
+		/* AMDGPU_MAX_WB * sizeof(uint32_t) * 8 = AMDGPU_MAX_WB 256bit slots */
+		r = amdgpu_bo_create_kernel(adev, AMDGPU_MAX_WB * sizeof(uint32_t) * 8,
 					    PAGE_SIZE, AMDGPU_GEM_DOMAIN_GTT,
 					    &adev->wb.wb_obj, &adev->wb.gpu_addr,
 					    (void **)&adev->wb.wb);
@@ -570,47 +536,10 @@ static int amdgpu_wb_init(struct amdgpu_device *adev)
 int amdgpu_wb_get(struct amdgpu_device *adev, u32 *wb)
 {
 	unsigned long offset = find_first_zero_bit(adev->wb.used, adev->wb.num_wb);
-	if (offset < adev->wb.num_wb) {
-		__set_bit(offset, adev->wb.used);
-		*wb = offset;
-		return 0;
-	} else {
-		return -EINVAL;
-	}
-}
 
-/**
- * amdgpu_wb_get_64bit - Allocate a wb entry
- *
- * @adev: amdgpu_device pointer
- * @wb: wb index
- *
- * Allocate a wb slot for use by the driver (all asics).
- * Returns 0 on success or -EINVAL on failure.
- */
-int amdgpu_wb_get_64bit(struct amdgpu_device *adev, u32 *wb)
-{
-	unsigned long offset = bitmap_find_next_zero_area_off(adev->wb.used,
-				adev->wb.num_wb, 0, 2, 7, 0);
-	if ((offset + 1) < adev->wb.num_wb) {
+	if (offset < adev->wb.num_wb) {
 		__set_bit(offset, adev->wb.used);
-		__set_bit(offset + 1, adev->wb.used);
-		*wb = offset;
-		return 0;
-	} else {
-		return -EINVAL;
-	}
-}
-
-int amdgpu_wb_get_256Bit(struct amdgpu_device *adev, u32 *wb)
-{
-	int i = 0;
-	unsigned long offset = bitmap_find_next_zero_area_off(adev->wb.used,
-				adev->wb.num_wb, 0, 8, 63, 0);
-	if ((offset + 7) < adev->wb.num_wb) {
-		for (i = 0; i < 8; i++)
-			__set_bit(offset + i, adev->wb.used);
-		*wb = offset;
+		*wb = offset * 8; /* convert to dw offset */
 		return 0;
 	} else {
 		return -EINVAL;
@@ -632,39 +561,6 @@ void amdgpu_wb_free(struct amdgpu_device *adev, u32 wb)
 }
 
 /**
- * amdgpu_wb_free_64bit - Free a wb entry
- *
- * @adev: amdgpu_device pointer
- * @wb: wb index
- *
- * Free a wb slot allocated for use by the driver (all asics)
- */
-void amdgpu_wb_free_64bit(struct amdgpu_device *adev, u32 wb)
-{
-	if ((wb + 1) < adev->wb.num_wb) {
-		__clear_bit(wb, adev->wb.used);
-		__clear_bit(wb + 1, adev->wb.used);
-	}
-}
-
-/**
- * amdgpu_wb_free_256bit - Free a wb entry
- *
- * @adev: amdgpu_device pointer
- * @wb: wb index
- *
- * Free a wb slot allocated for use by the driver (all asics)
- */
-void amdgpu_wb_free_256bit(struct amdgpu_device *adev, u32 wb)
-{
-	int i = 0;
-
-	if ((wb + 7) < adev->wb.num_wb)
-		for (i = 0; i < 8; i++)
-			__clear_bit(wb + i, adev->wb.used);
-}
-
-/**
  * amdgpu_vram_location - try to find VRAM location
  * @adev: amdgpu device structure holding all necessary informations
  * @mc: memory controller structure holding memory informations
@@ -1180,6 +1076,13 @@ static void amdgpu_check_arguments(struct amdgpu_device *adev)
 		amdgpu_gtt_size = -1;
 	}
 
+	/* valid range is between 4 and 9 inclusive */
+	if (amdgpu_vm_fragment_size != -1 &&
+	    (amdgpu_vm_fragment_size > 9 || amdgpu_vm_fragment_size < 4)) {
+		dev_warn(adev->dev, "valid range is between 4 and 9\n");
+		amdgpu_vm_fragment_size = -1;
+	}
+
 	amdgpu_check_vm_size(adev);
 
 	amdgpu_check_block_size(adev);
@@ -1948,7 +1851,8 @@ static int amdgpu_sriov_reinit_late(struct amdgpu_device *adev)
 		AMD_IP_BLOCK_TYPE_DCE,
 		AMD_IP_BLOCK_TYPE_GFX,
 		AMD_IP_BLOCK_TYPE_SDMA,
-		AMD_IP_BLOCK_TYPE_VCE,
+		AMD_IP_BLOCK_TYPE_UVD,
+		AMD_IP_BLOCK_TYPE_VCE
 	};
 
 	for (i = 0; i < ARRAY_SIZE(ip_order); i++) {
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c
index cdf2ab20166a..6ad243293a78 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c
@@ -482,7 +482,7 @@ static void amdgpu_user_framebuffer_destroy(struct drm_framebuffer *fb)
 {
 	struct amdgpu_framebuffer *amdgpu_fb = to_amdgpu_framebuffer(fb);
 
-	drm_gem_object_unreference_unlocked(amdgpu_fb->obj);
+	drm_gem_object_put_unlocked(amdgpu_fb->obj);
 	drm_framebuffer_cleanup(fb);
 	kfree(amdgpu_fb);
 }
@@ -542,14 +542,14 @@ amdgpu_user_framebuffer_create(struct drm_device *dev,
 
 	amdgpu_fb = kzalloc(sizeof(*amdgpu_fb), GFP_KERNEL);
 	if (amdgpu_fb == NULL) {
-		drm_gem_object_unreference_unlocked(obj);
+		drm_gem_object_put_unlocked(obj);
 		return ERR_PTR(-ENOMEM);
 	}
 
 	ret = amdgpu_framebuffer_init(dev, amdgpu_fb, mode_cmd, obj);
 	if (ret) {
 		kfree(amdgpu_fb);
-		drm_gem_object_unreference_unlocked(obj);
+		drm_gem_object_put_unlocked(obj);
 		return ERR_PTR(ret);
 	}
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index 5e9ce8a29669..e39ec981b11c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@ -68,9 +68,10 @@
  * - 3.16.0 - Add reserved vmid support
  * - 3.17.0 - Add AMDGPU_NUM_VRAM_CPU_PAGE_FAULTS.
  * - 3.18.0 - Export gpu always on cu bitmap
+ * - 3.19.0 - Add support for UVD MJPEG decode
  */
 #define KMS_DRIVER_MAJOR	3
-#define KMS_DRIVER_MINOR	18
+#define KMS_DRIVER_MINOR	19
 #define KMS_DRIVER_PATCHLEVEL	0
 
 int amdgpu_vram_limit = 0;
@@ -94,6 +95,7 @@ unsigned amdgpu_ip_block_mask = 0xffffffff;
 int amdgpu_bapm = -1;
 int amdgpu_deep_color = 0;
 int amdgpu_vm_size = -1;
+int amdgpu_vm_fragment_size = -1;
 int amdgpu_vm_block_size = -1;
 int amdgpu_vm_fault_stop = 0;
 int amdgpu_vm_debug = 0;
@@ -183,6 +185,9 @@ module_param_named(deep_color, amdgpu_deep_color, int, 0444);
 MODULE_PARM_DESC(vm_size, "VM address space size in gigabytes (default 64GB)");
 module_param_named(vm_size, amdgpu_vm_size, int, 0444);
 
+MODULE_PARM_DESC(vm_fragment_size, "VM fragment size in bits (4, 5, etc. 4 = 64K (default), Max 9 = 2M)");
+module_param_named(vm_fragment_size, amdgpu_vm_fragment_size, int, 0444);
+
 MODULE_PARM_DESC(vm_block_size, "VM page table size in bits (default depending on vm_size)");
 module_param_named(vm_block_size, amdgpu_vm_block_size, int, 0444);
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fb.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fb.c
index 0a8ee2411180..9afa9c097e1f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fb.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fb.c
@@ -118,7 +118,7 @@ static void amdgpufb_destroy_pinned_object(struct drm_gem_object *gobj)
 		amdgpu_bo_unpin(abo);
 		amdgpu_bo_unreserve(abo);
 	}
-	drm_gem_object_unreference_unlocked(gobj);
+	drm_gem_object_put_unlocked(gobj);
 }
 
 static int amdgpufb_create_pinned_object(struct amdgpu_fbdev *rfbdev,
@@ -250,7 +250,7 @@ static int amdgpufb_create(struct drm_fb_helper *helper,
 	tmp = amdgpu_bo_gpu_offset(abo) - adev->mc.vram_start;
 	info->fix.smem_start = adev->mc.aper_base + tmp;
 	info->fix.smem_len = amdgpu_bo_size(abo);
-	info->screen_base = abo->kptr;
+	info->screen_base = amdgpu_bo_kptr(abo);
 	info->screen_size = amdgpu_bo_size(abo);
 
 	drm_fb_helper_fill_var(info, &rfbdev->helper, sizes->fb_width, sizes->fb_height);
@@ -280,7 +280,7 @@ out:
 
 	}
 	if (fb && ret) {
-		drm_gem_object_unreference_unlocked(gobj);
+		drm_gem_object_put_unlocked(gobj);
 		drm_framebuffer_unregister_private(fb);
 		drm_framebuffer_cleanup(fb);
 		kfree(fb);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c
index 5cc4987cd887..94c1e2e8e34c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c
@@ -144,7 +144,7 @@ int amdgpu_gart_table_vram_alloc(struct amdgpu_device *adev)
 				     PAGE_SIZE, true, AMDGPU_GEM_DOMAIN_VRAM,
 				     AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED |
 				     AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS,
-				     NULL, NULL, &adev->gart.robj);
+				     NULL, NULL, 0, &adev->gart.robj);
 		if (r) {
 			return r;
 		}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
index 917ac5e074a0..7171968f261e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
@@ -59,7 +59,7 @@ int amdgpu_gem_object_create(struct amdgpu_device *adev, unsigned long size,
 
 retry:
 	r = amdgpu_bo_create(adev, size, alignment, kernel, initial_domain,
-			     flags, NULL, NULL, &robj);
+			     flags, NULL, NULL, 0, &robj);
 	if (r) {
 		if (r != -ERESTARTSYS) {
 			if (initial_domain == AMDGPU_GEM_DOMAIN_VRAM) {
@@ -91,7 +91,7 @@ void amdgpu_gem_force_release(struct amdgpu_device *adev)
 		spin_lock(&file->table_lock);
 		idr_for_each_entry(&file->object_idr, gobj, handle) {
 			WARN_ONCE(1, "And also active allocations!\n");
-			drm_gem_object_unreference_unlocked(gobj);
+			drm_gem_object_put_unlocked(gobj);
 		}
 		idr_destroy(&file->object_idr);
 		spin_unlock(&file->table_lock);
@@ -225,9 +225,7 @@ int amdgpu_gem_create_ioctl(struct drm_device *dev, void *data,
 	if (args->in.domain_flags & ~(AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED |
 				      AMDGPU_GEM_CREATE_NO_CPU_ACCESS |
 				      AMDGPU_GEM_CREATE_CPU_GTT_USWC |
-				      AMDGPU_GEM_CREATE_VRAM_CLEARED|
-				      AMDGPU_GEM_CREATE_SHADOW |
-				      AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS))
+				      AMDGPU_GEM_CREATE_VRAM_CLEARED))
 		return -EINVAL;
 
 	/* reject invalid gem domains */
@@ -263,7 +261,7 @@ int amdgpu_gem_create_ioctl(struct drm_device *dev, void *data,
 
 	r = drm_gem_handle_create(filp, gobj, &handle);
 	/* drop reference from allocate - handle holds it now */
-	drm_gem_object_unreference_unlocked(gobj);
+	drm_gem_object_put_unlocked(gobj);
 	if (r)
 		return r;
 
@@ -306,7 +304,7 @@ int amdgpu_gem_userptr_ioctl(struct drm_device *dev, void *data,
 		return r;
 
 	bo = gem_to_amdgpu_bo(gobj);
-	bo->prefered_domains = AMDGPU_GEM_DOMAIN_GTT;
+	bo->preferred_domains = AMDGPU_GEM_DOMAIN_GTT;
 	bo->allowed_domains = AMDGPU_GEM_DOMAIN_GTT;
 	r = amdgpu_ttm_tt_set_userptr(bo->tbo.ttm, args->addr, args->flags);
 	if (r)
@@ -341,7 +339,7 @@ int amdgpu_gem_userptr_ioctl(struct drm_device *dev, void *data,
 
 	r = drm_gem_handle_create(filp, gobj, &handle);
 	/* drop reference from allocate - handle holds it now */
-	drm_gem_object_unreference_unlocked(gobj);
+	drm_gem_object_put_unlocked(gobj);
 	if (r)
 		return r;
 
@@ -355,7 +353,7 @@ unlock_mmap_sem:
 	up_read(&current->mm->mmap_sem);
 
 release_object:
-	drm_gem_object_unreference_unlocked(gobj);
+	drm_gem_object_put_unlocked(gobj);
 
 	return r;
 }
@@ -374,11 +372,11 @@ int amdgpu_mode_dumb_mmap(struct drm_file *filp,
 	robj = gem_to_amdgpu_bo(gobj);
 	if (amdgpu_ttm_tt_get_usermm(robj->tbo.ttm) ||
 	    (robj->flags & AMDGPU_GEM_CREATE_NO_CPU_ACCESS)) {
-		drm_gem_object_unreference_unlocked(gobj);
+		drm_gem_object_put_unlocked(gobj);
 		return -EPERM;
 	}
 	*offset_p = amdgpu_bo_mmap_offset(robj);
-	drm_gem_object_unreference_unlocked(gobj);
+	drm_gem_object_put_unlocked(gobj);
 	return 0;
 }
 
@@ -448,7 +446,7 @@ int amdgpu_gem_wait_idle_ioctl(struct drm_device *dev, void *data,
 	} else
 		r = ret;
 
-	drm_gem_object_unreference_unlocked(gobj);
+	drm_gem_object_put_unlocked(gobj);
 	return r;
 }
 
@@ -491,7 +489,7 @@ int amdgpu_gem_metadata_ioctl(struct drm_device *dev, void *data,
 unreserve:
 	amdgpu_bo_unreserve(robj);
 out:
-	drm_gem_object_unreference_unlocked(gobj);
+	drm_gem_object_put_unlocked(gobj);
 	return r;
 }
 
@@ -623,7 +621,7 @@ int amdgpu_gem_va_ioctl(struct drm_device *dev, void *data,
 
 	switch (args->operation) {
 	case AMDGPU_VA_OP_MAP:
-		r = amdgpu_vm_alloc_pts(adev, bo_va->vm, args->va_address,
+		r = amdgpu_vm_alloc_pts(adev, bo_va->base.vm, args->va_address,
 					args->map_size);
 		if (r)
 			goto error_backoff;
@@ -643,7 +641,7 @@ int amdgpu_gem_va_ioctl(struct drm_device *dev, void *data,
 						args->map_size);
 		break;
 	case AMDGPU_VA_OP_REPLACE:
-		r = amdgpu_vm_alloc_pts(adev, bo_va->vm, args->va_address,
+		r = amdgpu_vm_alloc_pts(adev, bo_va->base.vm, args->va_address,
 					args->map_size);
 		if (r)
 			goto error_backoff;
@@ -664,7 +662,7 @@ error_backoff:
 	ttm_eu_backoff_reservation(&ticket, &list);
 
 error_unref:
-	drm_gem_object_unreference_unlocked(gobj);
+	drm_gem_object_put_unlocked(gobj);
 	return r;
 }
 
@@ -689,11 +687,11 @@ int amdgpu_gem_op_ioctl(struct drm_device *dev, void *data,
 	switch (args->op) {
 	case AMDGPU_GEM_OP_GET_GEM_CREATE_INFO: {
 		struct drm_amdgpu_gem_create_in info;
-		void __user *out = (void __user *)(uintptr_t)args->value;
+		void __user *out = u64_to_user_ptr(args->value);
 
 		info.bo_size = robj->gem_base.size;
 		info.alignment = robj->tbo.mem.page_alignment << PAGE_SHIFT;
-		info.domains = robj->prefered_domains;
+		info.domains = robj->preferred_domains;
 		info.domain_flags = robj->flags;
 		amdgpu_bo_unreserve(robj);
 		if (copy_to_user(out, &info, sizeof(info)))
@@ -711,10 +709,10 @@ int amdgpu_gem_op_ioctl(struct drm_device *dev, void *data,
 			amdgpu_bo_unreserve(robj);
 			break;
 		}
-		robj->prefered_domains = args->value & (AMDGPU_GEM_DOMAIN_VRAM |
+		robj->preferred_domains = args->value & (AMDGPU_GEM_DOMAIN_VRAM |
 							AMDGPU_GEM_DOMAIN_GTT |
 							AMDGPU_GEM_DOMAIN_CPU);
-		robj->allowed_domains = robj->prefered_domains;
+		robj->allowed_domains = robj->preferred_domains;
 		if (robj->allowed_domains == AMDGPU_GEM_DOMAIN_VRAM)
 			robj->allowed_domains |= AMDGPU_GEM_DOMAIN_GTT;
 
@@ -726,7 +724,7 @@ int amdgpu_gem_op_ioctl(struct drm_device *dev, void *data,
 	}
 
 out:
-	drm_gem_object_unreference_unlocked(gobj);
+	drm_gem_object_put_unlocked(gobj);
 	return r;
 }
 
@@ -754,7 +752,7 @@ int amdgpu_mode_dumb_create(struct drm_file *file_priv,
 
 	r = drm_gem_handle_create(file_priv, gobj, &handle);
 	/* drop reference from allocate - handle holds it now */
-	drm_gem_object_unreference_unlocked(gobj);
+	drm_gem_object_put_unlocked(gobj);
 	if (r) {
 		return r;
 	}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gtt_mgr.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gtt_mgr.c
index 5e6b90c6794f..9e05e257729f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gtt_mgr.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gtt_mgr.c
@@ -28,7 +28,7 @@
 struct amdgpu_gtt_mgr {
 	struct drm_mm mm;
 	spinlock_t lock;
-	uint64_t available;
+	atomic64_t available;
 };
 
 /**
@@ -54,7 +54,7 @@ static int amdgpu_gtt_mgr_init(struct ttm_mem_type_manager *man,
 	size = (adev->mc.gart_size >> PAGE_SHIFT) - start;
 	drm_mm_init(&mgr->mm, start, size);
 	spin_lock_init(&mgr->lock);
-	mgr->available = p_size;
+	atomic64_set(&mgr->available, p_size);
 	man->priv = mgr;
 	return 0;
 }
@@ -153,15 +153,6 @@ int amdgpu_gtt_mgr_alloc(struct ttm_mem_type_manager *man,
 	return r;
 }
 
-void amdgpu_gtt_mgr_print(struct seq_file *m, struct ttm_mem_type_manager *man)
-{
-	struct amdgpu_device *adev = amdgpu_ttm_adev(man->bdev);
-	struct amdgpu_gtt_mgr *mgr = man->priv;
-
-	seq_printf(m, "man size:%llu pages, gtt available:%llu pages, usage:%lluMB\n",
-		   man->size, mgr->available, (u64)atomic64_read(&adev->gtt_usage) >> 20);
-
-}
 /**
  * amdgpu_gtt_mgr_new - allocate a new node
  *
@@ -182,11 +173,11 @@ static int amdgpu_gtt_mgr_new(struct ttm_mem_type_manager *man,
 	int r;
 
 	spin_lock(&mgr->lock);
-	if (mgr->available < mem->num_pages) {
+	if (atomic64_read(&mgr->available) < mem->num_pages) {
 		spin_unlock(&mgr->lock);
 		return 0;
 	}
-	mgr->available -= mem->num_pages;
+	atomic64_sub(mem->num_pages, &mgr->available);
 	spin_unlock(&mgr->lock);
 
 	node = kzalloc(sizeof(*node), GFP_KERNEL);
@@ -213,9 +204,7 @@ static int amdgpu_gtt_mgr_new(struct ttm_mem_type_manager *man,
 
 	return 0;
 err_out:
-	spin_lock(&mgr->lock);
-	mgr->available += mem->num_pages;
-	spin_unlock(&mgr->lock);
+	atomic64_add(mem->num_pages, &mgr->available);
 
 	return r;
 }
@@ -242,30 +231,47 @@ static void amdgpu_gtt_mgr_del(struct ttm_mem_type_manager *man,
 	spin_lock(&mgr->lock);
 	if (node->start != AMDGPU_BO_INVALID_OFFSET)
 		drm_mm_remove_node(node);
-	mgr->available += mem->num_pages;
 	spin_unlock(&mgr->lock);
+	atomic64_add(mem->num_pages, &mgr->available);
 
 	kfree(node);
 	mem->mm_node = NULL;
 }
 
 /**
+ * amdgpu_gtt_mgr_usage - return usage of GTT domain
+ *
+ * @man: TTM memory type manager
+ *
+ * Return how many bytes are used in the GTT domain
+ */
+uint64_t amdgpu_gtt_mgr_usage(struct ttm_mem_type_manager *man)
+{
+	struct amdgpu_gtt_mgr *mgr = man->priv;
+
+	return (u64)(man->size - atomic64_read(&mgr->available)) * PAGE_SIZE;
+}
+
+/**
  * amdgpu_gtt_mgr_debug - dump VRAM table
  *
  * @man: TTM memory type manager
- * @prefix: text prefix
+ * @printer: DRM printer to use
  *
  * Dump the table content using printk.
  */
 static void amdgpu_gtt_mgr_debug(struct ttm_mem_type_manager *man,
-				  const char *prefix)
+				 struct drm_printer *printer)
 {
 	struct amdgpu_gtt_mgr *mgr = man->priv;
-	struct drm_printer p = drm_debug_printer(prefix);
 
 	spin_lock(&mgr->lock);
-	drm_mm_print(&mgr->mm, &p);
+	drm_mm_print(&mgr->mm, printer);
 	spin_unlock(&mgr->lock);
+
+	drm_printf(printer, "man size:%llu pages, gtt available:%llu pages, usage:%lluMB\n",
+		   man->size, (u64)atomic64_read(&mgr->available),
+		   amdgpu_gtt_mgr_usage(man) >> 20);
 }
 
 const struct ttm_mem_type_manager_func amdgpu_gtt_mgr_func = {
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
index 09f833255ba1..e16229000a98 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
@@ -158,7 +158,6 @@ int amdgpu_driver_load_kms(struct drm_device *dev, unsigned long flags)
 				"Error during ACPI methods call\n");
 	}
 
-	amdgpu_amdkfd_load_interface(adev);
 	amdgpu_amdkfd_device_probe(adev);
 	amdgpu_amdkfd_device_init(adev);
 
@@ -456,13 +455,13 @@ static int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file
 		ui64 = atomic64_read(&adev->num_vram_cpu_page_faults);
 		return copy_to_user(out, &ui64, min(size, 8u)) ? -EFAULT : 0;
 	case AMDGPU_INFO_VRAM_USAGE:
-		ui64 = atomic64_read(&adev->vram_usage);
+		ui64 = amdgpu_vram_mgr_usage(&adev->mman.bdev.man[TTM_PL_VRAM]);
 		return copy_to_user(out, &ui64, min(size, 8u)) ? -EFAULT : 0;
 	case AMDGPU_INFO_VIS_VRAM_USAGE:
-		ui64 = atomic64_read(&adev->vram_vis_usage);
+		ui64 = amdgpu_vram_mgr_vis_usage(&adev->mman.bdev.man[TTM_PL_VRAM]);
 		return copy_to_user(out, &ui64, min(size, 8u)) ? -EFAULT : 0;
 	case AMDGPU_INFO_GTT_USAGE:
-		ui64 = atomic64_read(&adev->gtt_usage);
+		ui64 = amdgpu_gtt_mgr_usage(&adev->mman.bdev.man[TTM_PL_TT]);
 		return copy_to_user(out, &ui64, min(size, 8u)) ? -EFAULT : 0;
 	case AMDGPU_INFO_GDS_CONFIG: {
 		struct drm_amdgpu_info_gds gds_info;
@@ -498,7 +497,8 @@ static int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file
 		mem.vram.total_heap_size = adev->mc.real_vram_size;
 		mem.vram.usable_heap_size =
 			adev->mc.real_vram_size - adev->vram_pin_size;
-		mem.vram.heap_usage = atomic64_read(&adev->vram_usage);
+		mem.vram.heap_usage =
+			amdgpu_vram_mgr_usage(&adev->mman.bdev.man[TTM_PL_VRAM]);
 		mem.vram.max_allocation = mem.vram.usable_heap_size * 3 / 4;
 
 		mem.cpu_accessible_vram.total_heap_size =
@@ -507,7 +507,7 @@ static int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file
 			adev->mc.visible_vram_size -
 			(adev->vram_pin_size - adev->invisible_pin_size);
 		mem.cpu_accessible_vram.heap_usage =
-			atomic64_read(&adev->vram_vis_usage);
+			amdgpu_vram_mgr_vis_usage(&adev->mman.bdev.man[TTM_PL_VRAM]);
 		mem.cpu_accessible_vram.max_allocation =
 			mem.cpu_accessible_vram.usable_heap_size * 3 / 4;
 
@@ -515,7 +515,8 @@ static int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file
 		mem.gtt.total_heap_size *= PAGE_SIZE;
 		mem.gtt.usable_heap_size = mem.gtt.total_heap_size
 			- adev->gart_pin_size;
-		mem.gtt.heap_usage = atomic64_read(&adev->gtt_usage);
+		mem.gtt.heap_usage =
+			amdgpu_gtt_mgr_usage(&adev->mman.bdev.man[TTM_PL_TT]);
 		mem.gtt.max_allocation = mem.gtt.usable_heap_size * 3 / 4;
 
 		return copy_to_user(out, &mem,
@@ -589,11 +590,8 @@ static int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file
 		dev_info.virtual_address_offset = AMDGPU_VA_RESERVED_SIZE;
 		dev_info.virtual_address_max = (uint64_t)adev->vm_manager.max_pfn * AMDGPU_GPU_PAGE_SIZE;
 		dev_info.virtual_address_alignment = max((int)PAGE_SIZE, AMDGPU_GPU_PAGE_SIZE);
-		dev_info.pte_fragment_size =
-			(1 << AMDGPU_LOG2_PAGES_PER_FRAG(adev)) *
-			AMDGPU_GPU_PAGE_SIZE;
+		dev_info.pte_fragment_size = (1 << adev->vm_manager.fragment_size) * AMDGPU_GPU_PAGE_SIZE;
 		dev_info.gart_page_size = AMDGPU_GPU_PAGE_SIZE;
-
 		dev_info.cu_active_number = adev->gfx.cu_info.number;
 		dev_info.cu_ao_mask = adev->gfx.cu_info.ao_cu_mask;
 		dev_info.ce_ram_size = adev->gfx.ce_ram_size;
@@ -842,7 +840,7 @@ int amdgpu_driver_open_kms(struct drm_device *dev, struct drm_file *file_priv)
 	}
 
 	if (amdgpu_sriov_vf(adev)) {
-		r = amdgpu_map_static_csa(adev, &fpriv->vm);
+		r = amdgpu_map_static_csa(adev, &fpriv->vm, &fpriv->csa_va);
 		if (r)
 			goto out_suspend;
 	}
@@ -895,8 +893,8 @@ void amdgpu_driver_postclose_kms(struct drm_device *dev,
 	if (amdgpu_sriov_vf(adev)) {
 		/* TODO: how to handle reserve failure */
 		BUG_ON(amdgpu_bo_reserve(adev->virt.csa_obj, true));
-		amdgpu_vm_bo_rmv(adev, fpriv->vm.csa_bo_va);
-		fpriv->vm.csa_bo_va = NULL;
+		amdgpu_vm_bo_rmv(adev, fpriv->csa_va);
+		fpriv->csa_va = NULL;
 		amdgpu_bo_unreserve(adev->virt.csa_obj);
 	}
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mn.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_mn.c
index 38f739fb727b..6558a3ed57a7 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mn.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mn.c
@@ -359,7 +359,7 @@ void amdgpu_mn_unregister(struct amdgpu_bo *bo)
 	head = bo->mn_list.next;
 
 	bo->mn = NULL;
-	list_del(&bo->mn_list);
+	list_del_init(&bo->mn_list);
 
 	if (list_empty(head)) {
 		struct amdgpu_mn_node *node;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
index 3ec43cf9ad78..e7e899190bef 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
@@ -37,55 +37,6 @@
 #include "amdgpu.h"
 #include "amdgpu_trace.h"
 
-
-
-static u64 amdgpu_get_vis_part_size(struct amdgpu_device *adev,
-						struct ttm_mem_reg *mem)
-{
-	if (mem->start << PAGE_SHIFT >= adev->mc.visible_vram_size)
-		return 0;
-
-	return ((mem->start << PAGE_SHIFT) + mem->size) >
-		adev->mc.visible_vram_size ?
-		adev->mc.visible_vram_size - (mem->start << PAGE_SHIFT) :
-		mem->size;
-}
-
-static void amdgpu_update_memory_usage(struct amdgpu_device *adev,
-		       struct ttm_mem_reg *old_mem,
-		       struct ttm_mem_reg *new_mem)
-{
-	u64 vis_size;
-	if (!adev)
-		return;
-
-	if (new_mem) {
-		switch (new_mem->mem_type) {
-		case TTM_PL_TT:
-			atomic64_add(new_mem->size, &adev->gtt_usage);
-			break;
-		case TTM_PL_VRAM:
-			atomic64_add(new_mem->size, &adev->vram_usage);
-			vis_size = amdgpu_get_vis_part_size(adev, new_mem);
-			atomic64_add(vis_size, &adev->vram_vis_usage);
-			break;
-		}
-	}
-
-	if (old_mem) {
-		switch (old_mem->mem_type) {
-		case TTM_PL_TT:
-			atomic64_sub(old_mem->size, &adev->gtt_usage);
-			break;
-		case TTM_PL_VRAM:
-			atomic64_sub(old_mem->size, &adev->vram_usage);
-			vis_size = amdgpu_get_vis_part_size(adev, old_mem);
-			atomic64_sub(vis_size, &adev->vram_vis_usage);
-			break;
-		}
-	}
-}
-
 static void amdgpu_ttm_bo_destroy(struct ttm_buffer_object *tbo)
 {
 	struct amdgpu_device *adev = amdgpu_ttm_adev(tbo->bdev);
@@ -94,7 +45,6 @@ static void amdgpu_ttm_bo_destroy(struct ttm_buffer_object *tbo)
 	bo = container_of(tbo, struct amdgpu_bo, tbo);
 
 	amdgpu_bo_kunmap(bo);
-	amdgpu_update_memory_usage(adev, &bo->tbo.mem, NULL);
 
 	drm_gem_object_release(&bo->gem_base);
 	amdgpu_bo_unref(&bo->parent);
@@ -220,7 +170,7 @@ static void amdgpu_fill_placement_to_bo(struct amdgpu_bo *bo,
 }
 
 /**
- * amdgpu_bo_create_kernel - create BO for kernel use
+ * amdgpu_bo_create_reserved - create reserved BO for kernel use
  *
  * @adev: amdgpu device object
  * @size: size for the new BO
@@ -230,24 +180,30 @@ static void amdgpu_fill_placement_to_bo(struct amdgpu_bo *bo,
  * @gpu_addr: GPU addr of the pinned BO
  * @cpu_addr: optional CPU address mapping
  *
- * Allocates and pins a BO for kernel internal use.
+ * Allocates and pins a BO for kernel internal use, and returns it still
+ * reserved.
  *
  * Returns 0 on success, negative error code otherwise.
  */
-int amdgpu_bo_create_kernel(struct amdgpu_device *adev,
-			    unsigned long size, int align,
-			    u32 domain, struct amdgpu_bo **bo_ptr,
-			    u64 *gpu_addr, void **cpu_addr)
+int amdgpu_bo_create_reserved(struct amdgpu_device *adev,
+			      unsigned long size, int align,
+			      u32 domain, struct amdgpu_bo **bo_ptr,
+			      u64 *gpu_addr, void **cpu_addr)
 {
+	bool free = false;
 	int r;
 
-	r = amdgpu_bo_create(adev, size, align, true, domain,
-			     AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED |
-			     AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS,
-			     NULL, NULL, bo_ptr);
-	if (r) {
-		dev_err(adev->dev, "(%d) failed to allocate kernel bo\n", r);
-		return r;
+	if (!*bo_ptr) {
+		r = amdgpu_bo_create(adev, size, align, true, domain,
+				     AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED |
+				     AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS,
+				     NULL, NULL, 0, bo_ptr);
+		if (r) {
+			dev_err(adev->dev, "(%d) failed to allocate kernel bo\n",
+				r);
+			return r;
+		}
+		free = true;
 	}
 
 	r = amdgpu_bo_reserve(*bo_ptr, false);
@@ -270,20 +226,52 @@ int amdgpu_bo_create_kernel(struct amdgpu_device *adev,
 		}
 	}
 
-	amdgpu_bo_unreserve(*bo_ptr);
-
 	return 0;
 
 error_unreserve:
 	amdgpu_bo_unreserve(*bo_ptr);
 
 error_free:
-	amdgpu_bo_unref(bo_ptr);
+	if (free)
+		amdgpu_bo_unref(bo_ptr);
 
 	return r;
 }
 
 /**
+ * amdgpu_bo_create_kernel - create BO for kernel use
+ *
+ * @adev: amdgpu device object
+ * @size: size for the new BO
+ * @align: alignment for the new BO
+ * @domain: where to place it
+ * @bo_ptr: resulting BO
+ * @gpu_addr: GPU addr of the pinned BO
+ * @cpu_addr: optional CPU address mapping
+ *
+ * Allocates and pins a BO for kernel internal use.
+ *
+ * Returns 0 on success, negative error code otherwise.
+ */
+int amdgpu_bo_create_kernel(struct amdgpu_device *adev,
+			    unsigned long size, int align,
+			    u32 domain, struct amdgpu_bo **bo_ptr,
+			    u64 *gpu_addr, void **cpu_addr)
+{
+	int r;
+
+	r = amdgpu_bo_create_reserved(adev, size, align, domain, bo_ptr,
+				      gpu_addr, cpu_addr);
+
+	if (r)
+		return r;
+
+	amdgpu_bo_unreserve(*bo_ptr);
+
+	return 0;
+}
+
+/**
  * amdgpu_bo_free_kernel - free BO for kernel use
  *
  * @bo: amdgpu BO to free
@@ -318,6 +306,7 @@ int amdgpu_bo_create_restricted(struct amdgpu_device *adev,
 				struct sg_table *sg,
 				struct ttm_placement *placement,
 				struct reservation_object *resv,
+				uint64_t init_value,
 				struct amdgpu_bo **bo_ptr)
 {
 	struct amdgpu_bo *bo;
@@ -352,13 +341,13 @@ int amdgpu_bo_create_restricted(struct amdgpu_device *adev,
 	}
 	INIT_LIST_HEAD(&bo->shadow_list);
 	INIT_LIST_HEAD(&bo->va);
-	bo->prefered_domains = domain & (AMDGPU_GEM_DOMAIN_VRAM |
+	bo->preferred_domains = domain & (AMDGPU_GEM_DOMAIN_VRAM |
 					 AMDGPU_GEM_DOMAIN_GTT |
 					 AMDGPU_GEM_DOMAIN_CPU |
 					 AMDGPU_GEM_DOMAIN_GDS |
 					 AMDGPU_GEM_DOMAIN_GWS |
 					 AMDGPU_GEM_DOMAIN_OA);
-	bo->allowed_domains = bo->prefered_domains;
+	bo->allowed_domains = bo->preferred_domains;
 	if (!kernel && bo->allowed_domains == AMDGPU_GEM_DOMAIN_VRAM)
 		bo->allowed_domains |= AMDGPU_GEM_DOMAIN_GTT;
 
@@ -418,7 +407,7 @@ int amdgpu_bo_create_restricted(struct amdgpu_device *adev,
 	    bo->tbo.mem.placement & TTM_PL_FLAG_VRAM) {
 		struct dma_fence *fence;
 
-		r = amdgpu_fill_buffer(bo, 0, bo->tbo.resv, &fence);
+		r = amdgpu_fill_buffer(bo, init_value, bo->tbo.resv, &fence);
 		if (unlikely(r))
 			goto fail_unreserve;
 
@@ -470,6 +459,7 @@ static int amdgpu_bo_create_shadow(struct amdgpu_device *adev,
 					AMDGPU_GEM_CREATE_CPU_GTT_USWC,
 					NULL, &placement,
 					bo->tbo.resv,
+					0,
 					&bo->shadow);
 	if (!r) {
 		bo->shadow->parent = amdgpu_bo_ref(bo);
@@ -481,11 +471,15 @@ static int amdgpu_bo_create_shadow(struct amdgpu_device *adev,
 	return r;
 }
 
+/* init_value will only take effect when flags contains
+ * AMDGPU_GEM_CREATE_VRAM_CLEARED.
+ */
 int amdgpu_bo_create(struct amdgpu_device *adev,
 		     unsigned long size, int byte_align,
 		     bool kernel, u32 domain, u64 flags,
 		     struct sg_table *sg,
 		     struct reservation_object *resv,
+		     uint64_t init_value,
 		     struct amdgpu_bo **bo_ptr)
 {
 	struct ttm_placement placement = {0};
@@ -500,7 +494,7 @@ int amdgpu_bo_create(struct amdgpu_device *adev,
 
 	r = amdgpu_bo_create_restricted(adev, size, byte_align, kernel,
 					domain, flags, sg, &placement,
-					resv, bo_ptr);
+					resv, init_value, bo_ptr);
 	if (r)
 		return r;
 
@@ -562,7 +556,7 @@ int amdgpu_bo_validate(struct amdgpu_bo *bo)
 	if (bo->pin_count)
 		return 0;
 
-	domain = bo->prefered_domains;
+	domain = bo->preferred_domains;
 
 retry:
 	amdgpu_ttm_placement_from_domain(bo, domain);
@@ -609,16 +603,16 @@ err:
 
 int amdgpu_bo_kmap(struct amdgpu_bo *bo, void **ptr)
 {
-	bool is_iomem;
+	void *kptr;
 	long r;
 
 	if (bo->flags & AMDGPU_GEM_CREATE_NO_CPU_ACCESS)
 		return -EPERM;
 
-	if (bo->kptr) {
-		if (ptr) {
-			*ptr = bo->kptr;
-		}
+	kptr = amdgpu_bo_kptr(bo);
+	if (kptr) {
+		if (ptr)
+			*ptr = kptr;
 		return 0;
 	}
 
@@ -631,19 +625,23 @@ int amdgpu_bo_kmap(struct amdgpu_bo *bo, void **ptr)
 	if (r)
 		return r;
 
-	bo->kptr = ttm_kmap_obj_virtual(&bo->kmap, &is_iomem);
 	if (ptr)
-		*ptr = bo->kptr;
+		*ptr = amdgpu_bo_kptr(bo);
 
 	return 0;
 }
 
+void *amdgpu_bo_kptr(struct amdgpu_bo *bo)
+{
+	bool is_iomem;
+
+	return ttm_kmap_obj_virtual(&bo->kmap, &is_iomem);
+}
+
 void amdgpu_bo_kunmap(struct amdgpu_bo *bo)
 {
-	if (bo->kptr == NULL)
-		return;
-	bo->kptr = NULL;
-	ttm_bo_kunmap(&bo->kmap);
+	if (bo->kmap.bo)
+		ttm_bo_kunmap(&bo->kmap);
 }
 
 struct amdgpu_bo *amdgpu_bo_ref(struct amdgpu_bo *bo)
@@ -944,8 +942,6 @@ void amdgpu_bo_move_notify(struct ttm_buffer_object *bo,
 		return;
 
 	/* move_notify is called before move happens */
-	amdgpu_update_memory_usage(adev, &bo->mem, new_mem);
-
 	trace_amdgpu_ttm_bo_move(abo, new_mem->mem_type, old_mem->mem_type);
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h
index 833b172a2c2a..a288fa6d72c8 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h
@@ -33,6 +33,61 @@
 
 #define AMDGPU_BO_INVALID_OFFSET	LONG_MAX
 
+/* bo virtual addresses in a vm */
+struct amdgpu_bo_va_mapping {
+	struct list_head		list;
+	struct rb_node			rb;
+	uint64_t			start;
+	uint64_t			last;
+	uint64_t			__subtree_last;
+	uint64_t			offset;
+	uint64_t			flags;
+};
+
+/* User space allocated BO in a VM */
+struct amdgpu_bo_va {
+	struct amdgpu_vm_bo_base	base;
+
+	/* protected by bo being reserved */
+	struct dma_fence	        *last_pt_update;
+	unsigned			ref_count;
+
+	/* mappings for this bo_va */
+	struct list_head		invalids;
+	struct list_head		valids;
+};
+
+struct amdgpu_bo {
+	/* Protected by tbo.reserved */
+	u32				preferred_domains;
+	u32				allowed_domains;
+	struct ttm_place		placements[AMDGPU_GEM_DOMAIN_MAX + 1];
+	struct ttm_placement		placement;
+	struct ttm_buffer_object	tbo;
+	struct ttm_bo_kmap_obj		kmap;
+	u64				flags;
+	unsigned			pin_count;
+	u64				tiling_flags;
+	u64				metadata_flags;
+	void				*metadata;
+	u32				metadata_size;
+	unsigned			prime_shared_count;
+	/* list of all virtual address to which this bo is associated to */
+	struct list_head		va;
+	/* Constant after initialization */
+	struct drm_gem_object		gem_base;
+	struct amdgpu_bo		*parent;
+	struct amdgpu_bo		*shadow;
+
+	struct ttm_bo_kmap_obj		dma_buf_vmap;
+	struct amdgpu_mn		*mn;
+
+	union {
+		struct list_head	mn_list;
+		struct list_head	shadow_list;
+	};
+};
+
 /**
  * amdgpu_mem_type_to_domain - return domain corresponding to mem_type
  * @mem_type:	ttm memory type
@@ -132,6 +187,7 @@ int amdgpu_bo_create(struct amdgpu_device *adev,
 			    bool kernel, u32 domain, u64 flags,
 			    struct sg_table *sg,
 			    struct reservation_object *resv,
+			    uint64_t init_value,
 			    struct amdgpu_bo **bo_ptr);
 int amdgpu_bo_create_restricted(struct amdgpu_device *adev,
 				unsigned long size, int byte_align,
@@ -139,7 +195,12 @@ int amdgpu_bo_create_restricted(struct amdgpu_device *adev,
 				struct sg_table *sg,
 				struct ttm_placement *placement,
 			        struct reservation_object *resv,
+				uint64_t init_value,
 				struct amdgpu_bo **bo_ptr);
+int amdgpu_bo_create_reserved(struct amdgpu_device *adev,
+			      unsigned long size, int align,
+			      u32 domain, struct amdgpu_bo **bo_ptr,
+			      u64 *gpu_addr, void **cpu_addr);
 int amdgpu_bo_create_kernel(struct amdgpu_device *adev,
 			    unsigned long size, int align,
 			    u32 domain, struct amdgpu_bo **bo_ptr,
@@ -147,6 +208,7 @@ int amdgpu_bo_create_kernel(struct amdgpu_device *adev,
 void amdgpu_bo_free_kernel(struct amdgpu_bo **bo, u64 *gpu_addr,
 			   void **cpu_addr);
 int amdgpu_bo_kmap(struct amdgpu_bo *bo, void **ptr);
+void *amdgpu_bo_kptr(struct amdgpu_bo *bo);
 void amdgpu_bo_kunmap(struct amdgpu_bo *bo);
 struct amdgpu_bo *amdgpu_bo_ref(struct amdgpu_bo *bo);
 void amdgpu_bo_unref(struct amdgpu_bo **bo);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.h
index c19c4d138751..f21a7716b90e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.h
@@ -30,6 +30,7 @@ struct cg_flag_name
 	const char *name;
 };
 
+void amdgpu_pm_acpi_event_handler(struct amdgpu_device *adev);
 int amdgpu_pm_sysfs_init(struct amdgpu_device *adev);
 void amdgpu_pm_sysfs_fini(struct amdgpu_device *adev);
 void amdgpu_pm_print_power_states(struct amdgpu_device *adev);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_prime.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_prime.c
index 6bdc866570ab..5b3f92891f89 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_prime.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_prime.c
@@ -69,7 +69,7 @@ amdgpu_gem_prime_import_sg_table(struct drm_device *dev,
 
 	ww_mutex_lock(&resv->lock, NULL);
 	ret = amdgpu_bo_create(adev, attach->dmabuf->size, PAGE_SIZE, false,
-			       AMDGPU_GEM_DOMAIN_GTT, 0, sg, resv, &bo);
+			       AMDGPU_GEM_DOMAIN_GTT, 0, sg, resv, 0, &bo);
 	ww_mutex_unlock(&resv->lock);
 	if (ret)
 		return ERR_PTR(ret);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
index 15b7149d1204..6c5646b48d1a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
@@ -184,47 +184,22 @@ int amdgpu_ring_init(struct amdgpu_device *adev, struct amdgpu_ring *ring,
 			return r;
 	}
 
-	if (ring->funcs->support_64bit_ptrs) {
-		r = amdgpu_wb_get_64bit(adev, &ring->rptr_offs);
-		if (r) {
-			dev_err(adev->dev, "(%d) ring rptr_offs wb alloc failed\n", r);
-			return r;
-		}
-
-		r = amdgpu_wb_get_64bit(adev, &ring->wptr_offs);
-		if (r) {
-			dev_err(adev->dev, "(%d) ring wptr_offs wb alloc failed\n", r);
-			return r;
-		}
-
-	} else {
-		r = amdgpu_wb_get(adev, &ring->rptr_offs);
-		if (r) {
-			dev_err(adev->dev, "(%d) ring rptr_offs wb alloc failed\n", r);
-			return r;
-		}
-
-		r = amdgpu_wb_get(adev, &ring->wptr_offs);
-		if (r) {
-			dev_err(adev->dev, "(%d) ring wptr_offs wb alloc failed\n", r);
-			return r;
-		}
-
+	r = amdgpu_wb_get(adev, &ring->rptr_offs);
+	if (r) {
+		dev_err(adev->dev, "(%d) ring rptr_offs wb alloc failed\n", r);
+		return r;
 	}
 
-	if (amdgpu_sriov_vf(adev) && ring->funcs->type == AMDGPU_RING_TYPE_GFX) {
-		r = amdgpu_wb_get_256Bit(adev, &ring->fence_offs);
-		if (r) {
-			dev_err(adev->dev, "(%d) ring fence_offs wb alloc failed\n", r);
-			return r;
-		}
+	r = amdgpu_wb_get(adev, &ring->wptr_offs);
+	if (r) {
+		dev_err(adev->dev, "(%d) ring wptr_offs wb alloc failed\n", r);
+		return r;
+	}
 
-	} else {
-		r = amdgpu_wb_get(adev, &ring->fence_offs);
-		if (r) {
-			dev_err(adev->dev, "(%d) ring fence_offs wb alloc failed\n", r);
-			return r;
-		}
+	r = amdgpu_wb_get(adev, &ring->fence_offs);
+	if (r) {
+		dev_err(adev->dev, "(%d) ring fence_offs wb alloc failed\n", r);
+		return r;
 	}
 
 	r = amdgpu_wb_get(adev, &ring->cond_exe_offs);
@@ -286,19 +261,15 @@ void amdgpu_ring_fini(struct amdgpu_ring *ring)
 {
 	ring->ready = false;
 
-	if (ring->funcs->support_64bit_ptrs) {
-		amdgpu_wb_free_64bit(ring->adev, ring->rptr_offs);
-		amdgpu_wb_free_64bit(ring->adev, ring->wptr_offs);
-	} else {
-		amdgpu_wb_free(ring->adev, ring->rptr_offs);
-		amdgpu_wb_free(ring->adev, ring->wptr_offs);
-	}
+	/* Not to finish a ring which is not initialized */
+	if (!(ring->adev) || !(ring->adev->rings[ring->idx]))
+		return;
+
+	amdgpu_wb_free(ring->adev, ring->rptr_offs);
+	amdgpu_wb_free(ring->adev, ring->wptr_offs);
 
 	amdgpu_wb_free(ring->adev, ring->cond_exe_offs);
-	if (amdgpu_sriov_vf(ring->adev) && ring->funcs->type == AMDGPU_RING_TYPE_GFX)
-		amdgpu_wb_free_256bit(ring->adev, ring->fence_offs);
-	else
-		amdgpu_wb_free(ring->adev, ring->fence_offs);
+	amdgpu_wb_free(ring->adev, ring->fence_offs);
 
 	amdgpu_bo_free_kernel(&ring->ring_obj,
 			      &ring->gpu_addr,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sa.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sa.c
index 5ca75a456ad2..3144400435b7 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sa.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sa.c
@@ -64,7 +64,7 @@ int amdgpu_sa_bo_manager_init(struct amdgpu_device *adev,
 		INIT_LIST_HEAD(&sa_manager->flist[i]);
 
 	r = amdgpu_bo_create(adev, size, align, true, domain,
-			     0, NULL, NULL, &sa_manager->bo);
+			     0, NULL, NULL, 0, &sa_manager->bo);
 	if (r) {
 		dev_err(adev->dev, "(%d) failed to allocate bo for manager\n", r);
 		return r;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c
index a6899180b265..c586f44312f9 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c
@@ -244,6 +244,12 @@ struct dma_fence *amdgpu_sync_peek_fence(struct amdgpu_sync *sync,
 		struct dma_fence *f = e->fence;
 		struct amd_sched_fence *s_fence = to_amd_sched_fence(f);
 
+		if (dma_fence_is_signaled(f)) {
+			hash_del(&e->node);
+			dma_fence_put(f);
+			kmem_cache_free(amdgpu_sync_slab, e);
+			continue;
+		}
 		if (ring && s_fence) {
 			/* For fences from the same ring it is sufficient
 			 * when they are scheduled.
@@ -256,13 +262,6 @@ struct dma_fence *amdgpu_sync_peek_fence(struct amdgpu_sync *sync,
 			}
 		}
 
-		if (dma_fence_is_signaled(f)) {
-			hash_del(&e->node);
-			dma_fence_put(f);
-			kmem_cache_free(amdgpu_sync_slab, e);
-			continue;
-		}
-
 		return f;
 	}
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_test.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_test.c
index 3c4d7574d704..ed8c3739015b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_test.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_test.c
@@ -61,7 +61,7 @@ static void amdgpu_do_test_moves(struct amdgpu_device *adev)
 
 	r = amdgpu_bo_create(adev, size, PAGE_SIZE, true,
 			     AMDGPU_GEM_DOMAIN_VRAM, 0,
-			     NULL, NULL, &vram_obj);
+			     NULL, NULL, 0, &vram_obj);
 	if (r) {
 		DRM_ERROR("Failed to create VRAM object\n");
 		goto out_cleanup;
@@ -82,7 +82,7 @@ static void amdgpu_do_test_moves(struct amdgpu_device *adev)
 
 		r = amdgpu_bo_create(adev, size, PAGE_SIZE, true,
 				     AMDGPU_GEM_DOMAIN_GTT, 0, NULL,
-				     NULL, gtt_obj + i);
+				     NULL, 0, gtt_obj + i);
 		if (r) {
 			DRM_ERROR("Failed to create GTT object %d\n", i);
 			goto out_lclean;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h
index 509f7a63d40c..1c88bd5e29ad 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h
@@ -14,6 +14,62 @@
 #define AMDGPU_JOB_GET_TIMELINE_NAME(job) \
 	 job->base.s_fence->finished.ops->get_timeline_name(&job->base.s_fence->finished)
 
+TRACE_EVENT(amdgpu_ttm_tt_populate,
+	    TP_PROTO(struct amdgpu_device *adev, uint64_t dma_address, uint64_t phys_address),
+	    TP_ARGS(adev, dma_address, phys_address),
+	    TP_STRUCT__entry(
+				__field(uint16_t, domain)
+				__field(uint8_t, bus)
+				__field(uint8_t, slot)
+				__field(uint8_t, func)
+				__field(uint64_t, dma)
+				__field(uint64_t, phys)
+			    ),
+	    TP_fast_assign(
+			   __entry->domain = pci_domain_nr(adev->pdev->bus);
+			   __entry->bus = adev->pdev->bus->number;
+			   __entry->slot = PCI_SLOT(adev->pdev->devfn);
+			   __entry->func = PCI_FUNC(adev->pdev->devfn);
+			   __entry->dma = dma_address;
+			   __entry->phys = phys_address;
+			   ),
+	    TP_printk("%04x:%02x:%02x.%x: 0x%llx => 0x%llx",
+		      (unsigned)__entry->domain,
+		      (unsigned)__entry->bus,
+		      (unsigned)__entry->slot,
+		      (unsigned)__entry->func,
+		      (unsigned long long)__entry->dma,
+		      (unsigned long long)__entry->phys)
+);
+
+TRACE_EVENT(amdgpu_ttm_tt_unpopulate,
+	    TP_PROTO(struct amdgpu_device *adev, uint64_t dma_address, uint64_t phys_address),
+	    TP_ARGS(adev, dma_address, phys_address),
+	    TP_STRUCT__entry(
+				__field(uint16_t, domain)
+				__field(uint8_t, bus)
+				__field(uint8_t, slot)
+				__field(uint8_t, func)
+				__field(uint64_t, dma)
+				__field(uint64_t, phys)
+			    ),
+	    TP_fast_assign(
+			   __entry->domain = pci_domain_nr(adev->pdev->bus);
+			   __entry->bus = adev->pdev->bus->number;
+			   __entry->slot = PCI_SLOT(adev->pdev->devfn);
+			   __entry->func = PCI_FUNC(adev->pdev->devfn);
+			   __entry->dma = dma_address;
+			   __entry->phys = phys_address;
+			   ),
+	    TP_printk("%04x:%02x:%02x.%x: 0x%llx => 0x%llx",
+		      (unsigned)__entry->domain,
+		      (unsigned)__entry->bus,
+		      (unsigned)__entry->slot,
+		      (unsigned)__entry->func,
+		      (unsigned long long)__entry->dma,
+		      (unsigned long long)__entry->phys)
+);
+
 TRACE_EVENT(amdgpu_mm_rreg,
 	    TP_PROTO(unsigned did, uint32_t reg, uint32_t value),
 	    TP_ARGS(did, reg, value),
@@ -105,12 +161,12 @@ TRACE_EVENT(amdgpu_bo_create,
 			   __entry->bo = bo;
 			   __entry->pages = bo->tbo.num_pages;
 			   __entry->type = bo->tbo.mem.mem_type;
-			   __entry->prefer = bo->prefered_domains;
+			   __entry->prefer = bo->preferred_domains;
 			   __entry->allow = bo->allowed_domains;
 			   __entry->visible = bo->flags;
 			   ),
 
-	    TP_printk("bo=%p, pages=%u, type=%d, prefered=%d, allowed=%d, visible=%d",
+	    TP_printk("bo=%p, pages=%u, type=%d, preferred=%d, allowed=%d, visible=%d",
 		       __entry->bo, __entry->pages, __entry->type,
 		       __entry->prefer, __entry->allow, __entry->visible)
 );
@@ -228,7 +284,7 @@ TRACE_EVENT(amdgpu_vm_bo_map,
 			     ),
 
 	    TP_fast_assign(
-			   __entry->bo = bo_va ? bo_va->bo : NULL;
+			   __entry->bo = bo_va ? bo_va->base.bo : NULL;
 			   __entry->start = mapping->start;
 			   __entry->last = mapping->last;
 			   __entry->offset = mapping->offset;
@@ -252,7 +308,7 @@ TRACE_EVENT(amdgpu_vm_bo_unmap,
 			     ),
 
 	    TP_fast_assign(
-			   __entry->bo = bo_va->bo;
+			   __entry->bo = bo_va->base.bo;
 			   __entry->start = mapping->start;
 			   __entry->last = mapping->last;
 			   __entry->offset = mapping->offset;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
index e6f9a54c959d..8b2c294f6f79 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
@@ -43,6 +43,7 @@
 #include <linux/pagemap.h>
 #include <linux/debugfs.h>
 #include "amdgpu.h"
+#include "amdgpu_trace.h"
 #include "bif/bif_4_1_d.h"
 
 #define DRM_FILE_PAGE_OFFSET (0x100000000ULL >> PAGE_SHIFT)
@@ -662,6 +663,38 @@ release_pages:
 	return r;
 }
 
+static void amdgpu_trace_dma_map(struct ttm_tt *ttm)
+{
+	struct amdgpu_device *adev = amdgpu_ttm_adev(ttm->bdev);
+	struct amdgpu_ttm_tt *gtt = (void *)ttm;
+	unsigned i;
+
+	if (unlikely(trace_amdgpu_ttm_tt_populate_enabled())) {
+		for (i = 0; i < ttm->num_pages; i++) {
+			trace_amdgpu_ttm_tt_populate(
+				adev,
+				gtt->ttm.dma_address[i],
+				page_to_phys(ttm->pages[i]));
+		}
+	}
+}
+
+static void amdgpu_trace_dma_unmap(struct ttm_tt *ttm)
+{
+	struct amdgpu_device *adev = amdgpu_ttm_adev(ttm->bdev);
+	struct amdgpu_ttm_tt *gtt = (void *)ttm;
+	unsigned i;
+
+	if (unlikely(trace_amdgpu_ttm_tt_unpopulate_enabled())) {
+		for (i = 0; i < ttm->num_pages; i++) {
+			trace_amdgpu_ttm_tt_unpopulate(
+				adev,
+				gtt->ttm.dma_address[i],
+				page_to_phys(ttm->pages[i]));
+		}
+	}
+}
+
 /* prepare the sg table with the user pages */
 static int amdgpu_ttm_tt_pin_userptr(struct ttm_tt *ttm)
 {
@@ -688,6 +721,8 @@ static int amdgpu_ttm_tt_pin_userptr(struct ttm_tt *ttm)
 	drm_prime_sg_to_page_addr_arrays(ttm->sg, ttm->pages,
 					 gtt->ttm.dma_address, ttm->num_pages);
 
+	amdgpu_trace_dma_map(ttm);
+
 	return 0;
 
 release_sg:
@@ -721,6 +756,8 @@ static void amdgpu_ttm_tt_unpin_userptr(struct ttm_tt *ttm)
 		put_page(page);
 	}
 
+	amdgpu_trace_dma_unmap(ttm);
+
 	sg_free_table(ttm->sg);
 }
 
@@ -753,7 +790,7 @@ static int amdgpu_ttm_backend_bind(struct ttm_tt *ttm,
 				   struct ttm_mem_reg *bo_mem)
 {
 	struct amdgpu_ttm_tt *gtt = (void*)ttm;
-	int r;
+	int r = 0;
 
 	if (gtt->userptr) {
 		r = amdgpu_ttm_tt_pin_userptr(ttm);
@@ -892,7 +929,7 @@ static struct ttm_tt *amdgpu_ttm_tt_create(struct ttm_bo_device *bdev,
 
 static int amdgpu_ttm_tt_populate(struct ttm_tt *ttm)
 {
-	struct amdgpu_device *adev;
+	struct amdgpu_device *adev = amdgpu_ttm_adev(ttm->bdev);
 	struct amdgpu_ttm_tt *gtt = (void *)ttm;
 	unsigned i;
 	int r;
@@ -915,14 +952,14 @@ static int amdgpu_ttm_tt_populate(struct ttm_tt *ttm)
 		drm_prime_sg_to_page_addr_arrays(ttm->sg, ttm->pages,
 						 gtt->ttm.dma_address, ttm->num_pages);
 		ttm->state = tt_unbound;
-		return 0;
+		r = 0;
+		goto trace_mappings;
 	}
 
-	adev = amdgpu_ttm_adev(ttm->bdev);
-
 #ifdef CONFIG_SWIOTLB
 	if (swiotlb_nr_tbl()) {
-		return ttm_dma_populate(&gtt->ttm, adev->dev);
+		r = ttm_dma_populate(&gtt->ttm, adev->dev);
+		goto trace_mappings;
 	}
 #endif
 
@@ -945,7 +982,12 @@ static int amdgpu_ttm_tt_populate(struct ttm_tt *ttm)
 			return -EFAULT;
 		}
 	}
-	return 0;
+
+	r = 0;
+trace_mappings:
+	if (likely(!r))
+		amdgpu_trace_dma_map(ttm);
+	return r;
 }
 
 static void amdgpu_ttm_tt_unpopulate(struct ttm_tt *ttm)
@@ -966,6 +1008,8 @@ static void amdgpu_ttm_tt_unpopulate(struct ttm_tt *ttm)
 
 	adev = amdgpu_ttm_adev(ttm->bdev);
 
+	amdgpu_trace_dma_unmap(ttm);
+
 #ifdef CONFIG_SWIOTLB
 	if (swiotlb_nr_tbl()) {
 		ttm_dma_unpopulate(&gtt->ttm, adev->dev);
@@ -1232,23 +1276,12 @@ int amdgpu_ttm_init(struct amdgpu_device *adev)
 	/* Change the size here instead of the init above so only lpfn is affected */
 	amdgpu_ttm_set_active_vram_size(adev, adev->mc.visible_vram_size);
 
-	r = amdgpu_bo_create(adev, adev->mc.stolen_size, PAGE_SIZE, true,
-			     AMDGPU_GEM_DOMAIN_VRAM,
-			     AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED |
-			     AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS,
-			     NULL, NULL, &adev->stollen_vga_memory);
-	if (r) {
-		return r;
-	}
-	r = amdgpu_bo_reserve(adev->stollen_vga_memory, false);
+	r = amdgpu_bo_create_kernel(adev, adev->mc.stolen_size, PAGE_SIZE,
+				    AMDGPU_GEM_DOMAIN_VRAM,
+				    &adev->stolen_vga_memory,
+				    NULL, NULL);
 	if (r)
 		return r;
-	r = amdgpu_bo_pin(adev->stollen_vga_memory, AMDGPU_GEM_DOMAIN_VRAM, NULL);
-	amdgpu_bo_unreserve(adev->stollen_vga_memory);
-	if (r) {
-		amdgpu_bo_unref(&adev->stollen_vga_memory);
-		return r;
-	}
 	DRM_INFO("amdgpu: %uM of VRAM memory ready\n",
 		 (unsigned) (adev->mc.real_vram_size / (1024 * 1024)));
 
@@ -1319,13 +1352,13 @@ void amdgpu_ttm_fini(struct amdgpu_device *adev)
 	if (!adev->mman.initialized)
 		return;
 	amdgpu_ttm_debugfs_fini(adev);
-	if (adev->stollen_vga_memory) {
-		r = amdgpu_bo_reserve(adev->stollen_vga_memory, true);
+	if (adev->stolen_vga_memory) {
+		r = amdgpu_bo_reserve(adev->stolen_vga_memory, true);
 		if (r == 0) {
-			amdgpu_bo_unpin(adev->stollen_vga_memory);
-			amdgpu_bo_unreserve(adev->stollen_vga_memory);
+			amdgpu_bo_unpin(adev->stolen_vga_memory);
+			amdgpu_bo_unreserve(adev->stolen_vga_memory);
 		}
-		amdgpu_bo_unref(&adev->stollen_vga_memory);
+		amdgpu_bo_unref(&adev->stolen_vga_memory);
 	}
 	ttm_bo_clean_mm(&adev->mman.bdev, TTM_PL_VRAM);
 	ttm_bo_clean_mm(&adev->mman.bdev, TTM_PL_TT);
@@ -1509,11 +1542,12 @@ error_free:
 }
 
 int amdgpu_fill_buffer(struct amdgpu_bo *bo,
-		       uint32_t src_data,
+		       uint64_t src_data,
 		       struct reservation_object *resv,
 		       struct dma_fence **fence)
 {
 	struct amdgpu_device *adev = amdgpu_ttm_adev(bo->tbo.bdev);
+	/* max_bytes applies to SDMA_OP_PTEPDE as well as SDMA_OP_CONST_FILL*/
 	uint32_t max_bytes = adev->mman.buffer_funcs->fill_max_bytes;
 	struct amdgpu_ring *ring = adev->mman.buffer_funcs_ring;
 
@@ -1545,7 +1579,9 @@ int amdgpu_fill_buffer(struct amdgpu_bo *bo,
 		num_pages -= mm_node->size;
 		++mm_node;
 	}
-	num_dw = num_loops * adev->mman.buffer_funcs->fill_num_dw;
+
+	/* 10 double words for each SDMA_OP_PTEPDE cmd */
+	num_dw = num_loops * 10;
 
 	/* for IB padding */
 	num_dw += 64;
@@ -1570,12 +1606,16 @@ int amdgpu_fill_buffer(struct amdgpu_bo *bo,
 		uint32_t byte_count = mm_node->size << PAGE_SHIFT;
 		uint64_t dst_addr;
 
+		WARN_ONCE(byte_count & 0x7, "size should be a multiple of 8");
+
 		dst_addr = amdgpu_mm_node_addr(&bo->tbo, mm_node, &bo->tbo.mem);
 		while (byte_count) {
 			uint32_t cur_size_in_bytes = min(byte_count, max_bytes);
 
-			amdgpu_emit_fill_buffer(adev, &job->ibs[0], src_data,
-						dst_addr, cur_size_in_bytes);
+			amdgpu_vm_set_pte_pde(adev, &job->ibs[0],
+					dst_addr, 0,
+					cur_size_in_bytes >> 3, 0,
+					src_data);
 
 			dst_addr += cur_size_in_bytes;
 			byte_count -= cur_size_in_bytes;
@@ -1601,32 +1641,16 @@ error_free:
 
 #if defined(CONFIG_DEBUG_FS)
 
-extern void amdgpu_gtt_mgr_print(struct seq_file *m, struct ttm_mem_type_manager
-				 *man);
 static int amdgpu_mm_dump_table(struct seq_file *m, void *data)
 {
 	struct drm_info_node *node = (struct drm_info_node *)m->private;
 	unsigned ttm_pl = *(int *)node->info_ent->data;
 	struct drm_device *dev = node->minor->dev;
 	struct amdgpu_device *adev = dev->dev_private;
-	struct drm_mm *mm = (struct drm_mm *)adev->mman.bdev.man[ttm_pl].priv;
-	struct ttm_bo_global *glob = adev->mman.bdev.glob;
+	struct ttm_mem_type_manager *man = &adev->mman.bdev.man[ttm_pl];
 	struct drm_printer p = drm_seq_file_printer(m);
 
-	spin_lock(&glob->lru_lock);
-	drm_mm_print(mm, &p);
-	spin_unlock(&glob->lru_lock);
-	switch (ttm_pl) {
-	case TTM_PL_VRAM:
-		seq_printf(m, "man size:%llu pages, ram usage:%lluMB, vis usage:%lluMB\n",
-			   adev->mman.bdev.man[ttm_pl].size,
-			   (u64)atomic64_read(&adev->vram_usage) >> 20,
-			   (u64)atomic64_read(&adev->vram_vis_usage) >> 20);
-		break;
-	case TTM_PL_TT:
-		amdgpu_gtt_mgr_print(m, &adev->mman.bdev.man[TTM_PL_TT]);
-		break;
-	}
+	man->func->debug(man, &p);
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h
index f137c2458ee8..f22a4758719d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h
@@ -66,6 +66,10 @@ int amdgpu_gtt_mgr_alloc(struct ttm_mem_type_manager *man,
 			 struct ttm_buffer_object *tbo,
 			 const struct ttm_place *place,
 			 struct ttm_mem_reg *mem);
+uint64_t amdgpu_gtt_mgr_usage(struct ttm_mem_type_manager *man);
+
+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);
 
 int amdgpu_copy_buffer(struct amdgpu_ring *ring, uint64_t src_offset,
 		       uint64_t dst_offset, uint32_t byte_count,
@@ -73,7 +77,7 @@ int amdgpu_copy_buffer(struct amdgpu_ring *ring, uint64_t src_offset,
 		       struct dma_fence **fence, bool direct_submit,
 		       bool vm_needs_flush);
 int amdgpu_fill_buffer(struct amdgpu_bo *bo,
-			uint32_t src_data,
+			uint64_t src_data,
 			struct reservation_object *resv,
 			struct dma_fence **fence);
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c
index fcfb9d4f7477..36c763310df5 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c
@@ -358,8 +358,6 @@ static int amdgpu_ucode_patch_jt(struct amdgpu_firmware_info *ucode,
 			   (le32_to_cpu(header->jt_offset) * 4);
 	memcpy(dst_addr, src_addr, le32_to_cpu(header->jt_size) * 4);
 
-	ucode->ucode_size += le32_to_cpu(header->jt_size) * 4;
-
 	return 0;
 }
 
@@ -381,7 +379,7 @@ int amdgpu_ucode_init_bo(struct amdgpu_device *adev)
 	err = amdgpu_bo_create(adev, adev->firmware.fw_size, PAGE_SIZE, true,
 				amdgpu_sriov_vf(adev) ? AMDGPU_GEM_DOMAIN_VRAM : AMDGPU_GEM_DOMAIN_GTT,
 				AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS,
-				NULL, NULL, bo);
+				NULL, NULL, 0, bo);
 	if (err) {
 		dev_err(adev->dev, "(%d) Firmware buffer allocate failed\n", err);
 		goto failed;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c
index 2ca09f111f08..e19928dae8e3 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c
@@ -588,6 +588,10 @@ static int amdgpu_uvd_cs_msg_decode(struct amdgpu_device *adev, uint32_t *msg,
 		}
 		break;
 
+	case 8: /* MJPEG */
+		min_dpb_size = 0;
+		break;
+
 	case 16: /* H265 */
 		image_size = (ALIGN(width, 16) * ALIGN(height, 16) * 3) / 2;
 		image_size = ALIGN(image_size, 256);
@@ -1051,7 +1055,7 @@ int amdgpu_uvd_get_create_msg(struct amdgpu_ring *ring, uint32_t handle,
 			     AMDGPU_GEM_DOMAIN_VRAM,
 			     AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED |
 			     AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS,
-			     NULL, NULL, &bo);
+			     NULL, NULL, 0, &bo);
 	if (r)
 		return r;
 
@@ -1101,7 +1105,7 @@ int amdgpu_uvd_get_destroy_msg(struct amdgpu_ring *ring, uint32_t handle,
 			     AMDGPU_GEM_DOMAIN_VRAM,
 			     AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED |
 			     AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS,
-			     NULL, NULL, &bo);
+			     NULL, NULL, 0, &bo);
 	if (r)
 		return r;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c
index b692ad402252..c855366521ab 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c
@@ -937,9 +937,9 @@ int amdgpu_vce_ring_test_ring(struct amdgpu_ring *ring)
 	unsigned i;
 	int r, timeout = adev->usec_timeout;
 
-	/* workaround VCE ring test slow issue for sriov*/
+	/* skip ring test for sriov*/
 	if (amdgpu_sriov_vf(adev))
-		timeout *= 10;
+		return 0;
 
 	r = amdgpu_ring_alloc(ring, 16);
 	if (r) {
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
index 09190fadd228..041e0121590c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
@@ -209,9 +209,9 @@ static void amdgpu_vcn_idle_work_handler(struct work_struct *work)
 
 	if (fences == 0) {
 		if (adev->pm.dpm_enabled) {
+			/* might be used when with pg/cg
 			amdgpu_dpm_enable_uvd(adev, false);
-		} else {
-			amdgpu_asic_set_uvd_clocks(adev, 0, 0);
+			*/
 		}
 	} else {
 		schedule_delayed_work(&adev->vcn.idle_work, VCN_IDLE_TIMEOUT);
@@ -223,12 +223,10 @@ void amdgpu_vcn_ring_begin_use(struct amdgpu_ring *ring)
 	struct amdgpu_device *adev = ring->adev;
 	bool set_clocks = !cancel_delayed_work_sync(&adev->vcn.idle_work);
 
-	if (set_clocks) {
-		if (adev->pm.dpm_enabled) {
-			amdgpu_dpm_enable_uvd(adev, true);
-		} else {
-			amdgpu_asic_set_uvd_clocks(adev, 53300, 40000);
-		}
+	if (set_clocks && adev->pm.dpm_enabled) {
+		/* might be used when with pg/cg
+		amdgpu_dpm_enable_uvd(adev, true);
+		*/
 	}
 }
 
@@ -361,7 +359,7 @@ static int amdgpu_vcn_dec_get_create_msg(struct amdgpu_ring *ring, uint32_t hand
 			     AMDGPU_GEM_DOMAIN_VRAM,
 			     AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED |
 			     AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS,
-			     NULL, NULL, &bo);
+			     NULL, NULL, 0, &bo);
 	if (r)
 		return r;
 
@@ -413,7 +411,7 @@ static int amdgpu_vcn_dec_get_destroy_msg(struct amdgpu_ring *ring, uint32_t han
 			     AMDGPU_GEM_DOMAIN_VRAM,
 			     AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED |
 			     AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS,
-			     NULL, NULL, &bo);
+			     NULL, NULL, 0, &bo);
 	if (r)
 		return r;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c
index 8a081e162d13..ab05121b9272 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c
@@ -46,14 +46,14 @@ int amdgpu_allocate_static_csa(struct amdgpu_device *adev)
  * address within META_DATA init package to support SRIOV gfx preemption.
  */
 
-int amdgpu_map_static_csa(struct amdgpu_device *adev, struct amdgpu_vm *vm)
+int amdgpu_map_static_csa(struct amdgpu_device *adev, struct amdgpu_vm *vm,
+			  struct amdgpu_bo_va **bo_va)
 {
-	int r;
-	struct amdgpu_bo_va *bo_va;
 	struct ww_acquire_ctx ticket;
 	struct list_head list;
 	struct amdgpu_bo_list_entry pd;
 	struct ttm_validate_buffer csa_tv;
+	int r;
 
 	INIT_LIST_HEAD(&list);
 	INIT_LIST_HEAD(&csa_tv.head);
@@ -69,34 +69,33 @@ int amdgpu_map_static_csa(struct amdgpu_device *adev, struct amdgpu_vm *vm)
 		return r;
 	}
 
-	bo_va = amdgpu_vm_bo_add(adev, vm, adev->virt.csa_obj);
-	if (!bo_va) {
+	*bo_va = amdgpu_vm_bo_add(adev, vm, adev->virt.csa_obj);
+	if (!*bo_va) {
 		ttm_eu_backoff_reservation(&ticket, &list);
 		DRM_ERROR("failed to create bo_va for static CSA\n");
 		return -ENOMEM;
 	}
 
-	r = amdgpu_vm_alloc_pts(adev, bo_va->vm, AMDGPU_CSA_VADDR,
-				   AMDGPU_CSA_SIZE);
+	r = amdgpu_vm_alloc_pts(adev, (*bo_va)->base.vm, AMDGPU_CSA_VADDR,
+				AMDGPU_CSA_SIZE);
 	if (r) {
 		DRM_ERROR("failed to allocate pts for static CSA, err=%d\n", r);
-		amdgpu_vm_bo_rmv(adev, bo_va);
+		amdgpu_vm_bo_rmv(adev, *bo_va);
 		ttm_eu_backoff_reservation(&ticket, &list);
 		return r;
 	}
 
-	r = amdgpu_vm_bo_map(adev, bo_va, AMDGPU_CSA_VADDR, 0,AMDGPU_CSA_SIZE,
-						AMDGPU_PTE_READABLE | AMDGPU_PTE_WRITEABLE |
-						AMDGPU_PTE_EXECUTABLE);
+	r = amdgpu_vm_bo_map(adev, *bo_va, AMDGPU_CSA_VADDR, 0, AMDGPU_CSA_SIZE,
+			     AMDGPU_PTE_READABLE | AMDGPU_PTE_WRITEABLE |
+			     AMDGPU_PTE_EXECUTABLE);
 
 	if (r) {
 		DRM_ERROR("failed to do bo_map on static CSA, err=%d\n", r);
-		amdgpu_vm_bo_rmv(adev, bo_va);
+		amdgpu_vm_bo_rmv(adev, *bo_va);
 		ttm_eu_backoff_reservation(&ticket, &list);
 		return r;
 	}
 
-	vm->csa_bo_va = bo_va;
 	ttm_eu_backoff_reservation(&ticket, &list);
 	return 0;
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h
index e5b1baf387c1..afcfb8bcfb65 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h
@@ -90,7 +90,8 @@ static inline bool is_virtual_machine(void)
 
 struct amdgpu_vm;
 int amdgpu_allocate_static_csa(struct amdgpu_device *adev);
-int amdgpu_map_static_csa(struct amdgpu_device *adev, struct amdgpu_vm *vm);
+int amdgpu_map_static_csa(struct amdgpu_device *adev, struct amdgpu_vm *vm,
+			  struct amdgpu_bo_va **bo_va);
 void amdgpu_virt_init_setting(struct amdgpu_device *adev);
 uint32_t amdgpu_virt_kiq_rreg(struct amdgpu_device *adev, uint32_t reg);
 void amdgpu_virt_kiq_wreg(struct amdgpu_device *adev, uint32_t reg, uint32_t v);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
index 250c8e80e646..6b1343e5541d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
@@ -159,11 +159,20 @@ void amdgpu_vm_get_pd_bo(struct amdgpu_vm *vm,
  */
 static int amdgpu_vm_validate_level(struct amdgpu_vm_pt *parent,
 				    int (*validate)(void *, struct amdgpu_bo *),
-				    void *param, bool use_cpu_for_update)
+				    void *param, bool use_cpu_for_update,
+				    struct ttm_bo_global *glob)
 {
 	unsigned i;
 	int r;
 
+	if (parent->bo->shadow) {
+		struct amdgpu_bo *shadow = parent->bo->shadow;
+
+		r = amdgpu_ttm_bind(&shadow->tbo, &shadow->tbo.mem);
+		if (r)
+			return r;
+	}
+
 	if (use_cpu_for_update) {
 		r = amdgpu_bo_kmap(parent->bo, NULL);
 		if (r)
@@ -183,12 +192,18 @@ static int amdgpu_vm_validate_level(struct amdgpu_vm_pt *parent,
 		if (r)
 			return r;
 
+		spin_lock(&glob->lru_lock);
+		ttm_bo_move_to_lru_tail(&entry->bo->tbo);
+		if (entry->bo->shadow)
+			ttm_bo_move_to_lru_tail(&entry->bo->shadow->tbo);
+		spin_unlock(&glob->lru_lock);
+
 		/*
 		 * Recurse into the sub directory. This is harmless because we
 		 * have only a maximum of 5 layers.
 		 */
 		r = amdgpu_vm_validate_level(entry, validate, param,
-					     use_cpu_for_update);
+					     use_cpu_for_update, glob);
 		if (r)
 			return r;
 	}
@@ -220,54 +235,11 @@ int amdgpu_vm_validate_pt_bos(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 		return 0;
 
 	return amdgpu_vm_validate_level(&vm->root, validate, param,
-					vm->use_cpu_for_update);
+					vm->use_cpu_for_update,
+					adev->mman.bdev.glob);
 }
 
 /**
- * amdgpu_vm_move_level_in_lru - move one level of PT BOs to the LRU tail
- *
- * @adev: amdgpu device instance
- * @vm: vm providing the BOs
- *
- * Move the PT BOs to the tail of the LRU.
- */
-static void amdgpu_vm_move_level_in_lru(struct amdgpu_vm_pt *parent)
-{
-	unsigned i;
-
-	if (!parent->entries)
-		return;
-
-	for (i = 0; i <= parent->last_entry_used; ++i) {
-		struct amdgpu_vm_pt *entry = &parent->entries[i];
-
-		if (!entry->bo)
-			continue;
-
-		ttm_bo_move_to_lru_tail(&entry->bo->tbo);
-		amdgpu_vm_move_level_in_lru(entry);
-	}
-}
-
-/**
- * amdgpu_vm_move_pt_bos_in_lru - move the PT BOs to the LRU tail
- *
- * @adev: amdgpu device instance
- * @vm: vm providing the BOs
- *
- * Move the PT BOs to the tail of the LRU.
- */
-void amdgpu_vm_move_pt_bos_in_lru(struct amdgpu_device *adev,
-				  struct amdgpu_vm *vm)
-{
-	struct ttm_bo_global *glob = adev->mman.bdev.glob;
-
-	spin_lock(&glob->lru_lock);
-	amdgpu_vm_move_level_in_lru(&vm->root);
-	spin_unlock(&glob->lru_lock);
-}
-
- /**
  * amdgpu_vm_alloc_levels - allocate the PD/PT levels
  *
  * @adev: amdgpu_device pointer
@@ -288,6 +260,7 @@ static int amdgpu_vm_alloc_levels(struct amdgpu_device *adev,
 	unsigned pt_idx, from, to;
 	int r;
 	u64 flags;
+	uint64_t init_value = 0;
 
 	if (!parent->entries) {
 		unsigned num_entries = amdgpu_vm_num_entries(adev, level);
@@ -321,6 +294,12 @@ static int amdgpu_vm_alloc_levels(struct amdgpu_device *adev,
 		flags |= (AMDGPU_GEM_CREATE_NO_CPU_ACCESS |
 				AMDGPU_GEM_CREATE_SHADOW);
 
+	if (vm->pte_support_ats) {
+		init_value = AMDGPU_PTE_SYSTEM;
+		if (level != adev->vm_manager.num_level - 1)
+			init_value |= AMDGPU_PDE_PTE;
+	}
+
 	/* walk over the address space and allocate the page tables */
 	for (pt_idx = from; pt_idx <= to; ++pt_idx) {
 		struct reservation_object *resv = vm->root.bo->tbo.resv;
@@ -333,7 +312,7 @@ static int amdgpu_vm_alloc_levels(struct amdgpu_device *adev,
 					     AMDGPU_GPU_PAGE_SIZE, true,
 					     AMDGPU_GEM_DOMAIN_VRAM,
 					     flags,
-					     NULL, resv, &pt);
+					     NULL, resv, init_value, &pt);
 			if (r)
 				return r;
 
@@ -352,7 +331,6 @@ static int amdgpu_vm_alloc_levels(struct amdgpu_device *adev,
 
 			entry->bo = pt;
 			entry->addr = 0;
-			entry->huge_page = false;
 		}
 
 		if (level < adev->vm_manager.num_level) {
@@ -892,8 +870,8 @@ struct amdgpu_bo_va *amdgpu_vm_bo_find(struct amdgpu_vm *vm,
 {
 	struct amdgpu_bo_va *bo_va;
 
-	list_for_each_entry(bo_va, &bo->va, bo_list) {
-		if (bo_va->vm == vm) {
+	list_for_each_entry(bo_va, &bo->va, base.bo_list) {
+		if (bo_va->base.vm == vm) {
 			return bo_va;
 		}
 	}
@@ -1060,18 +1038,13 @@ static int amdgpu_vm_update_level(struct amdgpu_device *adev,
 	shadow = parent->bo->shadow;
 
 	if (vm->use_cpu_for_update) {
-		pd_addr = (unsigned long)parent->bo->kptr;
+		pd_addr = (unsigned long)amdgpu_bo_kptr(parent->bo);
 		r = amdgpu_vm_wait_pd(adev, vm, AMDGPU_FENCE_OWNER_VM);
 		if (unlikely(r))
 			return r;
 
 		params.func = amdgpu_vm_cpu_set_ptes;
 	} else {
-		if (shadow) {
-			r = amdgpu_ttm_bind(&shadow->tbo, &shadow->tbo.mem);
-			if (r)
-				return r;
-		}
 		ring = container_of(vm->entity.sched, struct amdgpu_ring,
 				    sched);
 
@@ -1107,22 +1080,14 @@ static int amdgpu_vm_update_level(struct amdgpu_device *adev,
 		if (bo == NULL)
 			continue;
 
-		if (bo->shadow) {
-			struct amdgpu_bo *pt_shadow = bo->shadow;
-
-			r = amdgpu_ttm_bind(&pt_shadow->tbo,
-					    &pt_shadow->tbo.mem);
-			if (r)
-				return r;
-		}
-
 		pt = amdgpu_bo_gpu_offset(bo);
 		pt = amdgpu_gart_get_vm_pde(adev, pt);
-		if (parent->entries[pt_idx].addr == pt ||
-		    parent->entries[pt_idx].huge_page)
+		/* Don't update huge pages here */
+		if ((parent->entries[pt_idx].addr & AMDGPU_PDE_PTE) ||
+		    parent->entries[pt_idx].addr == (pt | AMDGPU_PTE_VALID))
 			continue;
 
-		parent->entries[pt_idx].addr = pt;
+		parent->entries[pt_idx].addr = pt | AMDGPU_PTE_VALID;
 
 		pde = pd_addr + pt_idx * 8;
 		if (((last_pde + 8 * count) != pde) ||
@@ -1300,15 +1265,14 @@ void amdgpu_vm_get_entry(struct amdgpu_pte_update_params *p, uint64_t addr,
  *
  * Check if we can update the PD with a huge page.
  */
-static int amdgpu_vm_handle_huge_pages(struct amdgpu_pte_update_params *p,
-				       struct amdgpu_vm_pt *entry,
-				       struct amdgpu_vm_pt *parent,
-				       unsigned nptes, uint64_t dst,
-				       uint64_t flags)
+static void amdgpu_vm_handle_huge_pages(struct amdgpu_pte_update_params *p,
+					struct amdgpu_vm_pt *entry,
+					struct amdgpu_vm_pt *parent,
+					unsigned nptes, uint64_t dst,
+					uint64_t flags)
 {
 	bool use_cpu_update = (p->func == amdgpu_vm_cpu_set_ptes);
 	uint64_t pd_addr, pde;
-	int r;
 
 	/* In the case of a mixed PT the PDE must point to it*/
 	if (p->adev->asic_type < CHIP_VEGA10 ||
@@ -1320,21 +1284,17 @@ static int amdgpu_vm_handle_huge_pages(struct amdgpu_pte_update_params *p,
 		dst = amdgpu_gart_get_vm_pde(p->adev, dst);
 		flags = AMDGPU_PTE_VALID;
 	} else {
+		/* Set the huge page flag to stop scanning at this PDE */
 		flags |= AMDGPU_PDE_PTE;
 	}
 
-	if (entry->addr == dst &&
-	    entry->huge_page == !!(flags & AMDGPU_PDE_PTE))
-		return 0;
+	if (entry->addr == (dst | flags))
+		return;
 
-	entry->addr = dst;
-	entry->huge_page = !!(flags & AMDGPU_PDE_PTE);
+	entry->addr = (dst | flags);
 
 	if (use_cpu_update) {
-		r = amdgpu_bo_kmap(parent->bo, (void *)&pd_addr);
-		if (r)
-			return r;
-
+		pd_addr = (unsigned long)amdgpu_bo_kptr(parent->bo);
 		pde = pd_addr + (entry - parent->entries) * 8;
 		amdgpu_vm_cpu_set_ptes(p, pde, dst, 1, 0, flags);
 	} else {
@@ -1347,8 +1307,6 @@ static int amdgpu_vm_handle_huge_pages(struct amdgpu_pte_update_params *p,
 		pde = pd_addr + (entry - parent->entries) * 8;
 		amdgpu_vm_do_set_ptes(p, pde, dst, 1, 0, flags);
 	}
-
-	return 0;
 }
 
 /**
@@ -1375,7 +1333,6 @@ static int amdgpu_vm_update_ptes(struct amdgpu_pte_update_params *params,
 	struct amdgpu_bo *pt;
 	unsigned nptes;
 	bool use_cpu_update = (params->func == amdgpu_vm_cpu_set_ptes);
-	int r;
 
 	/* walk over the address space and update the page tables */
 	for (addr = start; addr < end; addr += nptes,
@@ -1391,17 +1348,15 @@ static int amdgpu_vm_update_ptes(struct amdgpu_pte_update_params *params,
 		else
 			nptes = AMDGPU_VM_PTE_COUNT(adev) - (addr & mask);
 
-		r = amdgpu_vm_handle_huge_pages(params, entry, parent,
-						nptes, dst, flags);
-		if (r)
-			return r;
-
-		if (entry->huge_page)
+		amdgpu_vm_handle_huge_pages(params, entry, parent,
+					    nptes, dst, flags);
+		/* We don't need to update PTEs for huge pages */
+		if (entry->addr & AMDGPU_PDE_PTE)
 			continue;
 
 		pt = entry->bo;
 		if (use_cpu_update) {
-			pe_start = (unsigned long)pt->kptr;
+			pe_start = (unsigned long)amdgpu_bo_kptr(pt);
 		} else {
 			if (pt->shadow) {
 				pe_start = amdgpu_bo_gpu_offset(pt->shadow);
@@ -1455,9 +1410,7 @@ static int amdgpu_vm_frag_ptes(struct amdgpu_pte_update_params	*params,
 	 * Userspace can support this by aligning virtual base address and
 	 * allocation size to the fragment size.
 	 */
-
-	/* SI and newer are optimized for 64KB */
-	unsigned pages_per_frag = AMDGPU_LOG2_PAGES_PER_FRAG(params->adev);
+	unsigned pages_per_frag = params->adev->vm_manager.fragment_size;
 	uint64_t frag_flags = AMDGPU_PTE_FRAG(pages_per_frag);
 	uint64_t frag_align = 1 << pages_per_frag;
 
@@ -1771,7 +1724,8 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev,
 			struct amdgpu_bo_va *bo_va,
 			bool clear)
 {
-	struct amdgpu_vm *vm = bo_va->vm;
+	struct amdgpu_bo *bo = bo_va->base.bo;
+	struct amdgpu_vm *vm = bo_va->base.vm;
 	struct amdgpu_bo_va_mapping *mapping;
 	dma_addr_t *pages_addr = NULL;
 	uint64_t gtt_flags, flags;
@@ -1780,27 +1734,27 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev,
 	struct dma_fence *exclusive;
 	int r;
 
-	if (clear || !bo_va->bo) {
+	if (clear || !bo_va->base.bo) {
 		mem = NULL;
 		nodes = NULL;
 		exclusive = NULL;
 	} else {
 		struct ttm_dma_tt *ttm;
 
-		mem = &bo_va->bo->tbo.mem;
+		mem = &bo_va->base.bo->tbo.mem;
 		nodes = mem->mm_node;
 		if (mem->mem_type == TTM_PL_TT) {
-			ttm = container_of(bo_va->bo->tbo.ttm, struct
-					   ttm_dma_tt, ttm);
+			ttm = container_of(bo_va->base.bo->tbo.ttm,
+					   struct ttm_dma_tt, ttm);
 			pages_addr = ttm->dma_address;
 		}
-		exclusive = reservation_object_get_excl(bo_va->bo->tbo.resv);
+		exclusive = reservation_object_get_excl(bo->tbo.resv);
 	}
 
-	if (bo_va->bo) {
-		flags = amdgpu_ttm_tt_pte_flags(adev, bo_va->bo->tbo.ttm, mem);
-		gtt_flags = (amdgpu_ttm_is_bound(bo_va->bo->tbo.ttm) &&
-			adev == amdgpu_ttm_adev(bo_va->bo->tbo.bdev)) ?
+	if (bo) {
+		flags = amdgpu_ttm_tt_pte_flags(adev, bo->tbo.ttm, mem);
+		gtt_flags = (amdgpu_ttm_is_bound(bo->tbo.ttm) &&
+			adev == amdgpu_ttm_adev(bo->tbo.bdev)) ?
 			flags : 0;
 	} else {
 		flags = 0x0;
@@ -1808,7 +1762,7 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev,
 	}
 
 	spin_lock(&vm->status_lock);
-	if (!list_empty(&bo_va->vm_status))
+	if (!list_empty(&bo_va->base.vm_status))
 		list_splice_init(&bo_va->valids, &bo_va->invalids);
 	spin_unlock(&vm->status_lock);
 
@@ -1831,9 +1785,9 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev,
 
 	spin_lock(&vm->status_lock);
 	list_splice_init(&bo_va->invalids, &bo_va->valids);
-	list_del_init(&bo_va->vm_status);
+	list_del_init(&bo_va->base.vm_status);
 	if (clear)
-		list_add(&bo_va->vm_status, &vm->cleared);
+		list_add(&bo_va->base.vm_status, &vm->cleared);
 	spin_unlock(&vm->status_lock);
 
 	if (vm->use_cpu_for_update) {
@@ -1995,15 +1949,19 @@ int amdgpu_vm_clear_freed(struct amdgpu_device *adev,
 	struct amdgpu_bo_va_mapping *mapping;
 	struct dma_fence *f = NULL;
 	int r;
+	uint64_t init_pte_value = 0;
 
 	while (!list_empty(&vm->freed)) {
 		mapping = list_first_entry(&vm->freed,
 			struct amdgpu_bo_va_mapping, list);
 		list_del(&mapping->list);
 
+		if (vm->pte_support_ats)
+			init_pte_value = AMDGPU_PTE_SYSTEM;
+
 		r = amdgpu_vm_bo_update_mapping(adev, NULL, 0, NULL, vm,
 						mapping->start, mapping->last,
-						0, 0, &f);
+						init_pte_value, 0, &f);
 		amdgpu_vm_free_mapping(adev, vm, mapping, f);
 		if (r) {
 			dma_fence_put(f);
@@ -2023,26 +1981,26 @@ int amdgpu_vm_clear_freed(struct amdgpu_device *adev,
 }
 
 /**
- * amdgpu_vm_clear_invalids - clear invalidated BOs in the PT
+ * amdgpu_vm_clear_moved - clear moved BOs in the PT
  *
  * @adev: amdgpu_device pointer
  * @vm: requested vm
  *
- * Make sure all invalidated BOs are cleared in the PT.
+ * Make sure all moved BOs are cleared in the PT.
  * Returns 0 for success.
  *
  * PTs have to be reserved and mutex must be locked!
  */
-int amdgpu_vm_clear_invalids(struct amdgpu_device *adev,
-			     struct amdgpu_vm *vm, struct amdgpu_sync *sync)
+int amdgpu_vm_clear_moved(struct amdgpu_device *adev, struct amdgpu_vm *vm,
+			    struct amdgpu_sync *sync)
 {
 	struct amdgpu_bo_va *bo_va = NULL;
 	int r = 0;
 
 	spin_lock(&vm->status_lock);
-	while (!list_empty(&vm->invalidated)) {
-		bo_va = list_first_entry(&vm->invalidated,
-			struct amdgpu_bo_va, vm_status);
+	while (!list_empty(&vm->moved)) {
+		bo_va = list_first_entry(&vm->moved,
+			struct amdgpu_bo_va, base.vm_status);
 		spin_unlock(&vm->status_lock);
 
 		r = amdgpu_vm_bo_update(adev, bo_va, true);
@@ -2082,16 +2040,17 @@ struct amdgpu_bo_va *amdgpu_vm_bo_add(struct amdgpu_device *adev,
 	if (bo_va == NULL) {
 		return NULL;
 	}
-	bo_va->vm = vm;
-	bo_va->bo = bo;
+	bo_va->base.vm = vm;
+	bo_va->base.bo = bo;
+	INIT_LIST_HEAD(&bo_va->base.bo_list);
+	INIT_LIST_HEAD(&bo_va->base.vm_status);
+
 	bo_va->ref_count = 1;
-	INIT_LIST_HEAD(&bo_va->bo_list);
 	INIT_LIST_HEAD(&bo_va->valids);
 	INIT_LIST_HEAD(&bo_va->invalids);
-	INIT_LIST_HEAD(&bo_va->vm_status);
 
 	if (bo)
-		list_add_tail(&bo_va->bo_list, &bo->va);
+		list_add_tail(&bo_va->base.bo_list, &bo->va);
 
 	return bo_va;
 }
@@ -2116,7 +2075,8 @@ int amdgpu_vm_bo_map(struct amdgpu_device *adev,
 		     uint64_t size, uint64_t flags)
 {
 	struct amdgpu_bo_va_mapping *mapping, *tmp;
-	struct amdgpu_vm *vm = bo_va->vm;
+	struct amdgpu_bo *bo = bo_va->base.bo;
+	struct amdgpu_vm *vm = bo_va->base.vm;
 	uint64_t eaddr;
 
 	/* validate the parameters */
@@ -2127,7 +2087,7 @@ int amdgpu_vm_bo_map(struct amdgpu_device *adev,
 	/* make sure object fit at this offset */
 	eaddr = saddr + size - 1;
 	if (saddr >= eaddr ||
-	    (bo_va->bo && offset + size > amdgpu_bo_size(bo_va->bo)))
+	    (bo && offset + size > amdgpu_bo_size(bo)))
 		return -EINVAL;
 
 	saddr /= AMDGPU_GPU_PAGE_SIZE;
@@ -2137,7 +2097,7 @@ int amdgpu_vm_bo_map(struct amdgpu_device *adev,
 	if (tmp) {
 		/* bo and tmp overlap, invalid addr */
 		dev_err(adev->dev, "bo %p va 0x%010Lx-0x%010Lx conflict with "
-			"0x%010Lx-0x%010Lx\n", bo_va->bo, saddr, eaddr,
+			"0x%010Lx-0x%010Lx\n", bo, saddr, eaddr,
 			tmp->start, tmp->last + 1);
 		return -EINVAL;
 	}
@@ -2182,7 +2142,8 @@ int amdgpu_vm_bo_replace_map(struct amdgpu_device *adev,
 			     uint64_t size, uint64_t flags)
 {
 	struct amdgpu_bo_va_mapping *mapping;
-	struct amdgpu_vm *vm = bo_va->vm;
+	struct amdgpu_bo *bo = bo_va->base.bo;
+	struct amdgpu_vm *vm = bo_va->base.vm;
 	uint64_t eaddr;
 	int r;
 
@@ -2194,7 +2155,7 @@ int amdgpu_vm_bo_replace_map(struct amdgpu_device *adev,
 	/* make sure object fit at this offset */
 	eaddr = saddr + size - 1;
 	if (saddr >= eaddr ||
-	    (bo_va->bo && offset + size > amdgpu_bo_size(bo_va->bo)))
+	    (bo && offset + size > amdgpu_bo_size(bo)))
 		return -EINVAL;
 
 	/* Allocate all the needed memory */
@@ -2202,7 +2163,7 @@ int amdgpu_vm_bo_replace_map(struct amdgpu_device *adev,
 	if (!mapping)
 		return -ENOMEM;
 
-	r = amdgpu_vm_bo_clear_mappings(adev, bo_va->vm, saddr, size);
+	r = amdgpu_vm_bo_clear_mappings(adev, bo_va->base.vm, saddr, size);
 	if (r) {
 		kfree(mapping);
 		return r;
@@ -2242,7 +2203,7 @@ int amdgpu_vm_bo_unmap(struct amdgpu_device *adev,
 		       uint64_t saddr)
 {
 	struct amdgpu_bo_va_mapping *mapping;
-	struct amdgpu_vm *vm = bo_va->vm;
+	struct amdgpu_vm *vm = bo_va->base.vm;
 	bool valid = true;
 
 	saddr /= AMDGPU_GPU_PAGE_SIZE;
@@ -2390,12 +2351,12 @@ void amdgpu_vm_bo_rmv(struct amdgpu_device *adev,
 		      struct amdgpu_bo_va *bo_va)
 {
 	struct amdgpu_bo_va_mapping *mapping, *next;
-	struct amdgpu_vm *vm = bo_va->vm;
+	struct amdgpu_vm *vm = bo_va->base.vm;
 
-	list_del(&bo_va->bo_list);
+	list_del(&bo_va->base.bo_list);
 
 	spin_lock(&vm->status_lock);
-	list_del(&bo_va->vm_status);
+	list_del(&bo_va->base.vm_status);
 	spin_unlock(&vm->status_lock);
 
 	list_for_each_entry_safe(mapping, next, &bo_va->valids, list) {
@@ -2427,13 +2388,14 @@ void amdgpu_vm_bo_rmv(struct amdgpu_device *adev,
 void amdgpu_vm_bo_invalidate(struct amdgpu_device *adev,
 			     struct amdgpu_bo *bo)
 {
-	struct amdgpu_bo_va *bo_va;
+	struct amdgpu_vm_bo_base *bo_base;
 
-	list_for_each_entry(bo_va, &bo->va, bo_list) {
-		spin_lock(&bo_va->vm->status_lock);
-		if (list_empty(&bo_va->vm_status))
-			list_add(&bo_va->vm_status, &bo_va->vm->invalidated);
-		spin_unlock(&bo_va->vm->status_lock);
+	list_for_each_entry(bo_base, &bo->va, bo_list) {
+		spin_lock(&bo_base->vm->status_lock);
+		if (list_empty(&bo_base->vm_status))
+			list_add(&bo_base->vm_status,
+				 &bo_base->vm->moved);
+		spin_unlock(&bo_base->vm->status_lock);
 	}
 }
 
@@ -2451,12 +2413,26 @@ static uint32_t amdgpu_vm_get_block_size(uint64_t vm_size)
 }
 
 /**
- * amdgpu_vm_adjust_size - adjust vm size and block size
+ * amdgpu_vm_set_fragment_size - adjust fragment size in PTE
+ *
+ * @adev: amdgpu_device pointer
+ * @fragment_size_default: the default fragment size if it's set auto
+ */
+void amdgpu_vm_set_fragment_size(struct amdgpu_device *adev, uint32_t fragment_size_default)
+{
+	if (amdgpu_vm_fragment_size == -1)
+		adev->vm_manager.fragment_size = fragment_size_default;
+	else
+		adev->vm_manager.fragment_size = amdgpu_vm_fragment_size;
+}
+
+/**
+ * amdgpu_vm_adjust_size - adjust vm size, block size and fragment size
  *
  * @adev: amdgpu_device pointer
  * @vm_size: the default vm size if it's set auto
  */
-void amdgpu_vm_adjust_size(struct amdgpu_device *adev, uint64_t vm_size)
+void amdgpu_vm_adjust_size(struct amdgpu_device *adev, uint64_t vm_size, uint32_t fragment_size_default)
 {
 	/* adjust vm size firstly */
 	if (amdgpu_vm_size == -1)
@@ -2471,8 +2447,11 @@ void amdgpu_vm_adjust_size(struct amdgpu_device *adev, uint64_t vm_size)
 	else
 		adev->vm_manager.block_size = amdgpu_vm_block_size;
 
-	DRM_INFO("vm size is %llu GB, block size is %u-bit\n",
-		adev->vm_manager.vm_size, adev->vm_manager.block_size);
+	amdgpu_vm_set_fragment_size(adev, fragment_size_default);
+
+	DRM_INFO("vm size is %llu GB, block size is %u-bit, fragment size is %u-bit\n",
+		adev->vm_manager.vm_size, adev->vm_manager.block_size,
+		adev->vm_manager.fragment_size);
 }
 
 /**
@@ -2494,13 +2473,14 @@ int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 	struct amd_sched_rq *rq;
 	int r, i;
 	u64 flags;
+	uint64_t init_pde_value = 0;
 
 	vm->va = RB_ROOT;
 	vm->client_id = atomic64_inc_return(&adev->vm_manager.client_counter);
 	for (i = 0; i < AMDGPU_MAX_VMHUBS; i++)
 		vm->reserved_vmid[i] = NULL;
 	spin_lock_init(&vm->status_lock);
-	INIT_LIST_HEAD(&vm->invalidated);
+	INIT_LIST_HEAD(&vm->moved);
 	INIT_LIST_HEAD(&vm->cleared);
 	INIT_LIST_HEAD(&vm->freed);
 
@@ -2515,10 +2495,17 @@ int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 	if (r)
 		return r;
 
-	if (vm_context == AMDGPU_VM_CONTEXT_COMPUTE)
+	vm->pte_support_ats = false;
+
+	if (vm_context == AMDGPU_VM_CONTEXT_COMPUTE) {
 		vm->use_cpu_for_update = !!(adev->vm_manager.vm_update_mode &
 						AMDGPU_VM_USE_CPU_FOR_COMPUTE);
-	else
+
+		if (adev->asic_type == CHIP_RAVEN) {
+			vm->pte_support_ats = true;
+			init_pde_value = AMDGPU_PTE_SYSTEM | AMDGPU_PDE_PTE;
+		}
+	} else
 		vm->use_cpu_for_update = !!(adev->vm_manager.vm_update_mode &
 						AMDGPU_VM_USE_CPU_FOR_GFX);
 	DRM_DEBUG_DRIVER("VM update mode is %s\n",
@@ -2538,7 +2525,7 @@ int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 	r = amdgpu_bo_create(adev, amdgpu_vm_bo_size(adev, 0), align, true,
 			     AMDGPU_GEM_DOMAIN_VRAM,
 			     flags,
-			     NULL, NULL, &vm->root.bo);
+			     NULL, NULL, init_pde_value, &vm->root.bo);
 	if (r)
 		goto error_free_sched_entity;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
index 34d9174ebff2..ba6691b58ee7 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
@@ -50,11 +50,6 @@ struct amdgpu_bo_list_entry;
 /* PTBs (Page Table Blocks) need to be aligned to 32K */
 #define AMDGPU_VM_PTB_ALIGN_SIZE   32768
 
-/* LOG2 number of continuous pages for the fragment field */
-#define AMDGPU_LOG2_PAGES_PER_FRAG(adev) \
-	((adev)->asic_type < CHIP_VEGA10 ? 4 : \
-	 (adev)->vm_manager.block_size)
-
 #define AMDGPU_PTE_VALID	(1ULL << 0)
 #define AMDGPU_PTE_SYSTEM	(1ULL << 1)
 #define AMDGPU_PTE_SNOOPED	(1ULL << 2)
@@ -99,11 +94,22 @@ struct amdgpu_bo_list_entry;
 #define AMDGPU_VM_USE_CPU_FOR_GFX (1 << 0)
 #define AMDGPU_VM_USE_CPU_FOR_COMPUTE (1 << 1)
 
+/* base structure for tracking BO usage in a VM */
+struct amdgpu_vm_bo_base {
+	/* constant after initialization */
+	struct amdgpu_vm		*vm;
+	struct amdgpu_bo		*bo;
+
+	/* protected by bo being reserved */
+	struct list_head		bo_list;
+
+	/* protected by spinlock */
+	struct list_head		vm_status;
+};
 
 struct amdgpu_vm_pt {
 	struct amdgpu_bo	*bo;
 	uint64_t		addr;
-	bool			huge_page;
 
 	/* array of page tables, one for each directory entry */
 	struct amdgpu_vm_pt	*entries;
@@ -118,7 +124,7 @@ struct amdgpu_vm {
 	spinlock_t		status_lock;
 
 	/* BOs moved, but not yet updated in the PT */
-	struct list_head	invalidated;
+	struct list_head	moved;
 
 	/* BOs cleared in the PT because of a move */
 	struct list_head	cleared;
@@ -141,11 +147,12 @@ struct amdgpu_vm {
 	u64                     client_id;
 	/* dedicated to vm */
 	struct amdgpu_vm_id	*reserved_vmid[AMDGPU_MAX_VMHUBS];
-	/* each VM will map on CSA */
-	struct amdgpu_bo_va *csa_bo_va;
 
 	/* Flag to indicate if VM tables are updated by CPU or GPU (SDMA) */
 	bool                    use_cpu_for_update;
+
+	/* Flag to indicate ATS support from PTE for GFX9 */
+	bool			pte_support_ats;
 };
 
 struct amdgpu_vm_id {
@@ -188,6 +195,7 @@ struct amdgpu_vm_manager {
 	uint32_t				num_level;
 	uint64_t				vm_size;
 	uint32_t				block_size;
+	uint32_t				fragment_size;
 	/* vram base address for page table entry  */
 	u64					vram_base_offset;
 	/* vm pte handling */
@@ -220,8 +228,6 @@ void amdgpu_vm_get_pd_bo(struct amdgpu_vm *vm,
 int amdgpu_vm_validate_pt_bos(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 			      int (*callback)(void *p, struct amdgpu_bo *bo),
 			      void *param);
-void amdgpu_vm_move_pt_bos_in_lru(struct amdgpu_device *adev,
-				  struct amdgpu_vm *vm);
 int amdgpu_vm_alloc_pts(struct amdgpu_device *adev,
 			struct amdgpu_vm *vm,
 			uint64_t saddr, uint64_t size);
@@ -237,8 +243,8 @@ int amdgpu_vm_update_directories(struct amdgpu_device *adev,
 int amdgpu_vm_clear_freed(struct amdgpu_device *adev,
 			  struct amdgpu_vm *vm,
 			  struct dma_fence **fence);
-int amdgpu_vm_clear_invalids(struct amdgpu_device *adev, struct amdgpu_vm *vm,
-			     struct amdgpu_sync *sync);
+int amdgpu_vm_clear_moved(struct amdgpu_device *adev, struct amdgpu_vm *vm,
+			  struct amdgpu_sync *sync);
 int amdgpu_vm_bo_update(struct amdgpu_device *adev,
 			struct amdgpu_bo_va *bo_va,
 			bool clear);
@@ -265,7 +271,10 @@ int amdgpu_vm_bo_clear_mappings(struct amdgpu_device *adev,
 				uint64_t saddr, uint64_t size);
 void amdgpu_vm_bo_rmv(struct amdgpu_device *adev,
 		      struct amdgpu_bo_va *bo_va);
-void amdgpu_vm_adjust_size(struct amdgpu_device *adev, uint64_t vm_size);
+void amdgpu_vm_set_fragment_size(struct amdgpu_device *adev,
+				uint32_t fragment_size_default);
+void amdgpu_vm_adjust_size(struct amdgpu_device *adev, uint64_t vm_size,
+				uint32_t fragment_size_default);
 int amdgpu_vm_ioctl(struct drm_device *dev, void *data, struct drm_file *filp);
 bool amdgpu_vm_need_pipeline_sync(struct amdgpu_ring *ring,
 				  struct amdgpu_job *job);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
index a2c59a08b2bd..26e900627971 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
@@ -28,6 +28,8 @@
 struct amdgpu_vram_mgr {
 	struct drm_mm mm;
 	spinlock_t lock;
+	atomic64_t usage;
+	atomic64_t vis_usage;
 };
 
 /**
@@ -79,6 +81,27 @@ static int amdgpu_vram_mgr_fini(struct ttm_mem_type_manager *man)
 }
 
 /**
+ * amdgpu_vram_mgr_vis_size - Calculate visible node size
+ *
+ * @adev: amdgpu device structure
+ * @node: MM node structure
+ *
+ * Calculate how many bytes of the MM node are inside visible VRAM
+ */
+static u64 amdgpu_vram_mgr_vis_size(struct amdgpu_device *adev,
+				    struct drm_mm_node *node)
+{
+	uint64_t start = node->start << PAGE_SHIFT;
+	uint64_t end = (node->size + node->start) << PAGE_SHIFT;
+
+	if (start >= adev->mc.visible_vram_size)
+		return 0;
+
+	return (end > adev->mc.visible_vram_size ?
+		adev->mc.visible_vram_size : end) - start;
+}
+
+/**
  * amdgpu_vram_mgr_new - allocate new ranges
  *
  * @man: TTM memory type manager
@@ -93,11 +116,13 @@ static int amdgpu_vram_mgr_new(struct ttm_mem_type_manager *man,
 			       const struct ttm_place *place,
 			       struct ttm_mem_reg *mem)
 {
+	struct amdgpu_device *adev = amdgpu_ttm_adev(man->bdev);
 	struct amdgpu_vram_mgr *mgr = man->priv;
 	struct drm_mm *mm = &mgr->mm;
 	struct drm_mm_node *nodes;
 	enum drm_mm_insert_mode mode;
 	unsigned long lpfn, num_nodes, pages_per_node, pages_left;
+	uint64_t usage = 0, vis_usage = 0;
 	unsigned i;
 	int r;
 
@@ -142,6 +167,9 @@ static int amdgpu_vram_mgr_new(struct ttm_mem_type_manager *man,
 		if (unlikely(r))
 			goto error;
 
+		usage += nodes[i].size << PAGE_SHIFT;
+		vis_usage += amdgpu_vram_mgr_vis_size(adev, &nodes[i]);
+
 		/* Calculate a virtual BO start address to easily check if
 		 * everything is CPU accessible.
 		 */
@@ -155,6 +183,9 @@ static int amdgpu_vram_mgr_new(struct ttm_mem_type_manager *man,
 	}
 	spin_unlock(&mgr->lock);
 
+	atomic64_add(usage, &mgr->usage);
+	atomic64_add(vis_usage, &mgr->vis_usage);
+
 	mem->mm_node = nodes;
 
 	return 0;
@@ -181,8 +212,10 @@ error:
 static void amdgpu_vram_mgr_del(struct ttm_mem_type_manager *man,
 				struct ttm_mem_reg *mem)
 {
+	struct amdgpu_device *adev = amdgpu_ttm_adev(man->bdev);
 	struct amdgpu_vram_mgr *mgr = man->priv;
 	struct drm_mm_node *nodes = mem->mm_node;
+	uint64_t usage = 0, vis_usage = 0;
 	unsigned pages = mem->num_pages;
 
 	if (!mem->mm_node)
@@ -192,31 +225,67 @@ static void amdgpu_vram_mgr_del(struct ttm_mem_type_manager *man,
 	while (pages) {
 		pages -= nodes->size;
 		drm_mm_remove_node(nodes);
+		usage += nodes->size << PAGE_SHIFT;
+		vis_usage += amdgpu_vram_mgr_vis_size(adev, nodes);
 		++nodes;
 	}
 	spin_unlock(&mgr->lock);
 
+	atomic64_sub(usage, &mgr->usage);
+	atomic64_sub(vis_usage, &mgr->vis_usage);
+
 	kfree(mem->mm_node);
 	mem->mm_node = NULL;
 }
 
 /**
+ * amdgpu_vram_mgr_usage - how many bytes are used in this domain
+ *
+ * @man: TTM memory type manager
+ *
+ * Returns how many bytes are used in this domain.
+ */
+uint64_t amdgpu_vram_mgr_usage(struct ttm_mem_type_manager *man)
+{
+	struct amdgpu_vram_mgr *mgr = man->priv;
+
+	return atomic64_read(&mgr->usage);
+}
+
+/**
+ * amdgpu_vram_mgr_vis_usage - how many bytes are used in the visible part
+ *
+ * @man: TTM memory type manager
+ *
+ * Returns how many bytes are used in the visible part of VRAM
+ */
+uint64_t amdgpu_vram_mgr_vis_usage(struct ttm_mem_type_manager *man)
+{
+	struct amdgpu_vram_mgr *mgr = man->priv;
+
+	return atomic64_read(&mgr->vis_usage);
+}
+
+/**
  * amdgpu_vram_mgr_debug - dump VRAM table
  *
  * @man: TTM memory type manager
- * @prefix: text prefix
+ * @printer: DRM printer to use
  *
  * Dump the table content using printk.
  */
 static void amdgpu_vram_mgr_debug(struct ttm_mem_type_manager *man,
-				  const char *prefix)
+				  struct drm_printer *printer)
 {
 	struct amdgpu_vram_mgr *mgr = man->priv;
-	struct drm_printer p = drm_debug_printer(prefix);
 
 	spin_lock(&mgr->lock);
-	drm_mm_print(&mgr->mm, &p);
+	drm_mm_print(&mgr->mm, printer);
 	spin_unlock(&mgr->lock);
+
+	drm_printf(printer, "man size:%llu pages, ram usage:%lluMB, vis usage:%lluMB\n",
+		   man->size, amdgpu_vram_mgr_usage(man) >> 20,
+		   amdgpu_vram_mgr_vis_usage(man) >> 20);
 }
 
 const struct ttm_mem_type_manager_func amdgpu_vram_mgr_func = {
diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
index 490e84944851..4e519dc42916 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
@@ -2431,7 +2431,7 @@ static int dce_v10_0_crtc_cursor_set2(struct drm_crtc *crtc,
 	aobj = gem_to_amdgpu_bo(obj);
 	ret = amdgpu_bo_reserve(aobj, false);
 	if (ret != 0) {
-		drm_gem_object_unreference_unlocked(obj);
+		drm_gem_object_put_unlocked(obj);
 		return ret;
 	}
 
@@ -2439,7 +2439,7 @@ static int dce_v10_0_crtc_cursor_set2(struct drm_crtc *crtc,
 	amdgpu_bo_unreserve(aobj);
 	if (ret) {
 		DRM_ERROR("Failed to pin new cursor BO (%d)\n", ret);
-		drm_gem_object_unreference_unlocked(obj);
+		drm_gem_object_put_unlocked(obj);
 		return ret;
 	}
 
@@ -2473,7 +2473,7 @@ unpin:
 			amdgpu_bo_unpin(aobj);
 			amdgpu_bo_unreserve(aobj);
 		}
-		drm_gem_object_unreference_unlocked(amdgpu_crtc->cursor_bo);
+		drm_gem_object_put_unlocked(amdgpu_crtc->cursor_bo);
 	}
 
 	amdgpu_crtc->cursor_bo = obj;
diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
index 921c6f772f11..11edc75edaa9 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
@@ -2506,7 +2506,7 @@ static int dce_v11_0_crtc_cursor_set2(struct drm_crtc *crtc,
 	aobj = gem_to_amdgpu_bo(obj);
 	ret = amdgpu_bo_reserve(aobj, false);
 	if (ret != 0) {
-		drm_gem_object_unreference_unlocked(obj);
+		drm_gem_object_put_unlocked(obj);
 		return ret;
 	}
 
@@ -2514,7 +2514,7 @@ static int dce_v11_0_crtc_cursor_set2(struct drm_crtc *crtc,
 	amdgpu_bo_unreserve(aobj);
 	if (ret) {
 		DRM_ERROR("Failed to pin new cursor BO (%d)\n", ret);
-		drm_gem_object_unreference_unlocked(obj);
+		drm_gem_object_put_unlocked(obj);
 		return ret;
 	}
 
@@ -2548,7 +2548,7 @@ unpin:
 			amdgpu_bo_unpin(aobj);
 			amdgpu_bo_unreserve(aobj);
 		}
-		drm_gem_object_unreference_unlocked(amdgpu_crtc->cursor_bo);
+		drm_gem_object_put_unlocked(amdgpu_crtc->cursor_bo);
 	}
 
 	amdgpu_crtc->cursor_bo = obj;
diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c
index bcd9521237f4..a51e35f824a1 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c
@@ -42,6 +42,7 @@
 #include "dce/dce_6_0_d.h"
 #include "dce/dce_6_0_sh_mask.h"
 #include "gca/gfx_7_2_enum.h"
+#include "dce_v6_0.h"
 #include "si_enums.h"
 
 static void dce_v6_0_set_display_funcs(struct amdgpu_device *adev);
@@ -2321,7 +2322,7 @@ static int dce_v6_0_crtc_cursor_set2(struct drm_crtc *crtc,
 	aobj = gem_to_amdgpu_bo(obj);
 	ret = amdgpu_bo_reserve(aobj, false);
 	if (ret != 0) {
-		drm_gem_object_unreference_unlocked(obj);
+		drm_gem_object_put_unlocked(obj);
 		return ret;
 	}
 
@@ -2329,7 +2330,7 @@ static int dce_v6_0_crtc_cursor_set2(struct drm_crtc *crtc,
 	amdgpu_bo_unreserve(aobj);
 	if (ret) {
 		DRM_ERROR("Failed to pin new cursor BO (%d)\n", ret);
-		drm_gem_object_unreference_unlocked(obj);
+		drm_gem_object_put_unlocked(obj);
 		return ret;
 	}
 
@@ -2363,7 +2364,7 @@ unpin:
 			amdgpu_bo_unpin(aobj);
 			amdgpu_bo_unreserve(aobj);
 		}
-		drm_gem_object_unreference_unlocked(amdgpu_crtc->cursor_bo);
+		drm_gem_object_put_unlocked(amdgpu_crtc->cursor_bo);
 	}
 
 	amdgpu_crtc->cursor_bo = obj;
diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c
index 609438fe8584..9cf14b8b2db9 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c
@@ -2335,7 +2335,7 @@ static int dce_v8_0_crtc_cursor_set2(struct drm_crtc *crtc,
 	aobj = gem_to_amdgpu_bo(obj);
 	ret = amdgpu_bo_reserve(aobj, false);
 	if (ret != 0) {
-		drm_gem_object_unreference_unlocked(obj);
+		drm_gem_object_put_unlocked(obj);
 		return ret;
 	}
 
@@ -2343,7 +2343,7 @@ static int dce_v8_0_crtc_cursor_set2(struct drm_crtc *crtc,
 	amdgpu_bo_unreserve(aobj);
 	if (ret) {
 		DRM_ERROR("Failed to pin new cursor BO (%d)\n", ret);
-		drm_gem_object_unreference_unlocked(obj);
+		drm_gem_object_put_unlocked(obj);
 		return ret;
 	}
 
@@ -2377,7 +2377,7 @@ unpin:
 			amdgpu_bo_unpin(aobj);
 			amdgpu_bo_unreserve(aobj);
 		}
-		drm_gem_object_unreference_unlocked(amdgpu_crtc->cursor_bo);
+		drm_gem_object_put_unlocked(amdgpu_crtc->cursor_bo);
 	}
 
 	amdgpu_crtc->cursor_bo = obj;
diff --git a/drivers/gpu/drm/amd/amdgpu/dce_virtual.c b/drivers/gpu/drm/amd/amdgpu/dce_virtual.c
index 5ed919e45351..b9ee9073cb0d 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_virtual.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_virtual.c
@@ -479,6 +479,8 @@ static int dce_virtual_hw_init(void *handle)
 #endif
 		/* no DCE */
 		break;
+	case CHIP_VEGA10:
+		break;
 	default:
 		DRM_ERROR("Virtual display unsupported ASIC type: 0x%X\n", adev->asic_type);
 	}
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c
index 4ac85f47f287..d228f5a99044 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c
@@ -2217,40 +2217,9 @@ static void gfx_v6_0_ring_emit_vm_flush(struct amdgpu_ring *ring,
 
 static void gfx_v6_0_rlc_fini(struct amdgpu_device *adev)
 {
-	int r;
-
-	if (adev->gfx.rlc.save_restore_obj) {
-		r = amdgpu_bo_reserve(adev->gfx.rlc.save_restore_obj, true);
-		if (unlikely(r != 0))
-			dev_warn(adev->dev, "(%d) reserve RLC sr bo failed\n", r);
-		amdgpu_bo_unpin(adev->gfx.rlc.save_restore_obj);
-		amdgpu_bo_unreserve(adev->gfx.rlc.save_restore_obj);
-
-		amdgpu_bo_unref(&adev->gfx.rlc.save_restore_obj);
-		adev->gfx.rlc.save_restore_obj = NULL;
-	}
-
-	if (adev->gfx.rlc.clear_state_obj) {
-		r = amdgpu_bo_reserve(adev->gfx.rlc.clear_state_obj, true);
-		if (unlikely(r != 0))
-			dev_warn(adev->dev, "(%d) reserve RLC c bo failed\n", r);
-		amdgpu_bo_unpin(adev->gfx.rlc.clear_state_obj);
-		amdgpu_bo_unreserve(adev->gfx.rlc.clear_state_obj);
-
-		amdgpu_bo_unref(&adev->gfx.rlc.clear_state_obj);
-		adev->gfx.rlc.clear_state_obj = NULL;
-	}
-
-	if (adev->gfx.rlc.cp_table_obj) {
-		r = amdgpu_bo_reserve(adev->gfx.rlc.cp_table_obj, true);
-		if (unlikely(r != 0))
-			dev_warn(adev->dev, "(%d) reserve RLC cp table bo failed\n", r);
-		amdgpu_bo_unpin(adev->gfx.rlc.cp_table_obj);
-		amdgpu_bo_unreserve(adev->gfx.rlc.cp_table_obj);
-
-		amdgpu_bo_unref(&adev->gfx.rlc.cp_table_obj);
-		adev->gfx.rlc.cp_table_obj = NULL;
-	}
+	amdgpu_bo_free_kernel(&adev->gfx.rlc.save_restore_obj, NULL, NULL);
+	amdgpu_bo_free_kernel(&adev->gfx.rlc.clear_state_obj, NULL, NULL);
+	amdgpu_bo_free_kernel(&adev->gfx.rlc.cp_table_obj, NULL, NULL);
 }
 
 static int gfx_v6_0_rlc_init(struct amdgpu_device *adev)
@@ -2273,43 +2242,23 @@ static int gfx_v6_0_rlc_init(struct amdgpu_device *adev)
 
 	if (src_ptr) {
 		/* save restore block */
-		if (adev->gfx.rlc.save_restore_obj == NULL) {
-			r = amdgpu_bo_create(adev, dws * 4, PAGE_SIZE, true,
-					     AMDGPU_GEM_DOMAIN_VRAM,
-					     AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED,
-					     NULL, NULL,
-					     &adev->gfx.rlc.save_restore_obj);
-
-			if (r) {
-				dev_warn(adev->dev, "(%d) create RLC sr bo failed\n", r);
-				return r;
-			}
-		}
-
-		r = amdgpu_bo_reserve(adev->gfx.rlc.save_restore_obj, false);
-		if (unlikely(r != 0)) {
-			gfx_v6_0_rlc_fini(adev);
-			return r;
-		}
-		r = amdgpu_bo_pin(adev->gfx.rlc.save_restore_obj, AMDGPU_GEM_DOMAIN_VRAM,
-				  &adev->gfx.rlc.save_restore_gpu_addr);
+		r = amdgpu_bo_create_reserved(adev, dws * 4, PAGE_SIZE,
+					      AMDGPU_GEM_DOMAIN_VRAM,
+					      &adev->gfx.rlc.save_restore_obj,
+					      &adev->gfx.rlc.save_restore_gpu_addr,
+					      (void **)&adev->gfx.rlc.sr_ptr);
 		if (r) {
-			amdgpu_bo_unreserve(adev->gfx.rlc.save_restore_obj);
-			dev_warn(adev->dev, "(%d) pin RLC sr bo failed\n", r);
+			dev_warn(adev->dev, "(%d) create RLC sr bo failed\n",
+				 r);
 			gfx_v6_0_rlc_fini(adev);
 			return r;
 		}
 
-		r = amdgpu_bo_kmap(adev->gfx.rlc.save_restore_obj, (void **)&adev->gfx.rlc.sr_ptr);
-		if (r) {
-			dev_warn(adev->dev, "(%d) map RLC sr bo failed\n", r);
-			gfx_v6_0_rlc_fini(adev);
-			return r;
-		}
 		/* write the sr buffer */
 		dst_ptr = adev->gfx.rlc.sr_ptr;
 		for (i = 0; i < adev->gfx.rlc.reg_list_size; i++)
 			dst_ptr[i] = cpu_to_le32(src_ptr[i]);
+
 		amdgpu_bo_kunmap(adev->gfx.rlc.save_restore_obj);
 		amdgpu_bo_unreserve(adev->gfx.rlc.save_restore_obj);
 	}
@@ -2319,39 +2268,17 @@ static int gfx_v6_0_rlc_init(struct amdgpu_device *adev)
 		adev->gfx.rlc.clear_state_size = gfx_v6_0_get_csb_size(adev);
 		dws = adev->gfx.rlc.clear_state_size + (256 / 4);
 
-		if (adev->gfx.rlc.clear_state_obj == NULL) {
-			r = amdgpu_bo_create(adev, dws * 4, PAGE_SIZE, true,
-					     AMDGPU_GEM_DOMAIN_VRAM,
-					     AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED,
-					     NULL, NULL,
-					     &adev->gfx.rlc.clear_state_obj);
-
-			if (r) {
-				dev_warn(adev->dev, "(%d) create RLC c bo failed\n", r);
-				gfx_v6_0_rlc_fini(adev);
-				return r;
-			}
-		}
-		r = amdgpu_bo_reserve(adev->gfx.rlc.clear_state_obj, false);
-		if (unlikely(r != 0)) {
-			gfx_v6_0_rlc_fini(adev);
-			return r;
-		}
-		r = amdgpu_bo_pin(adev->gfx.rlc.clear_state_obj, AMDGPU_GEM_DOMAIN_VRAM,
-				  &adev->gfx.rlc.clear_state_gpu_addr);
+		r = amdgpu_bo_create_reserved(adev, dws * 4, PAGE_SIZE,
+					      AMDGPU_GEM_DOMAIN_VRAM,
+					      &adev->gfx.rlc.clear_state_obj,
+					      &adev->gfx.rlc.clear_state_gpu_addr,
+					      (void **)&adev->gfx.rlc.cs_ptr);
 		if (r) {
-			amdgpu_bo_unreserve(adev->gfx.rlc.clear_state_obj);
-			dev_warn(adev->dev, "(%d) pin RLC c bo failed\n", r);
+			dev_warn(adev->dev, "(%d) create RLC c bo failed\n", r);
 			gfx_v6_0_rlc_fini(adev);
 			return r;
 		}
 
-		r = amdgpu_bo_kmap(adev->gfx.rlc.clear_state_obj, (void **)&adev->gfx.rlc.cs_ptr);
-		if (r) {
-			dev_warn(adev->dev, "(%d) map RLC c bo failed\n", r);
-			gfx_v6_0_rlc_fini(adev);
-			return r;
-		}
 		/* set up the cs buffer */
 		dst_ptr = adev->gfx.rlc.cs_ptr;
 		reg_list_mc_addr = adev->gfx.rlc.clear_state_gpu_addr + 256;
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
index 17b7c6934b0a..00868764a0dd 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
@@ -1823,7 +1823,7 @@ static void gfx_v7_0_setup_rb(struct amdgpu_device *adev)
 }
 
 /**
- * gmc_v7_0_init_compute_vmid - gart enable
+ * gfx_v7_0_init_compute_vmid - gart enable
  *
  * @adev: amdgpu_device pointer
  *
@@ -1833,7 +1833,7 @@ static void gfx_v7_0_setup_rb(struct amdgpu_device *adev)
 #define DEFAULT_SH_MEM_BASES	(0x6000)
 #define FIRST_COMPUTE_VMID	(8)
 #define LAST_COMPUTE_VMID	(16)
-static void gmc_v7_0_init_compute_vmid(struct amdgpu_device *adev)
+static void gfx_v7_0_init_compute_vmid(struct amdgpu_device *adev)
 {
 	int i;
 	uint32_t sh_mem_config;
@@ -1921,6 +1921,7 @@ static void gfx_v7_0_gpu_init(struct amdgpu_device *adev)
 				   ELEMENT_SIZE, 1);
 	sh_static_mem_cfg = REG_SET_FIELD(sh_static_mem_cfg, SH_STATIC_MEM_CONFIG,
 				   INDEX_STRIDE, 3);
+	WREG32(mmSH_STATIC_MEM_CONFIG, sh_static_mem_cfg);
 
 	mutex_lock(&adev->srbm_mutex);
 	for (i = 0; i < adev->vm_manager.id_mgr[0].num_ids; i++) {
@@ -1934,12 +1935,11 @@ static void gfx_v7_0_gpu_init(struct amdgpu_device *adev)
 		WREG32(mmSH_MEM_APE1_BASE, 1);
 		WREG32(mmSH_MEM_APE1_LIMIT, 0);
 		WREG32(mmSH_MEM_BASES, sh_mem_base);
-		WREG32(mmSH_STATIC_MEM_CONFIG, sh_static_mem_cfg);
 	}
 	cik_srbm_select(adev, 0, 0, 0, 0);
 	mutex_unlock(&adev->srbm_mutex);
 
-	gmc_v7_0_init_compute_vmid(adev);
+	gfx_v7_0_init_compute_vmid(adev);
 
 	WREG32(mmSX_DEBUG_1, 0x20);
 
@@ -2774,39 +2774,18 @@ static int gfx_v7_0_cp_compute_load_microcode(struct amdgpu_device *adev)
  */
 static void gfx_v7_0_cp_compute_fini(struct amdgpu_device *adev)
 {
-	int i, r;
+	int i;
 
 	for (i = 0; i < adev->gfx.num_compute_rings; i++) {
 		struct amdgpu_ring *ring = &adev->gfx.compute_ring[i];
 
-		if (ring->mqd_obj) {
-			r = amdgpu_bo_reserve(ring->mqd_obj, true);
-			if (unlikely(r != 0))
-				dev_warn(adev->dev, "(%d) reserve MQD bo failed\n", r);
-
-			amdgpu_bo_unpin(ring->mqd_obj);
-			amdgpu_bo_unreserve(ring->mqd_obj);
-
-			amdgpu_bo_unref(&ring->mqd_obj);
-			ring->mqd_obj = NULL;
-		}
+		amdgpu_bo_free_kernel(&ring->mqd_obj, NULL, NULL);
 	}
 }
 
 static void gfx_v7_0_mec_fini(struct amdgpu_device *adev)
 {
-	int r;
-
-	if (adev->gfx.mec.hpd_eop_obj) {
-		r = amdgpu_bo_reserve(adev->gfx.mec.hpd_eop_obj, true);
-		if (unlikely(r != 0))
-			dev_warn(adev->dev, "(%d) reserve HPD EOP bo failed\n", r);
-		amdgpu_bo_unpin(adev->gfx.mec.hpd_eop_obj);
-		amdgpu_bo_unreserve(adev->gfx.mec.hpd_eop_obj);
-
-		amdgpu_bo_unref(&adev->gfx.mec.hpd_eop_obj);
-		adev->gfx.mec.hpd_eop_obj = NULL;
-	}
+	amdgpu_bo_free_kernel(&adev->gfx.mec.hpd_eop_obj, NULL, NULL);
 }
 
 static int gfx_v7_0_mec_init(struct amdgpu_device *adev)
@@ -2823,33 +2802,14 @@ static int gfx_v7_0_mec_init(struct amdgpu_device *adev)
 	/* allocate space for ALL pipes (even the ones we don't own) */
 	mec_hpd_size = adev->gfx.mec.num_mec * adev->gfx.mec.num_pipe_per_mec
 		* GFX7_MEC_HPD_SIZE * 2;
-	if (adev->gfx.mec.hpd_eop_obj == NULL) {
-		r = amdgpu_bo_create(adev,
-				     mec_hpd_size,
-				     PAGE_SIZE, true,
-				     AMDGPU_GEM_DOMAIN_GTT, 0, NULL, NULL,
-				     &adev->gfx.mec.hpd_eop_obj);
-		if (r) {
-			dev_warn(adev->dev, "(%d) create HDP EOP bo failed\n", r);
-			return r;
-		}
-	}
 
-	r = amdgpu_bo_reserve(adev->gfx.mec.hpd_eop_obj, false);
-	if (unlikely(r != 0)) {
-		gfx_v7_0_mec_fini(adev);
-		return r;
-	}
-	r = amdgpu_bo_pin(adev->gfx.mec.hpd_eop_obj, AMDGPU_GEM_DOMAIN_GTT,
-			  &adev->gfx.mec.hpd_eop_gpu_addr);
-	if (r) {
-		dev_warn(adev->dev, "(%d) pin HDP EOP bo failed\n", r);
-		gfx_v7_0_mec_fini(adev);
-		return r;
-	}
-	r = amdgpu_bo_kmap(adev->gfx.mec.hpd_eop_obj, (void **)&hpd);
+	r = amdgpu_bo_create_reserved(adev, mec_hpd_size, PAGE_SIZE,
+				      AMDGPU_GEM_DOMAIN_GTT,
+				      &adev->gfx.mec.hpd_eop_obj,
+				      &adev->gfx.mec.hpd_eop_gpu_addr,
+				      (void **)&hpd);
 	if (r) {
-		dev_warn(adev->dev, "(%d) map HDP EOP bo failed\n", r);
+		dev_warn(adev->dev, "(%d) create, pin or map of HDP EOP bo failed\n", r);
 		gfx_v7_0_mec_fini(adev);
 		return r;
 	}
@@ -3108,32 +3068,12 @@ static int gfx_v7_0_compute_queue_init(struct amdgpu_device *adev, int ring_id)
 	struct cik_mqd *mqd;
 	struct amdgpu_ring *ring = &adev->gfx.compute_ring[ring_id];
 
-	if (ring->mqd_obj == NULL) {
-		r = amdgpu_bo_create(adev,
-				sizeof(struct cik_mqd),
-				PAGE_SIZE, true,
-				AMDGPU_GEM_DOMAIN_GTT, 0, NULL, NULL,
-				&ring->mqd_obj);
-		if (r) {
-			dev_warn(adev->dev, "(%d) create MQD bo failed\n", r);
-			return r;
-		}
-	}
-
-	r = amdgpu_bo_reserve(ring->mqd_obj, false);
-	if (unlikely(r != 0))
-		goto out;
-
-	r = amdgpu_bo_pin(ring->mqd_obj, AMDGPU_GEM_DOMAIN_GTT,
-			&mqd_gpu_addr);
-	if (r) {
-		dev_warn(adev->dev, "(%d) pin MQD bo failed\n", r);
-		goto out_unreserve;
-	}
-	r = amdgpu_bo_kmap(ring->mqd_obj, (void **)&mqd);
+	r = amdgpu_bo_create_reserved(adev, sizeof(struct cik_mqd), PAGE_SIZE,
+				      AMDGPU_GEM_DOMAIN_GTT, &ring->mqd_obj,
+				      &mqd_gpu_addr, (void **)&mqd);
 	if (r) {
-		dev_warn(adev->dev, "(%d) map MQD bo failed\n", r);
-		goto out_unreserve;
+		dev_warn(adev->dev, "(%d) create MQD bo failed\n", r);
+		return r;
 	}
 
 	mutex_lock(&adev->srbm_mutex);
@@ -3147,9 +3087,7 @@ static int gfx_v7_0_compute_queue_init(struct amdgpu_device *adev, int ring_id)
 	mutex_unlock(&adev->srbm_mutex);
 
 	amdgpu_bo_kunmap(ring->mqd_obj);
-out_unreserve:
 	amdgpu_bo_unreserve(ring->mqd_obj);
-out:
 	return 0;
 }
 
@@ -3361,43 +3299,9 @@ static void gfx_v7_0_ring_emit_vm_flush(struct amdgpu_ring *ring,
  */
 static void gfx_v7_0_rlc_fini(struct amdgpu_device *adev)
 {
-	int r;
-
-	/* save restore block */
-	if (adev->gfx.rlc.save_restore_obj) {
-		r = amdgpu_bo_reserve(adev->gfx.rlc.save_restore_obj, true);
-		if (unlikely(r != 0))
-			dev_warn(adev->dev, "(%d) reserve RLC sr bo failed\n", r);
-		amdgpu_bo_unpin(adev->gfx.rlc.save_restore_obj);
-		amdgpu_bo_unreserve(adev->gfx.rlc.save_restore_obj);
-
-		amdgpu_bo_unref(&adev->gfx.rlc.save_restore_obj);
-		adev->gfx.rlc.save_restore_obj = NULL;
-	}
-
-	/* clear state block */
-	if (adev->gfx.rlc.clear_state_obj) {
-		r = amdgpu_bo_reserve(adev->gfx.rlc.clear_state_obj, true);
-		if (unlikely(r != 0))
-			dev_warn(adev->dev, "(%d) reserve RLC c bo failed\n", r);
-		amdgpu_bo_unpin(adev->gfx.rlc.clear_state_obj);
-		amdgpu_bo_unreserve(adev->gfx.rlc.clear_state_obj);
-
-		amdgpu_bo_unref(&adev->gfx.rlc.clear_state_obj);
-		adev->gfx.rlc.clear_state_obj = NULL;
-	}
-
-	/* clear state block */
-	if (adev->gfx.rlc.cp_table_obj) {
-		r = amdgpu_bo_reserve(adev->gfx.rlc.cp_table_obj, true);
-		if (unlikely(r != 0))
-			dev_warn(adev->dev, "(%d) reserve RLC cp table bo failed\n", r);
-		amdgpu_bo_unpin(adev->gfx.rlc.cp_table_obj);
-		amdgpu_bo_unreserve(adev->gfx.rlc.cp_table_obj);
-
-		amdgpu_bo_unref(&adev->gfx.rlc.cp_table_obj);
-		adev->gfx.rlc.cp_table_obj = NULL;
-	}
+	amdgpu_bo_free_kernel(&adev->gfx.rlc.save_restore_obj, NULL, NULL);
+	amdgpu_bo_free_kernel(&adev->gfx.rlc.clear_state_obj, NULL, NULL);
+	amdgpu_bo_free_kernel(&adev->gfx.rlc.cp_table_obj, NULL, NULL);
 }
 
 static int gfx_v7_0_rlc_init(struct amdgpu_device *adev)
@@ -3432,39 +3336,17 @@ static int gfx_v7_0_rlc_init(struct amdgpu_device *adev)
 
 	if (src_ptr) {
 		/* save restore block */
-		if (adev->gfx.rlc.save_restore_obj == NULL) {
-			r = amdgpu_bo_create(adev, dws * 4, PAGE_SIZE, true,
-					     AMDGPU_GEM_DOMAIN_VRAM,
-					     AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED |
-					     AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS,
-					     NULL, NULL,
-					     &adev->gfx.rlc.save_restore_obj);
-			if (r) {
-				dev_warn(adev->dev, "(%d) create RLC sr bo failed\n", r);
-				return r;
-			}
-		}
-
-		r = amdgpu_bo_reserve(adev->gfx.rlc.save_restore_obj, false);
-		if (unlikely(r != 0)) {
-			gfx_v7_0_rlc_fini(adev);
-			return r;
-		}
-		r = amdgpu_bo_pin(adev->gfx.rlc.save_restore_obj, AMDGPU_GEM_DOMAIN_VRAM,
-				  &adev->gfx.rlc.save_restore_gpu_addr);
+		r = amdgpu_bo_create_reserved(adev, dws * 4, PAGE_SIZE,
+					      AMDGPU_GEM_DOMAIN_VRAM,
+					      &adev->gfx.rlc.save_restore_obj,
+					      &adev->gfx.rlc.save_restore_gpu_addr,
+					      (void **)&adev->gfx.rlc.sr_ptr);
 		if (r) {
-			amdgpu_bo_unreserve(adev->gfx.rlc.save_restore_obj);
-			dev_warn(adev->dev, "(%d) pin RLC sr bo failed\n", r);
+			dev_warn(adev->dev, "(%d) create, pin or map of RLC sr bo failed\n", r);
 			gfx_v7_0_rlc_fini(adev);
 			return r;
 		}
 
-		r = amdgpu_bo_kmap(adev->gfx.rlc.save_restore_obj, (void **)&adev->gfx.rlc.sr_ptr);
-		if (r) {
-			dev_warn(adev->dev, "(%d) map RLC sr bo failed\n", r);
-			gfx_v7_0_rlc_fini(adev);
-			return r;
-		}
 		/* write the sr buffer */
 		dst_ptr = adev->gfx.rlc.sr_ptr;
 		for (i = 0; i < adev->gfx.rlc.reg_list_size; i++)
@@ -3477,39 +3359,17 @@ static int gfx_v7_0_rlc_init(struct amdgpu_device *adev)
 		/* clear state block */
 		adev->gfx.rlc.clear_state_size = dws = gfx_v7_0_get_csb_size(adev);
 
-		if (adev->gfx.rlc.clear_state_obj == NULL) {
-			r = amdgpu_bo_create(adev, dws * 4, PAGE_SIZE, true,
-					     AMDGPU_GEM_DOMAIN_VRAM,
-					     AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED |
-					     AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS,
-					     NULL, NULL,
-					     &adev->gfx.rlc.clear_state_obj);
-			if (r) {
-				dev_warn(adev->dev, "(%d) create RLC c bo failed\n", r);
-				gfx_v7_0_rlc_fini(adev);
-				return r;
-			}
-		}
-		r = amdgpu_bo_reserve(adev->gfx.rlc.clear_state_obj, false);
-		if (unlikely(r != 0)) {
-			gfx_v7_0_rlc_fini(adev);
-			return r;
-		}
-		r = amdgpu_bo_pin(adev->gfx.rlc.clear_state_obj, AMDGPU_GEM_DOMAIN_VRAM,
-				  &adev->gfx.rlc.clear_state_gpu_addr);
+		r = amdgpu_bo_create_reserved(adev, dws * 4, PAGE_SIZE,
+					      AMDGPU_GEM_DOMAIN_VRAM,
+					      &adev->gfx.rlc.clear_state_obj,
+					      &adev->gfx.rlc.clear_state_gpu_addr,
+					      (void **)&adev->gfx.rlc.cs_ptr);
 		if (r) {
-			amdgpu_bo_unreserve(adev->gfx.rlc.clear_state_obj);
-			dev_warn(adev->dev, "(%d) pin RLC c bo failed\n", r);
+			dev_warn(adev->dev, "(%d) create RLC c bo failed\n", r);
 			gfx_v7_0_rlc_fini(adev);
 			return r;
 		}
 
-		r = amdgpu_bo_kmap(adev->gfx.rlc.clear_state_obj, (void **)&adev->gfx.rlc.cs_ptr);
-		if (r) {
-			dev_warn(adev->dev, "(%d) map RLC c bo failed\n", r);
-			gfx_v7_0_rlc_fini(adev);
-			return r;
-		}
 		/* set up the cs buffer */
 		dst_ptr = adev->gfx.rlc.cs_ptr;
 		gfx_v7_0_get_csb_buffer(adev, dst_ptr);
@@ -3518,37 +3378,14 @@ static int gfx_v7_0_rlc_init(struct amdgpu_device *adev)
 	}
 
 	if (adev->gfx.rlc.cp_table_size) {
-		if (adev->gfx.rlc.cp_table_obj == NULL) {
-			r = amdgpu_bo_create(adev, adev->gfx.rlc.cp_table_size, PAGE_SIZE, true,
-					     AMDGPU_GEM_DOMAIN_VRAM,
-					     AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED |
-					     AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS,
-					     NULL, NULL,
-					     &adev->gfx.rlc.cp_table_obj);
-			if (r) {
-				dev_warn(adev->dev, "(%d) create RLC cp table bo failed\n", r);
-				gfx_v7_0_rlc_fini(adev);
-				return r;
-			}
-		}
 
-		r = amdgpu_bo_reserve(adev->gfx.rlc.cp_table_obj, false);
-		if (unlikely(r != 0)) {
-			dev_warn(adev->dev, "(%d) reserve RLC cp table bo failed\n", r);
-			gfx_v7_0_rlc_fini(adev);
-			return r;
-		}
-		r = amdgpu_bo_pin(adev->gfx.rlc.cp_table_obj, AMDGPU_GEM_DOMAIN_VRAM,
-				  &adev->gfx.rlc.cp_table_gpu_addr);
-		if (r) {
-			amdgpu_bo_unreserve(adev->gfx.rlc.cp_table_obj);
-			dev_warn(adev->dev, "(%d) pin RLC cp_table bo failed\n", r);
-			gfx_v7_0_rlc_fini(adev);
-			return r;
-		}
-		r = amdgpu_bo_kmap(adev->gfx.rlc.cp_table_obj, (void **)&adev->gfx.rlc.cp_table_ptr);
+		r = amdgpu_bo_create_reserved(adev, adev->gfx.rlc.cp_table_size,
+					      PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM,
+					      &adev->gfx.rlc.cp_table_obj,
+					      &adev->gfx.rlc.cp_table_gpu_addr,
+					      (void **)&adev->gfx.rlc.cp_table_ptr);
 		if (r) {
-			dev_warn(adev->dev, "(%d) map RLC cp table bo failed\n", r);
+			dev_warn(adev->dev, "(%d) create RLC cp table bo failed\n", r);
 			gfx_v7_0_rlc_fini(adev);
 			return r;
 		}
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
index 05436b8730b4..832e592fcd07 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
@@ -1238,29 +1238,8 @@ static void cz_init_cp_jump_table(struct amdgpu_device *adev)
 
 static void gfx_v8_0_rlc_fini(struct amdgpu_device *adev)
 {
-	int r;
-
-	/* clear state block */
-	if (adev->gfx.rlc.clear_state_obj) {
-		r = amdgpu_bo_reserve(adev->gfx.rlc.clear_state_obj, true);
-		if (unlikely(r != 0))
-			dev_warn(adev->dev, "(%d) reserve RLC cbs bo failed\n", r);
-		amdgpu_bo_unpin(adev->gfx.rlc.clear_state_obj);
-		amdgpu_bo_unreserve(adev->gfx.rlc.clear_state_obj);
-		amdgpu_bo_unref(&adev->gfx.rlc.clear_state_obj);
-		adev->gfx.rlc.clear_state_obj = NULL;
-	}
-
-	/* jump table block */
-	if (adev->gfx.rlc.cp_table_obj) {
-		r = amdgpu_bo_reserve(adev->gfx.rlc.cp_table_obj, true);
-		if (unlikely(r != 0))
-			dev_warn(adev->dev, "(%d) reserve RLC cp table bo failed\n", r);
-		amdgpu_bo_unpin(adev->gfx.rlc.cp_table_obj);
-		amdgpu_bo_unreserve(adev->gfx.rlc.cp_table_obj);
-		amdgpu_bo_unref(&adev->gfx.rlc.cp_table_obj);
-		adev->gfx.rlc.cp_table_obj = NULL;
-	}
+	amdgpu_bo_free_kernel(&adev->gfx.rlc.clear_state_obj, NULL, NULL);
+	amdgpu_bo_free_kernel(&adev->gfx.rlc.cp_table_obj, NULL, NULL);
 }
 
 static int gfx_v8_0_rlc_init(struct amdgpu_device *adev)
@@ -1278,39 +1257,17 @@ static int gfx_v8_0_rlc_init(struct amdgpu_device *adev)
 		/* clear state block */
 		adev->gfx.rlc.clear_state_size = dws = gfx_v8_0_get_csb_size(adev);
 
-		if (adev->gfx.rlc.clear_state_obj == NULL) {
-			r = amdgpu_bo_create(adev, dws * 4, PAGE_SIZE, true,
-					     AMDGPU_GEM_DOMAIN_VRAM,
-					     AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED |
-					     AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS,
-					     NULL, NULL,
-					     &adev->gfx.rlc.clear_state_obj);
-			if (r) {
-				dev_warn(adev->dev, "(%d) create RLC c bo failed\n", r);
-				gfx_v8_0_rlc_fini(adev);
-				return r;
-			}
-		}
-		r = amdgpu_bo_reserve(adev->gfx.rlc.clear_state_obj, false);
-		if (unlikely(r != 0)) {
-			gfx_v8_0_rlc_fini(adev);
-			return r;
-		}
-		r = amdgpu_bo_pin(adev->gfx.rlc.clear_state_obj, AMDGPU_GEM_DOMAIN_VRAM,
-				  &adev->gfx.rlc.clear_state_gpu_addr);
+		r = amdgpu_bo_create_reserved(adev, dws * 4, PAGE_SIZE,
+					      AMDGPU_GEM_DOMAIN_VRAM,
+					      &adev->gfx.rlc.clear_state_obj,
+					      &adev->gfx.rlc.clear_state_gpu_addr,
+					      (void **)&adev->gfx.rlc.cs_ptr);
 		if (r) {
-			amdgpu_bo_unreserve(adev->gfx.rlc.clear_state_obj);
-			dev_warn(adev->dev, "(%d) pin RLC cbs bo failed\n", r);
+			dev_warn(adev->dev, "(%d) create RLC c bo failed\n", r);
 			gfx_v8_0_rlc_fini(adev);
 			return r;
 		}
 
-		r = amdgpu_bo_kmap(adev->gfx.rlc.clear_state_obj, (void **)&adev->gfx.rlc.cs_ptr);
-		if (r) {
-			dev_warn(adev->dev, "(%d) map RLC cbs bo failed\n", r);
-			gfx_v8_0_rlc_fini(adev);
-			return r;
-		}
 		/* set up the cs buffer */
 		dst_ptr = adev->gfx.rlc.cs_ptr;
 		gfx_v8_0_get_csb_buffer(adev, dst_ptr);
@@ -1321,34 +1278,13 @@ static int gfx_v8_0_rlc_init(struct amdgpu_device *adev)
 	if ((adev->asic_type == CHIP_CARRIZO) ||
 	    (adev->asic_type == CHIP_STONEY)) {
 		adev->gfx.rlc.cp_table_size = ALIGN(96 * 5 * 4, 2048) + (64 * 1024); /* JT + GDS */
-		if (adev->gfx.rlc.cp_table_obj == NULL) {
-			r = amdgpu_bo_create(adev, adev->gfx.rlc.cp_table_size, PAGE_SIZE, true,
-					     AMDGPU_GEM_DOMAIN_VRAM,
-					     AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED |
-					     AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS,
-					     NULL, NULL,
-					     &adev->gfx.rlc.cp_table_obj);
-			if (r) {
-				dev_warn(adev->dev, "(%d) create RLC cp table bo failed\n", r);
-				return r;
-			}
-		}
-
-		r = amdgpu_bo_reserve(adev->gfx.rlc.cp_table_obj, false);
-		if (unlikely(r != 0)) {
-			dev_warn(adev->dev, "(%d) reserve RLC cp table bo failed\n", r);
-			return r;
-		}
-		r = amdgpu_bo_pin(adev->gfx.rlc.cp_table_obj, AMDGPU_GEM_DOMAIN_VRAM,
-				  &adev->gfx.rlc.cp_table_gpu_addr);
-		if (r) {
-			amdgpu_bo_unreserve(adev->gfx.rlc.cp_table_obj);
-			dev_warn(adev->dev, "(%d) pin RLC cp table bo failed\n", r);
-			return r;
-		}
-		r = amdgpu_bo_kmap(adev->gfx.rlc.cp_table_obj, (void **)&adev->gfx.rlc.cp_table_ptr);
+		r = amdgpu_bo_create_reserved(adev, adev->gfx.rlc.cp_table_size,
+					      PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM,
+					      &adev->gfx.rlc.cp_table_obj,
+					      &adev->gfx.rlc.cp_table_gpu_addr,
+					      (void **)&adev->gfx.rlc.cp_table_ptr);
 		if (r) {
-			dev_warn(adev->dev, "(%d) map RLC cp table bo failed\n", r);
+			dev_warn(adev->dev, "(%d) create RLC cp table bo failed\n", r);
 			return r;
 		}
 
@@ -1363,17 +1299,7 @@ static int gfx_v8_0_rlc_init(struct amdgpu_device *adev)
 
 static void gfx_v8_0_mec_fini(struct amdgpu_device *adev)
 {
-	int r;
-
-	if (adev->gfx.mec.hpd_eop_obj) {
-		r = amdgpu_bo_reserve(adev->gfx.mec.hpd_eop_obj, true);
-		if (unlikely(r != 0))
-			dev_warn(adev->dev, "(%d) reserve HPD EOP bo failed\n", r);
-		amdgpu_bo_unpin(adev->gfx.mec.hpd_eop_obj);
-		amdgpu_bo_unreserve(adev->gfx.mec.hpd_eop_obj);
-		amdgpu_bo_unref(&adev->gfx.mec.hpd_eop_obj);
-		adev->gfx.mec.hpd_eop_obj = NULL;
-	}
+	amdgpu_bo_free_kernel(&adev->gfx.mec.hpd_eop_obj, NULL, NULL);
 }
 
 static int gfx_v8_0_mec_init(struct amdgpu_device *adev)
@@ -1389,34 +1315,13 @@ static int gfx_v8_0_mec_init(struct amdgpu_device *adev)
 
 	mec_hpd_size = adev->gfx.num_compute_rings * GFX8_MEC_HPD_SIZE;
 
-	if (adev->gfx.mec.hpd_eop_obj == NULL) {
-		r = amdgpu_bo_create(adev,
-				     mec_hpd_size,
-				     PAGE_SIZE, true,
-				     AMDGPU_GEM_DOMAIN_GTT, 0, NULL, NULL,
-				     &adev->gfx.mec.hpd_eop_obj);
-		if (r) {
-			dev_warn(adev->dev, "(%d) create HDP EOP bo failed\n", r);
-			return r;
-		}
-	}
-
-	r = amdgpu_bo_reserve(adev->gfx.mec.hpd_eop_obj, false);
-	if (unlikely(r != 0)) {
-		gfx_v8_0_mec_fini(adev);
-		return r;
-	}
-	r = amdgpu_bo_pin(adev->gfx.mec.hpd_eop_obj, AMDGPU_GEM_DOMAIN_GTT,
-			  &adev->gfx.mec.hpd_eop_gpu_addr);
+	r = amdgpu_bo_create_reserved(adev, mec_hpd_size, PAGE_SIZE,
+				      AMDGPU_GEM_DOMAIN_GTT,
+				      &adev->gfx.mec.hpd_eop_obj,
+				      &adev->gfx.mec.hpd_eop_gpu_addr,
+				      (void **)&hpd);
 	if (r) {
-		dev_warn(adev->dev, "(%d) pin HDP EOP bo failed\n", r);
-		gfx_v8_0_mec_fini(adev);
-		return r;
-	}
-	r = amdgpu_bo_kmap(adev->gfx.mec.hpd_eop_obj, (void **)&hpd);
-	if (r) {
-		dev_warn(adev->dev, "(%d) map HDP EOP bo failed\n", r);
-		gfx_v8_0_mec_fini(adev);
+		dev_warn(adev->dev, "(%d) create HDP EOP bo failed\n", r);
 		return r;
 	}
 
@@ -3802,6 +3707,8 @@ static void gfx_v8_0_gpu_init(struct amdgpu_device *adev)
 				   ELEMENT_SIZE, 1);
 	sh_static_mem_cfg = REG_SET_FIELD(sh_static_mem_cfg, SH_STATIC_MEM_CONFIG,
 				   INDEX_STRIDE, 3);
+	WREG32(mmSH_STATIC_MEM_CONFIG, sh_static_mem_cfg);
+
 	mutex_lock(&adev->srbm_mutex);
 	for (i = 0; i < adev->vm_manager.id_mgr[0].num_ids; i++) {
 		vi_srbm_select(adev, 0, 0, 0, i);
@@ -3825,7 +3732,6 @@ static void gfx_v8_0_gpu_init(struct amdgpu_device *adev)
 
 		WREG32(mmSH_MEM_APE1_BASE, 1);
 		WREG32(mmSH_MEM_APE1_LIMIT, 0);
-		WREG32(mmSH_STATIC_MEM_CONFIG, sh_static_mem_cfg);
 	}
 	vi_srbm_select(adev, 0, 0, 0, 0);
 	mutex_unlock(&adev->srbm_mutex);
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
index 435db6f5efcf..69182eeca264 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
@@ -116,7 +116,9 @@ static const u32 golden_settings_gc_9_0[] =
 	SOC15_REG_OFFSET(GC, 0, mmRLC_GPM_UTCL1_CNTL_2), 0x08000000, 0x08000080,
 	SOC15_REG_OFFSET(GC, 0, mmRLC_PREWALKER_UTCL1_CNTL), 0x08000000, 0x08000080,
 	SOC15_REG_OFFSET(GC, 0, mmRLC_SPM_UTCL1_CNTL), 0x08000000, 0x08000080,
+	SOC15_REG_OFFSET(GC, 0, mmSH_MEM_CONFIG), 0x00001000, 0x00001000,
 	SOC15_REG_OFFSET(GC, 0, mmSPI_CONFIG_CNTL_1), 0x0000000f, 0x01000107,
+	SOC15_REG_OFFSET(GC, 0, mmSQC_CONFIG), 0x03000000, 0x020a2000,
 	SOC15_REG_OFFSET(GC, 0, mmTA_CNTL_AUX), 0xfffffeef, 0x010b0000,
 	SOC15_REG_OFFSET(GC, 0, mmTCP_CHAN_STEER_HI), 0xffffffff, 0x4a2c0e68,
 	SOC15_REG_OFFSET(GC, 0, mmTCP_CHAN_STEER_LO), 0xffffffff, 0xb5d3f197,
@@ -772,18 +774,16 @@ static int gfx_v9_0_rlc_init(struct amdgpu_device *adev)
 	if (cs_data) {
 		/* clear state block */
 		adev->gfx.rlc.clear_state_size = dws = gfx_v9_0_get_csb_size(adev);
-		if (adev->gfx.rlc.clear_state_obj == NULL) {
-			r = amdgpu_bo_create_kernel(adev, dws * 4, PAGE_SIZE,
-						AMDGPU_GEM_DOMAIN_VRAM,
-						&adev->gfx.rlc.clear_state_obj,
-						&adev->gfx.rlc.clear_state_gpu_addr,
-						(void **)&adev->gfx.rlc.cs_ptr);
-			if (r) {
-				dev_err(adev->dev,
-					"(%d) failed to create rlc csb bo\n", r);
-				gfx_v9_0_rlc_fini(adev);
-				return r;
-			}
+		r = amdgpu_bo_create_reserved(adev, dws * 4, PAGE_SIZE,
+					      AMDGPU_GEM_DOMAIN_VRAM,
+					      &adev->gfx.rlc.clear_state_obj,
+					      &adev->gfx.rlc.clear_state_gpu_addr,
+					      (void **)&adev->gfx.rlc.cs_ptr);
+		if (r) {
+			dev_err(adev->dev, "(%d) failed to create rlc csb bo\n",
+				r);
+			gfx_v9_0_rlc_fini(adev);
+			return r;
 		}
 		/* set up the cs buffer */
 		dst_ptr = adev->gfx.rlc.cs_ptr;
@@ -795,18 +795,16 @@ static int gfx_v9_0_rlc_init(struct amdgpu_device *adev)
 	if (adev->asic_type == CHIP_RAVEN) {
 		/* TODO: double check the cp_table_size for RV */
 		adev->gfx.rlc.cp_table_size = ALIGN(96 * 5 * 4, 2048) + (64 * 1024); /* JT + GDS */
-		if (adev->gfx.rlc.cp_table_obj == NULL) {
-			r = amdgpu_bo_create_kernel(adev, adev->gfx.rlc.cp_table_size,
-						PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM,
-						&adev->gfx.rlc.cp_table_obj,
-						&adev->gfx.rlc.cp_table_gpu_addr,
-						(void **)&adev->gfx.rlc.cp_table_ptr);
-			if (r) {
-				dev_err(adev->dev,
-					"(%d) failed to create cp table bo\n", r);
-				gfx_v9_0_rlc_fini(adev);
-				return r;
-			}
+		r = amdgpu_bo_create_reserved(adev, adev->gfx.rlc.cp_table_size,
+					      PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM,
+					      &adev->gfx.rlc.cp_table_obj,
+					      &adev->gfx.rlc.cp_table_gpu_addr,
+					      (void **)&adev->gfx.rlc.cp_table_ptr);
+		if (r) {
+			dev_err(adev->dev,
+				"(%d) failed to create cp table bo\n", r);
+			gfx_v9_0_rlc_fini(adev);
+			return r;
 		}
 
 		rv_init_cp_jump_table(adev);
@@ -821,28 +819,8 @@ static int gfx_v9_0_rlc_init(struct amdgpu_device *adev)
 
 static void gfx_v9_0_mec_fini(struct amdgpu_device *adev)
 {
-	int r;
-
-	if (adev->gfx.mec.hpd_eop_obj) {
-		r = amdgpu_bo_reserve(adev->gfx.mec.hpd_eop_obj, true);
-		if (unlikely(r != 0))
-			dev_warn(adev->dev, "(%d) reserve HPD EOP bo failed\n", r);
-		amdgpu_bo_unpin(adev->gfx.mec.hpd_eop_obj);
-		amdgpu_bo_unreserve(adev->gfx.mec.hpd_eop_obj);
-
-		amdgpu_bo_unref(&adev->gfx.mec.hpd_eop_obj);
-		adev->gfx.mec.hpd_eop_obj = NULL;
-	}
-	if (adev->gfx.mec.mec_fw_obj) {
-		r = amdgpu_bo_reserve(adev->gfx.mec.mec_fw_obj, true);
-		if (unlikely(r != 0))
-			dev_warn(adev->dev, "(%d) reserve mec firmware bo failed\n", r);
-		amdgpu_bo_unpin(adev->gfx.mec.mec_fw_obj);
-		amdgpu_bo_unreserve(adev->gfx.mec.mec_fw_obj);
-
-		amdgpu_bo_unref(&adev->gfx.mec.mec_fw_obj);
-		adev->gfx.mec.mec_fw_obj = NULL;
-	}
+	amdgpu_bo_free_kernel(&adev->gfx.mec.hpd_eop_obj, NULL, NULL);
+	amdgpu_bo_free_kernel(&adev->gfx.mec.mec_fw_obj, NULL, NULL);
 }
 
 static int gfx_v9_0_mec_init(struct amdgpu_device *adev)
@@ -862,33 +840,13 @@ static int gfx_v9_0_mec_init(struct amdgpu_device *adev)
 	amdgpu_gfx_compute_queue_acquire(adev);
 	mec_hpd_size = adev->gfx.num_compute_rings * GFX9_MEC_HPD_SIZE;
 
-	if (adev->gfx.mec.hpd_eop_obj == NULL) {
-		r = amdgpu_bo_create(adev,
-				     mec_hpd_size,
-				     PAGE_SIZE, true,
-				     AMDGPU_GEM_DOMAIN_GTT, 0, NULL, NULL,
-				     &adev->gfx.mec.hpd_eop_obj);
-		if (r) {
-			dev_warn(adev->dev, "(%d) create HDP EOP bo failed\n", r);
-			return r;
-		}
-	}
-
-	r = amdgpu_bo_reserve(adev->gfx.mec.hpd_eop_obj, false);
-	if (unlikely(r != 0)) {
-		gfx_v9_0_mec_fini(adev);
-		return r;
-	}
-	r = amdgpu_bo_pin(adev->gfx.mec.hpd_eop_obj, AMDGPU_GEM_DOMAIN_GTT,
-			  &adev->gfx.mec.hpd_eop_gpu_addr);
+	r = amdgpu_bo_create_reserved(adev, mec_hpd_size, PAGE_SIZE,
+				      AMDGPU_GEM_DOMAIN_GTT,
+				      &adev->gfx.mec.hpd_eop_obj,
+				      &adev->gfx.mec.hpd_eop_gpu_addr,
+				      (void **)&hpd);
 	if (r) {
-		dev_warn(adev->dev, "(%d) pin HDP EOP bo failed\n", r);
-		gfx_v9_0_mec_fini(adev);
-		return r;
-	}
-	r = amdgpu_bo_kmap(adev->gfx.mec.hpd_eop_obj, (void **)&hpd);
-	if (r) {
-		dev_warn(adev->dev, "(%d) map HDP EOP bo failed\n", r);
+		dev_warn(adev->dev, "(%d) create HDP EOP bo failed\n", r);
 		gfx_v9_0_mec_fini(adev);
 		return r;
 	}
@@ -905,42 +863,22 @@ static int gfx_v9_0_mec_init(struct amdgpu_device *adev)
 		 le32_to_cpu(mec_hdr->header.ucode_array_offset_bytes));
 	fw_size = le32_to_cpu(mec_hdr->header.ucode_size_bytes) / 4;
 
-	if (adev->gfx.mec.mec_fw_obj == NULL) {
-		r = amdgpu_bo_create(adev,
-			mec_hdr->header.ucode_size_bytes,
-			PAGE_SIZE, true,
-			AMDGPU_GEM_DOMAIN_GTT, 0, NULL, NULL,
-			&adev->gfx.mec.mec_fw_obj);
-		if (r) {
-			dev_warn(adev->dev, "(%d) create mec firmware bo failed\n", r);
-			return r;
-		}
-	}
-
-	r = amdgpu_bo_reserve(adev->gfx.mec.mec_fw_obj, false);
-	if (unlikely(r != 0)) {
-		gfx_v9_0_mec_fini(adev);
-		return r;
-	}
-	r = amdgpu_bo_pin(adev->gfx.mec.mec_fw_obj, AMDGPU_GEM_DOMAIN_GTT,
-			&adev->gfx.mec.mec_fw_gpu_addr);
-	if (r) {
-		dev_warn(adev->dev, "(%d) pin mec firmware bo failed\n", r);
-		gfx_v9_0_mec_fini(adev);
-		return r;
-	}
-	r = amdgpu_bo_kmap(adev->gfx.mec.mec_fw_obj, (void **)&fw);
+	r = amdgpu_bo_create_reserved(adev, mec_hdr->header.ucode_size_bytes,
+				      PAGE_SIZE, AMDGPU_GEM_DOMAIN_GTT,
+				      &adev->gfx.mec.mec_fw_obj,
+				      &adev->gfx.mec.mec_fw_gpu_addr,
+				      (void **)&fw);
 	if (r) {
-		dev_warn(adev->dev, "(%d) map firmware bo failed\n", r);
+		dev_warn(adev->dev, "(%d) create mec firmware bo failed\n", r);
 		gfx_v9_0_mec_fini(adev);
 		return r;
 	}
+
 	memcpy(fw, fw_data, fw_size);
 
 	amdgpu_bo_kunmap(adev->gfx.mec.mec_fw_obj);
 	amdgpu_bo_unreserve(adev->gfx.mec.mec_fw_obj);
 
-
 	return 0;
 }
 
@@ -2219,7 +2157,7 @@ static int gfx_v9_0_cp_gfx_start(struct amdgpu_device *adev)
 	struct amdgpu_ring *ring = &adev->gfx.gfx_ring[0];
 	const struct cs_section_def *sect = NULL;
 	const struct cs_extent_def *ext = NULL;
-	int r, i;
+	int r, i, tmp;
 
 	/* init the CP */
 	WREG32_SOC15(GC, 0, mmCP_MAX_CONTEXT, adev->gfx.config.max_hw_contexts - 1);
@@ -2227,7 +2165,7 @@ static int gfx_v9_0_cp_gfx_start(struct amdgpu_device *adev)
 
 	gfx_v9_0_cp_gfx_enable(adev, true);
 
-	r = amdgpu_ring_alloc(ring, gfx_v9_0_get_csb_size(adev) + 4);
+	r = amdgpu_ring_alloc(ring, gfx_v9_0_get_csb_size(adev) + 4 + 3);
 	if (r) {
 		DRM_ERROR("amdgpu: cp failed to lock ring (%d).\n", r);
 		return r;
@@ -2265,6 +2203,12 @@ static int gfx_v9_0_cp_gfx_start(struct amdgpu_device *adev)
 	amdgpu_ring_write(ring, 0x8000);
 	amdgpu_ring_write(ring, 0x8000);
 
+	amdgpu_ring_write(ring, PACKET3(PACKET3_SET_UCONFIG_REG,1));
+	tmp = (PACKET3_SET_UCONFIG_REG_INDEX_TYPE |
+		(SOC15_REG_OFFSET(GC, 0, mmVGT_INDEX_TYPE) - PACKET3_SET_UCONFIG_REG_START));
+	amdgpu_ring_write(ring, tmp);
+	amdgpu_ring_write(ring, 0);
+
 	amdgpu_ring_commit(ring);
 
 	return 0;
@@ -4158,7 +4102,7 @@ static int gfx_v9_0_kiq_irq(struct amdgpu_device *adev,
 	return 0;
 }
 
-const struct amd_ip_funcs gfx_v9_0_ip_funcs = {
+static const struct amd_ip_funcs gfx_v9_0_ip_funcs = {
 	.name = "gfx_v9_0",
 	.early_init = gfx_v9_0_early_init,
 	.late_init = gfx_v9_0_late_init,
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.h b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.h
index 56ef652a575d..fa5a3fbaf6ab 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.h
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.h
@@ -24,7 +24,6 @@
 #ifndef __GFX_V9_0_H__
 #define __GFX_V9_0_H__
 
-extern const struct amd_ip_funcs gfx_v9_0_ip_funcs;
 extern const struct amdgpu_ip_block_version gfx_v9_0_ip_block;
 
 void gfx_v9_0_select_se_sh(struct amdgpu_device *adev, u32 se_num, u32 sh_num);
diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c
index 408723ef157c..4f2788b61a08 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c
@@ -124,7 +124,7 @@ static void gfxhub_v1_0_init_tlb_regs(struct amdgpu_device *adev)
 
 static void gfxhub_v1_0_init_cache_regs(struct amdgpu_device *adev)
 {
-	uint32_t tmp;
+	uint32_t tmp, field;
 
 	/* Setup L2 cache */
 	tmp = RREG32_SOC15(GC, 0, mmVM_L2_CNTL);
@@ -143,9 +143,10 @@ static void gfxhub_v1_0_init_cache_regs(struct amdgpu_device *adev)
 	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL2, INVALIDATE_L2_CACHE, 1);
 	WREG32_SOC15(GC, 0, mmVM_L2_CNTL2, tmp);
 
+	field = adev->vm_manager.fragment_size;
 	tmp = mmVM_L2_CNTL3_DEFAULT;
-	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, BANK_SELECT, 12);
-	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, L2_CACHE_BIGK_FRAGMENT_SIZE, 9);
+	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, BANK_SELECT, field);
+	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, L2_CACHE_BIGK_FRAGMENT_SIZE, 6);
 	WREG32_SOC15(GC, 0, mmVM_L2_CNTL3, tmp);
 
 	tmp = mmVM_L2_CNTL4_DEFAULT;
diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.h b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.h
index d2dbb085f480..206e29cad753 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.h
+++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.h
@@ -30,7 +30,5 @@ void gfxhub_v1_0_set_fault_enable_default(struct amdgpu_device *adev,
 					  bool value);
 void gfxhub_v1_0_init(struct amdgpu_device *adev);
 u64 gfxhub_v1_0_get_mc_fb_offset(struct amdgpu_device *adev);
-extern const struct amd_ip_funcs gfxhub_v1_0_ip_funcs;
-extern const struct amdgpu_ip_block_version gfxhub_v1_0_ip_block;
 
 #endif
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c
index 93c45f26b7c8..12b0c4cd7a5a 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c
@@ -461,6 +461,7 @@ static void gmc_v6_0_set_prt(struct amdgpu_device *adev, bool enable)
 static int gmc_v6_0_gart_enable(struct amdgpu_device *adev)
 {
 	int r, i;
+	u32 field;
 
 	if (adev->gart.robj == NULL) {
 		dev_err(adev->dev, "No VRAM object for PCIE GART.\n");
@@ -488,10 +489,12 @@ static int gmc_v6_0_gart_enable(struct amdgpu_device *adev)
 	WREG32(mmVM_L2_CNTL2,
 	       VM_L2_CNTL2__INVALIDATE_ALL_L1_TLBS_MASK |
 	       VM_L2_CNTL2__INVALIDATE_L2_CACHE_MASK);
+
+	field = adev->vm_manager.fragment_size;
 	WREG32(mmVM_L2_CNTL3,
 	       VM_L2_CNTL3__L2_CACHE_BIGK_ASSOCIATIVITY_MASK |
-	       (4UL << VM_L2_CNTL3__BANK_SELECT__SHIFT) |
-	       (4UL << VM_L2_CNTL3__L2_CACHE_BIGK_FRAGMENT_SIZE__SHIFT));
+	       (field << VM_L2_CNTL3__BANK_SELECT__SHIFT) |
+	       (field << VM_L2_CNTL3__L2_CACHE_BIGK_FRAGMENT_SIZE__SHIFT));
 	/* setup context0 */
 	WREG32(mmVM_CONTEXT0_PAGE_TABLE_START_ADDR, adev->mc.gart_start >> 12);
 	WREG32(mmVM_CONTEXT0_PAGE_TABLE_END_ADDR, adev->mc.gart_end >> 12);
@@ -811,7 +814,7 @@ static int gmc_v6_0_sw_init(void *handle)
 	if (r)
 		return r;
 
-	amdgpu_vm_adjust_size(adev, 64);
+	amdgpu_vm_adjust_size(adev, 64, 4);
 	adev->vm_manager.max_pfn = adev->vm_manager.vm_size << 18;
 
 	adev->mc.mc_mask = 0xffffffffffULL;
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c
index 4a9e84062874..e42c1ad3af5e 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c
@@ -562,7 +562,7 @@ static void gmc_v7_0_set_prt(struct amdgpu_device *adev, bool enable)
 static int gmc_v7_0_gart_enable(struct amdgpu_device *adev)
 {
 	int r, i;
-	u32 tmp;
+	u32 tmp, field;
 
 	if (adev->gart.robj == NULL) {
 		dev_err(adev->dev, "No VRAM object for PCIE GART.\n");
@@ -592,10 +592,12 @@ static int gmc_v7_0_gart_enable(struct amdgpu_device *adev)
 	tmp = REG_SET_FIELD(0, VM_L2_CNTL2, INVALIDATE_ALL_L1_TLBS, 1);
 	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL2, INVALIDATE_L2_CACHE, 1);
 	WREG32(mmVM_L2_CNTL2, tmp);
+
+	field = adev->vm_manager.fragment_size;
 	tmp = RREG32(mmVM_L2_CNTL3);
 	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, L2_CACHE_BIGK_ASSOCIATIVITY, 1);
-	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, BANK_SELECT, 4);
-	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, L2_CACHE_BIGK_FRAGMENT_SIZE, 4);
+	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, BANK_SELECT, field);
+	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, L2_CACHE_BIGK_FRAGMENT_SIZE, field);
 	WREG32(mmVM_L2_CNTL3, tmp);
 	/* setup context0 */
 	WREG32(mmVM_CONTEXT0_PAGE_TABLE_START_ADDR, adev->mc.gart_start >> 12);
@@ -948,7 +950,7 @@ static int gmc_v7_0_sw_init(void *handle)
 	 * Currently set to 4GB ((1 << 20) 4k pages).
 	 * Max GPUVM size for cayman and SI is 40 bits.
 	 */
-	amdgpu_vm_adjust_size(adev, 64);
+	amdgpu_vm_adjust_size(adev, 64, 4);
 	adev->vm_manager.max_pfn = adev->vm_manager.vm_size << 18;
 
 	/* Set the internal MC address mask
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c
index 85c937b5e40b..7ca2dae8237a 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c
@@ -762,7 +762,7 @@ static void gmc_v8_0_set_prt(struct amdgpu_device *adev, bool enable)
 static int gmc_v8_0_gart_enable(struct amdgpu_device *adev)
 {
 	int r, i;
-	u32 tmp;
+	u32 tmp, field;
 
 	if (adev->gart.robj == NULL) {
 		dev_err(adev->dev, "No VRAM object for PCIE GART.\n");
@@ -793,10 +793,12 @@ static int gmc_v8_0_gart_enable(struct amdgpu_device *adev)
 	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL2, INVALIDATE_ALL_L1_TLBS, 1);
 	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL2, INVALIDATE_L2_CACHE, 1);
 	WREG32(mmVM_L2_CNTL2, tmp);
+
+	field = adev->vm_manager.fragment_size;
 	tmp = RREG32(mmVM_L2_CNTL3);
 	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, L2_CACHE_BIGK_ASSOCIATIVITY, 1);
-	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, BANK_SELECT, 4);
-	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, L2_CACHE_BIGK_FRAGMENT_SIZE, 4);
+	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, BANK_SELECT, field);
+	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, L2_CACHE_BIGK_FRAGMENT_SIZE, field);
 	WREG32(mmVM_L2_CNTL3, tmp);
 	/* XXX: set to enable PTE/PDE in system memory */
 	tmp = RREG32(mmVM_L2_CNTL4);
@@ -1046,7 +1048,7 @@ static int gmc_v8_0_sw_init(void *handle)
 	 * Currently set to 4GB ((1 << 20) 4k pages).
 	 * Max GPUVM size for cayman and SI is 40 bits.
 	 */
-	amdgpu_vm_adjust_size(adev, 64);
+	amdgpu_vm_adjust_size(adev, 64, 4);
 	adev->vm_manager.max_pfn = adev->vm_manager.vm_size << 18;
 
 	/* Set the internal MC address mask
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index c22899a08106..2769c2b3b56e 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -541,9 +541,10 @@ static int gmc_v9_0_sw_init(void *handle)
 			adev->vm_manager.vm_size = 1U << 18;
 			adev->vm_manager.block_size = 9;
 			adev->vm_manager.num_level = 3;
+			amdgpu_vm_set_fragment_size(adev, 9);
 		} else {
-			/* vm_size is 64GB for legacy 2-level page support*/
-			amdgpu_vm_adjust_size(adev, 64);
+			/* vm_size is 64GB for legacy 2-level page support */
+			amdgpu_vm_adjust_size(adev, 64, 9);
 			adev->vm_manager.num_level = 1;
 		}
 		break;
@@ -558,14 +559,16 @@ static int gmc_v9_0_sw_init(void *handle)
 		adev->vm_manager.vm_size = 1U << 18;
 		adev->vm_manager.block_size = 9;
 		adev->vm_manager.num_level = 3;
+		amdgpu_vm_set_fragment_size(adev, 9);
 		break;
 	default:
 		break;
 	}
 
-	DRM_INFO("vm size is %llu GB, block size is %u-bit\n",
+	DRM_INFO("vm size is %llu GB, block size is %u-bit,fragment size is %u-bit\n",
 			adev->vm_manager.vm_size,
-			adev->vm_manager.block_size);
+			adev->vm_manager.block_size,
+			adev->vm_manager.fragment_size);
 
 	/* This interrupt is VMC page fault.*/
 	r = amdgpu_irq_add_id(adev, AMDGPU_IH_CLIENTID_VMC, 0,
diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c
index ad8def3cc343..4395a4f12149 100644
--- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c
@@ -138,7 +138,7 @@ static void mmhub_v1_0_init_tlb_regs(struct amdgpu_device *adev)
 
 static void mmhub_v1_0_init_cache_regs(struct amdgpu_device *adev)
 {
-	uint32_t tmp;
+	uint32_t tmp, field;
 
 	/* Setup L2 cache */
 	tmp = RREG32_SOC15(MMHUB, 0, mmVM_L2_CNTL);
@@ -157,9 +157,10 @@ static void mmhub_v1_0_init_cache_regs(struct amdgpu_device *adev)
 	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL2, INVALIDATE_L2_CACHE, 1);
 	WREG32_SOC15(MMHUB, 0, mmVM_L2_CNTL2, tmp);
 
+	field = adev->vm_manager.fragment_size;
 	tmp = mmVM_L2_CNTL3_DEFAULT;
-	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, BANK_SELECT, 12);
-	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, L2_CACHE_BIGK_FRAGMENT_SIZE, 9);
+	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, BANK_SELECT, field);
+	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, L2_CACHE_BIGK_FRAGMENT_SIZE, 6);
 	WREG32_SOC15(MMHUB, 0, mmVM_L2_CNTL3, tmp);
 
 	tmp = mmVM_L2_CNTL4_DEFAULT;
diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.h b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.h
index 57bb940c0ecd..5d38229baf69 100644
--- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.h
+++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.h
@@ -36,7 +36,4 @@ void mmhub_v1_0_initialize_power_gating(struct amdgpu_device *adev);
 void mmhub_v1_0_update_power_gating(struct amdgpu_device *adev,
                                 bool enable);
 
-extern const struct amd_ip_funcs mmhub_v1_0_ip_funcs;
-extern const struct amdgpu_ip_block_version mmhub_v1_0_ip_block;
-
 #endif
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c
index 591f3e7fb508..fd7c72aaafa6 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c
@@ -291,6 +291,8 @@ static void sdma_v4_0_ring_set_wptr(struct amdgpu_ring *ring)
 
 	DRM_DEBUG("Setting write pointer\n");
 	if (ring->use_doorbell) {
+		u64 *wb = (u64 *)&adev->wb.wb[ring->wptr_offs];
+
 		DRM_DEBUG("Using doorbell -- "
 				"wptr_offs == 0x%08x "
 				"lower_32_bits(ring->wptr) << 2 == 0x%08x "
@@ -299,8 +301,7 @@ static void sdma_v4_0_ring_set_wptr(struct amdgpu_ring *ring)
 				lower_32_bits(ring->wptr << 2),
 				upper_32_bits(ring->wptr << 2));
 		/* XXX check if swapping is necessary on BE */
-		adev->wb.wb[ring->wptr_offs] = lower_32_bits(ring->wptr << 2);
-		adev->wb.wb[ring->wptr_offs + 1] = upper_32_bits(ring->wptr << 2);
+		WRITE_ONCE(*wb, (ring->wptr << 2));
 		DRM_DEBUG("calling WDOORBELL64(0x%08x, 0x%016llx)\n",
 				ring->doorbell_index, ring->wptr << 2);
 		WDOORBELL64(ring->doorbell_index, ring->wptr << 2);
@@ -573,12 +574,13 @@ static void sdma_v4_0_enable(struct amdgpu_device *adev, bool enable)
 static int sdma_v4_0_gfx_resume(struct amdgpu_device *adev)
 {
 	struct amdgpu_ring *ring;
-	u32 rb_cntl, ib_cntl;
+	u32 rb_cntl, ib_cntl, wptr_poll_cntl;
 	u32 rb_bufsz;
 	u32 wb_offset;
 	u32 doorbell;
 	u32 doorbell_offset;
 	u32 temp;
+	u64 wptr_gpu_addr;
 	int i, r;
 
 	for (i = 0; i < adev->sdma.num_instances; i++) {
@@ -660,6 +662,19 @@ static int sdma_v4_0_gfx_resume(struct amdgpu_device *adev)
 			WREG32(sdma_v4_0_get_reg_offset(i, mmSDMA0_F32_CNTL), temp);
 		}
 
+		/* setup the wptr shadow polling */
+		wptr_gpu_addr = adev->wb.gpu_addr + (ring->wptr_offs * 4);
+		WREG32(sdma_v4_0_get_reg_offset(i, mmSDMA0_GFX_RB_WPTR_POLL_ADDR_LO),
+		       lower_32_bits(wptr_gpu_addr));
+		WREG32(sdma_v4_0_get_reg_offset(i, mmSDMA0_GFX_RB_WPTR_POLL_ADDR_HI),
+		       upper_32_bits(wptr_gpu_addr));
+		wptr_poll_cntl = RREG32(sdma_v4_0_get_reg_offset(i, mmSDMA0_GFX_RB_WPTR_POLL_CNTL));
+		if (amdgpu_sriov_vf(adev))
+			wptr_poll_cntl = REG_SET_FIELD(wptr_poll_cntl, SDMA0_GFX_RB_WPTR_POLL_CNTL, F32_POLL_ENABLE, 1);
+		else
+			wptr_poll_cntl = REG_SET_FIELD(wptr_poll_cntl, SDMA0_GFX_RB_WPTR_POLL_CNTL, F32_POLL_ENABLE, 0);
+		WREG32(sdma_v4_0_get_reg_offset(i, mmSDMA0_GFX_RB_WPTR_POLL_CNTL), wptr_poll_cntl);
+
 		/* enable DMA RB */
 		rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, RB_ENABLE, 1);
 		WREG32(sdma_v4_0_get_reg_offset(i, mmSDMA0_GFX_RB_CNTL), rb_cntl);
@@ -687,6 +702,7 @@ static int sdma_v4_0_gfx_resume(struct amdgpu_device *adev)
 
 		if (adev->mman.buffer_funcs_ring == ring)
 			amdgpu_ttm_set_active_vram_size(adev, adev->mc.real_vram_size);
+
 	}
 
 	return 0;
@@ -783,15 +799,12 @@ static int sdma_v4_0_load_microcode(struct amdgpu_device *adev)
 	const struct sdma_firmware_header_v1_0 *hdr;
 	const __le32 *fw_data;
 	u32 fw_size;
-	u32 digest_size = 0;
 	int i, j;
 
 	/* halt the MEs */
 	sdma_v4_0_enable(adev, false);
 
 	for (i = 0; i < adev->sdma.num_instances; i++) {
-		uint16_t version_major;
-		uint16_t version_minor;
 		if (!adev->sdma.instance[i].fw)
 			return -EINVAL;
 
@@ -799,23 +812,12 @@ static int sdma_v4_0_load_microcode(struct amdgpu_device *adev)
 		amdgpu_ucode_print_sdma_hdr(&hdr->header);
 		fw_size = le32_to_cpu(hdr->header.ucode_size_bytes) / 4;
 
-		version_major = le16_to_cpu(hdr->header.header_version_major);
-		version_minor = le16_to_cpu(hdr->header.header_version_minor);
-
-		if (version_major == 1 && version_minor >= 1) {
-			const struct sdma_firmware_header_v1_1 *sdma_v1_1_hdr = (const struct sdma_firmware_header_v1_1 *) hdr;
-			digest_size = le32_to_cpu(sdma_v1_1_hdr->digest_size);
-		}
-
-		fw_size -= digest_size;
-
 		fw_data = (const __le32 *)
 			(adev->sdma.instance[i].fw->data +
 				le32_to_cpu(hdr->header.ucode_array_offset_bytes));
 
 		WREG32(sdma_v4_0_get_reg_offset(i, mmSDMA0_UCODE_ADDR), 0);
 
-
 		for (j = 0; j < fw_size; j++)
 			WREG32(sdma_v4_0_get_reg_offset(i, mmSDMA0_UCODE_DATA), le32_to_cpup(fw_data++));
 
diff --git a/drivers/gpu/drm/amd/amdgpu/si.c b/drivers/gpu/drm/amd/amdgpu/si.c
index 812a24dd1204..8284d5dbfc30 100644
--- a/drivers/gpu/drm/amd/amdgpu/si.c
+++ b/drivers/gpu/drm/amd/amdgpu/si.c
@@ -1413,6 +1413,7 @@ static void si_init_golden_registers(struct amdgpu_device *adev)
 		amdgpu_program_register_sequence(adev,
 						 pitcairn_mgcg_cgcg_init,
 						 (const u32)ARRAY_SIZE(pitcairn_mgcg_cgcg_init));
+		break;
 	case CHIP_VERDE:
 		amdgpu_program_register_sequence(adev,
 						 verde_golden_registers,
@@ -1437,6 +1438,7 @@ static void si_init_golden_registers(struct amdgpu_device *adev)
 		amdgpu_program_register_sequence(adev,
 						 oland_mgcg_cgcg_init,
 						 (const u32)ARRAY_SIZE(oland_mgcg_cgcg_init));
+		break;
 	case CHIP_HAINAN:
 		amdgpu_program_register_sequence(adev,
 						 hainan_golden_registers,
diff --git a/drivers/gpu/drm/amd/amdgpu/soc15d.h b/drivers/gpu/drm/amd/amdgpu/soc15d.h
index e79befd80eed..7f408f85fdb6 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc15d.h
+++ b/drivers/gpu/drm/amd/amdgpu/soc15d.h
@@ -250,6 +250,7 @@
 #define	PACKET3_SET_UCONFIG_REG				0x79
 #define		PACKET3_SET_UCONFIG_REG_START			0x0000c000
 #define		PACKET3_SET_UCONFIG_REG_END			0x0000c400
+#define		PACKET3_SET_UCONFIG_REG_INDEX_TYPE		(2 << 28)
 #define	PACKET3_SCRATCH_RAM_WRITE			0x7D
 #define	PACKET3_SCRATCH_RAM_READ			0x7E
 #define	PACKET3_LOAD_CONST_RAM				0x80
diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c b/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c
index 987b958368ac..23a85750edd6 100644
--- a/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c
@@ -165,6 +165,9 @@ static int uvd_v7_0_enc_ring_test_ring(struct amdgpu_ring *ring)
 	unsigned i;
 	int r;
 
+	if (amdgpu_sriov_vf(adev))
+		return 0;
+
 	r = amdgpu_ring_alloc(ring, 16);
 	if (r) {
 		DRM_ERROR("amdgpu: uvd enc failed to lock ring %d (%d).\n",
@@ -432,13 +435,19 @@ static int uvd_v7_0_sw_init(void *handle)
 			return r;
 	}
 
-
 	for (i = 0; i < adev->uvd.num_enc_rings; ++i) {
 		ring = &adev->uvd.ring_enc[i];
 		sprintf(ring->name, "uvd_enc%d", i);
 		if (amdgpu_sriov_vf(adev)) {
 			ring->use_doorbell = true;
-			ring->doorbell_index = AMDGPU_DOORBELL64_UVD_RING0_1 * 2;
+
+			/* currently only use the first enconding ring for
+			 * sriov, so set unused location for other unused rings.
+			 */
+			if (i == 0)
+				ring->doorbell_index = AMDGPU_DOORBELL64_UVD_RING0_1 * 2;
+			else
+				ring->doorbell_index = AMDGPU_DOORBELL64_UVD_RING2_3 * 2 + 1;
 		}
 		r = amdgpu_ring_init(adev, ring, 512, &adev->uvd.irq, 0);
 		if (r)
@@ -685,6 +694,11 @@ static int uvd_v7_0_mmsch_start(struct amdgpu_device *adev,
 	/* 4, set resp to zero */
 	WREG32_SOC15(VCE, 0, mmVCE_MMSCH_VF_MAILBOX_RESP, 0);
 
+	WDOORBELL32(adev->uvd.ring_enc[0].doorbell_index, 0);
+	adev->wb.wb[adev->uvd.ring_enc[0].wptr_offs] = 0;
+	adev->uvd.ring_enc[0].wptr = 0;
+	adev->uvd.ring_enc[0].wptr_old = 0;
+
 	/* 5, kick off the initialization and wait until VCE_MMSCH_VF_MAILBOX_RESP becomes non-zero */
 	WREG32_SOC15(VCE, 0, mmVCE_MMSCH_VF_MAILBOX_HOST, 0x10000001);
 
@@ -702,7 +716,6 @@ static int uvd_v7_0_mmsch_start(struct amdgpu_device *adev,
 		dev_err(adev->dev, "failed to init MMSCH, mmVCE_MMSCH_VF_MAILBOX_RESP = %x\n", data);
 		return -EBUSY;
 	}
-	WDOORBELL32(adev->uvd.ring_enc[0].doorbell_index, 0);
 
 	return 0;
 }
@@ -736,11 +749,9 @@ static int uvd_v7_0_sriov_start(struct amdgpu_device *adev)
 		init_table += header->uvd_table_offset;
 
 		ring = &adev->uvd.ring;
+		ring->wptr = 0;
 		size = AMDGPU_GPU_PAGE_ALIGN(adev->uvd.fw->size + 4);
 
-		/* disable clock gating */
-		MMSCH_V1_0_INSERT_DIRECT_RD_MOD_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_POWER_STATUS),
-						   ~UVD_POWER_STATUS__UVD_PG_MODE_MASK, 0);
 		MMSCH_V1_0_INSERT_DIRECT_RD_MOD_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_STATUS),
 						   0xFFFFFFFF, 0x00000004);
 		/* mc resume*/
@@ -777,12 +788,6 @@ static int uvd_v7_0_sriov_start(struct amdgpu_device *adev)
 		MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_VCPU_CACHE_SIZE2),
 					    AMDGPU_UVD_STACK_SIZE + (AMDGPU_UVD_SESSION_SIZE * 40));
 
-		MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_UDEC_ADDR_CONFIG),
-					    adev->gfx.config.gb_addr_config);
-		MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_UDEC_DB_ADDR_CONFIG),
-					    adev->gfx.config.gb_addr_config);
-		MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_UDEC_DBW_ADDR_CONFIG),
-					    adev->gfx.config.gb_addr_config);
 		MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_GP_SCRATCH4), adev->uvd.max_handles);
 		/* mc resume end*/
 
@@ -819,17 +824,6 @@ static int uvd_v7_0_sriov_start(struct amdgpu_device *adev)
 						       UVD_LMI_CTRL__REQ_MODE_MASK |
 						       0x00100000L));
 
-		/* disable byte swapping */
-		MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_LMI_SWAP_CNTL), 0);
-		MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_MP_SWAP_CNTL), 0);
-
-		MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_MPC_SET_MUXA0), 0x40c2040);
-		MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_MPC_SET_MUXA1), 0x0);
-		MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_MPC_SET_MUXB0), 0x40c2040);
-		MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_MPC_SET_MUXB1), 0x0);
-		MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_MPC_SET_ALU), 0);
-		MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_MPC_SET_MUX), 0x88);
-
 		/* take all subblocks out of reset, except VCPU */
 		MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_SOFT_RESET),
 					    UVD_SOFT_RESET__VCPU_SOFT_RESET_MASK);
@@ -838,15 +832,6 @@ static int uvd_v7_0_sriov_start(struct amdgpu_device *adev)
 		MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_VCPU_CNTL),
 					    UVD_VCPU_CNTL__CLK_EN_MASK);
 
-		/* enable UMC */
-		MMSCH_V1_0_INSERT_DIRECT_RD_MOD_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_LMI_CTRL2),
-						   ~UVD_LMI_CTRL2__STALL_ARB_UMC_MASK, 0);
-
-		/* boot up the VCPU */
-		MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_SOFT_RESET), 0);
-
-		MMSCH_V1_0_INSERT_DIRECT_POLL(SOC15_REG_OFFSET(UVD, 0, mmUVD_STATUS), 0x02, 0x02);
-
 		/* enable master interrupt */
 		MMSCH_V1_0_INSERT_DIRECT_RD_MOD_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_MASTINT_EN),
 						   ~(UVD_MASTINT_EN__VCPU_EN_MASK|UVD_MASTINT_EN__SYS_EN_MASK),
@@ -859,40 +844,31 @@ static int uvd_v7_0_sriov_start(struct amdgpu_device *adev)
 		/* force RBC into idle state */
 		size = order_base_2(ring->ring_size);
 		tmp = REG_SET_FIELD(0, UVD_RBC_RB_CNTL, RB_BUFSZ, size);
-		tmp = REG_SET_FIELD(tmp, UVD_RBC_RB_CNTL, RB_BLKSZ, 1);
 		tmp = REG_SET_FIELD(tmp, UVD_RBC_RB_CNTL, RB_NO_FETCH, 1);
-		tmp = REG_SET_FIELD(tmp, UVD_RBC_RB_CNTL, RB_WPTR_POLL_EN, 0);
-		tmp = REG_SET_FIELD(tmp, UVD_RBC_RB_CNTL, RB_NO_UPDATE, 1);
-		tmp = REG_SET_FIELD(tmp, UVD_RBC_RB_CNTL, RB_RPTR_WR_EN, 1);
 		MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_RBC_RB_CNTL), tmp);
 
-		/* set the write pointer delay */
-		MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_RBC_RB_WPTR_CNTL), 0);
-
-		/* set the wb address */
-		MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_RBC_RB_RPTR_ADDR),
-					    (upper_32_bits(ring->gpu_addr) >> 2));
-
-		/* programm the RB_BASE for ring buffer */
-		MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_LMI_RBC_RB_64BIT_BAR_LOW),
-					    lower_32_bits(ring->gpu_addr));
-		MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_LMI_RBC_RB_64BIT_BAR_HIGH),
-					    upper_32_bits(ring->gpu_addr));
-
-		ring->wptr = 0;
 		ring = &adev->uvd.ring_enc[0];
+		ring->wptr = 0;
 		MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_RB_BASE_LO), ring->gpu_addr);
 		MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_RB_BASE_HI), upper_32_bits(ring->gpu_addr));
 		MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_RB_SIZE), ring->ring_size / 4);
 
+		/* boot up the VCPU */
+		MMSCH_V1_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_SOFT_RESET), 0);
+
+		/* enable UMC */
+		MMSCH_V1_0_INSERT_DIRECT_RD_MOD_WT(SOC15_REG_OFFSET(UVD, 0, mmUVD_LMI_CTRL2),
+										   ~UVD_LMI_CTRL2__STALL_ARB_UMC_MASK, 0);
+
+		MMSCH_V1_0_INSERT_DIRECT_POLL(SOC15_REG_OFFSET(UVD, 0, mmUVD_STATUS), 0x02, 0x02);
+
 		/* add end packet */
 		memcpy((void *)init_table, &end, sizeof(struct mmsch_v1_0_cmd_end));
 		table_size += sizeof(struct mmsch_v1_0_cmd_end) / 4;
 		header->uvd_table_size = table_size;
 
-		return uvd_v7_0_mmsch_start(adev, &adev->virt.mm_table);
 	}
-	return -EINVAL; /* already initializaed ? */
+	return uvd_v7_0_mmsch_start(adev, &adev->virt.mm_table);
 }
 
 /**
diff --git a/drivers/gpu/drm/amd/amdgpu/vce_v4_0.c b/drivers/gpu/drm/amd/amdgpu/vce_v4_0.c
index 1ecd6bb90c1f..11134d5f7443 100644
--- a/drivers/gpu/drm/amd/amdgpu/vce_v4_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vce_v4_0.c
@@ -173,6 +173,11 @@ static int vce_v4_0_mmsch_start(struct amdgpu_device *adev,
 	/* 4, set resp to zero */
 	WREG32(SOC15_REG_OFFSET(VCE, 0, mmVCE_MMSCH_VF_MAILBOX_RESP), 0);
 
+	WDOORBELL32(adev->vce.ring[0].doorbell_index, 0);
+	adev->wb.wb[adev->vce.ring[0].wptr_offs] = 0;
+	adev->vce.ring[0].wptr = 0;
+	adev->vce.ring[0].wptr_old = 0;
+
 	/* 5, kick off the initialization and wait until VCE_MMSCH_VF_MAILBOX_RESP becomes non-zero */
 	WREG32(SOC15_REG_OFFSET(VCE, 0, mmVCE_MMSCH_VF_MAILBOX_HOST), 0x10000001);
 
@@ -190,7 +195,6 @@ static int vce_v4_0_mmsch_start(struct amdgpu_device *adev,
 		dev_err(adev->dev, "failed to init MMSCH, mmVCE_MMSCH_VF_MAILBOX_RESP = %x\n", data);
 		return -EBUSY;
 	}
-	WDOORBELL32(adev->vce.ring[0].doorbell_index, 0);
 
 	return 0;
 }
@@ -274,7 +278,8 @@ static int vce_v4_0_sriov_start(struct amdgpu_device *adev)
 
 		MMSCH_V1_0_INSERT_DIRECT_RD_MOD_WT(SOC15_REG_OFFSET(VCE, 0, mmVCE_LMI_CTRL2), ~0x100, 0);
 		MMSCH_V1_0_INSERT_DIRECT_RD_MOD_WT(SOC15_REG_OFFSET(VCE, 0, mmVCE_SYS_INT_EN),
-						   0xffffffff, VCE_SYS_INT_EN__VCE_SYS_INT_TRAP_INTERRUPT_EN_MASK);
+						   VCE_SYS_INT_EN__VCE_SYS_INT_TRAP_INTERRUPT_EN_MASK,
+						   VCE_SYS_INT_EN__VCE_SYS_INT_TRAP_INTERRUPT_EN_MASK);
 
 		/* end of MC_RESUME */
 		MMSCH_V1_0_INSERT_DIRECT_RD_MOD_WT(SOC15_REG_OFFSET(VCE, 0, mmVCE_STATUS),
@@ -296,11 +301,9 @@ static int vce_v4_0_sriov_start(struct amdgpu_device *adev)
 		memcpy((void *)init_table, &end, sizeof(struct mmsch_v1_0_cmd_end));
 		table_size += sizeof(struct mmsch_v1_0_cmd_end) / 4;
 		header->vce_table_size = table_size;
-
-		return vce_v4_0_mmsch_start(adev, &adev->virt.mm_table);
 	}
 
-	return -EINVAL; /* already initializaed ? */
+	return vce_v4_0_mmsch_start(adev, &adev->virt.mm_table);
 }
 
 /**
@@ -443,12 +446,14 @@ static int vce_v4_0_sw_init(void *handle)
 		if (amdgpu_sriov_vf(adev)) {
 			/* DOORBELL only works under SRIOV */
 			ring->use_doorbell = true;
+
+			/* currently only use the first encoding ring for sriov,
+			 * so set unused location for other unused rings.
+			 */
 			if (i == 0)
-				ring->doorbell_index = AMDGPU_DOORBELL64_RING0_1 * 2;
-			else if (i == 1)
-				ring->doorbell_index = AMDGPU_DOORBELL64_RING2_3 * 2;
+				ring->doorbell_index = AMDGPU_DOORBELL64_VCE_RING0_1 * 2;
 			else
-				ring->doorbell_index = AMDGPU_DOORBELL64_RING2_3 * 2 + 1;
+				ring->doorbell_index = AMDGPU_DOORBELL64_VCE_RING2_3 * 2 + 1;
 		}
 		r = amdgpu_ring_init(adev, ring, 512, &adev->vce.irq, 0);
 		if (r)
@@ -990,11 +995,13 @@ static int vce_v4_0_set_interrupt_state(struct amdgpu_device *adev,
 {
 	uint32_t val = 0;
 
-	if (state == AMDGPU_IRQ_STATE_ENABLE)
-		val |= VCE_SYS_INT_EN__VCE_SYS_INT_TRAP_INTERRUPT_EN_MASK;
+	if (!amdgpu_sriov_vf(adev)) {
+		if (state == AMDGPU_IRQ_STATE_ENABLE)
+			val |= VCE_SYS_INT_EN__VCE_SYS_INT_TRAP_INTERRUPT_EN_MASK;
 
-	WREG32_P(SOC15_REG_OFFSET(VCE, 0, mmVCE_SYS_INT_EN), val,
-			~VCE_SYS_INT_EN__VCE_SYS_INT_TRAP_INTERRUPT_EN_MASK);
+		WREG32_P(SOC15_REG_OFFSET(VCE, 0, mmVCE_SYS_INT_EN), val,
+				~VCE_SYS_INT_EN__VCE_SYS_INT_TRAP_INTERRUPT_EN_MASK);
+	}
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/vi.c b/drivers/gpu/drm/amd/amdgpu/vi.c
index 6cac291c96da..9ff69b90df36 100644
--- a/drivers/gpu/drm/amd/amdgpu/vi.c
+++ b/drivers/gpu/drm/amd/amdgpu/vi.c
@@ -1028,8 +1028,7 @@ static int vi_common_early_init(void *handle)
 		/* rev0 hardware requires workarounds to support PG */
 		adev->pg_flags = 0;
 		if (adev->rev_id != 0x00 || CZ_REV_BRISTOL(adev->pdev->revision)) {
-			adev->pg_flags |= AMD_PG_SUPPORT_GFX_PG |
-				AMD_PG_SUPPORT_GFX_SMG |
+			adev->pg_flags |= AMD_PG_SUPPORT_GFX_SMG |
 				AMD_PG_SUPPORT_GFX_PIPELINE |
 				AMD_PG_SUPPORT_CP |
 				AMD_PG_SUPPORT_UVD |
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
index 6316aad43a73..e4a8c2e52cb2 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
@@ -142,12 +142,12 @@ static int set_queue_properties_from_user(struct queue_properties *q_properties,
 				struct kfd_ioctl_create_queue_args *args)
 {
 	if (args->queue_percentage > KFD_MAX_QUEUE_PERCENTAGE) {
-		pr_err("kfd: queue percentage must be between 0 to KFD_MAX_QUEUE_PERCENTAGE\n");
+		pr_err("Queue percentage must be between 0 to KFD_MAX_QUEUE_PERCENTAGE\n");
 		return -EINVAL;
 	}
 
 	if (args->queue_priority > KFD_MAX_QUEUE_PRIORITY) {
-		pr_err("kfd: queue priority must be between 0 to KFD_MAX_QUEUE_PRIORITY\n");
+		pr_err("Queue priority must be between 0 to KFD_MAX_QUEUE_PRIORITY\n");
 		return -EINVAL;
 	}
 
@@ -155,26 +155,26 @@ static int set_queue_properties_from_user(struct queue_properties *q_properties,
 		(!access_ok(VERIFY_WRITE,
 			(const void __user *) args->ring_base_address,
 			sizeof(uint64_t)))) {
-		pr_err("kfd: can't access ring base address\n");
+		pr_err("Can't access ring base address\n");
 		return -EFAULT;
 	}
 
 	if (!is_power_of_2(args->ring_size) && (args->ring_size != 0)) {
-		pr_err("kfd: ring size must be a power of 2 or 0\n");
+		pr_err("Ring size must be a power of 2 or 0\n");
 		return -EINVAL;
 	}
 
 	if (!access_ok(VERIFY_WRITE,
 			(const void __user *) args->read_pointer_address,
 			sizeof(uint32_t))) {
-		pr_err("kfd: can't access read pointer\n");
+		pr_err("Can't access read pointer\n");
 		return -EFAULT;
 	}
 
 	if (!access_ok(VERIFY_WRITE,
 			(const void __user *) args->write_pointer_address,
 			sizeof(uint32_t))) {
-		pr_err("kfd: can't access write pointer\n");
+		pr_err("Can't access write pointer\n");
 		return -EFAULT;
 	}
 
@@ -182,7 +182,7 @@ static int set_queue_properties_from_user(struct queue_properties *q_properties,
 		!access_ok(VERIFY_WRITE,
 			(const void __user *) args->eop_buffer_address,
 			sizeof(uint32_t))) {
-		pr_debug("kfd: can't access eop buffer");
+		pr_debug("Can't access eop buffer");
 		return -EFAULT;
 	}
 
@@ -190,7 +190,7 @@ static int set_queue_properties_from_user(struct queue_properties *q_properties,
 		!access_ok(VERIFY_WRITE,
 			(const void __user *) args->ctx_save_restore_address,
 			sizeof(uint32_t))) {
-		pr_debug("kfd: can't access ctx save restore buffer");
+		pr_debug("Can't access ctx save restore buffer");
 		return -EFAULT;
 	}
 
@@ -219,27 +219,27 @@ static int set_queue_properties_from_user(struct queue_properties *q_properties,
 	else
 		q_properties->format = KFD_QUEUE_FORMAT_PM4;
 
-	pr_debug("Queue Percentage (%d, %d)\n",
+	pr_debug("Queue Percentage: %d, %d\n",
 			q_properties->queue_percent, args->queue_percentage);
 
-	pr_debug("Queue Priority (%d, %d)\n",
+	pr_debug("Queue Priority: %d, %d\n",
 			q_properties->priority, args->queue_priority);
 
-	pr_debug("Queue Address (0x%llX, 0x%llX)\n",
+	pr_debug("Queue Address: 0x%llX, 0x%llX\n",
 			q_properties->queue_address, args->ring_base_address);
 
-	pr_debug("Queue Size (0x%llX, %u)\n",
+	pr_debug("Queue Size: 0x%llX, %u\n",
 			q_properties->queue_size, args->ring_size);
 
-	pr_debug("Queue r/w Pointers (0x%llX, 0x%llX)\n",
-			(uint64_t) q_properties->read_ptr,
-			(uint64_t) q_properties->write_ptr);
+	pr_debug("Queue r/w Pointers: %p, %p\n",
+			q_properties->read_ptr,
+			q_properties->write_ptr);
 
-	pr_debug("Queue Format (%d)\n", q_properties->format);
+	pr_debug("Queue Format: %d\n", q_properties->format);
 
-	pr_debug("Queue EOP (0x%llX)\n", q_properties->eop_ring_buffer_address);
+	pr_debug("Queue EOP: 0x%llX\n", q_properties->eop_ring_buffer_address);
 
-	pr_debug("Queue CTX save arex (0x%llX)\n",
+	pr_debug("Queue CTX save area: 0x%llX\n",
 			q_properties->ctx_save_restore_area_address);
 
 	return 0;
@@ -257,16 +257,16 @@ static int kfd_ioctl_create_queue(struct file *filep, struct kfd_process *p,
 
 	memset(&q_properties, 0, sizeof(struct queue_properties));
 
-	pr_debug("kfd: creating queue ioctl\n");
+	pr_debug("Creating queue ioctl\n");
 
 	err = set_queue_properties_from_user(&q_properties, args);
 	if (err)
 		return err;
 
-	pr_debug("kfd: looking for gpu id 0x%x\n", args->gpu_id);
+	pr_debug("Looking for gpu id 0x%x\n", args->gpu_id);
 	dev = kfd_device_by_id(args->gpu_id);
-	if (dev == NULL) {
-		pr_debug("kfd: gpu id 0x%x was not found\n", args->gpu_id);
+	if (!dev) {
+		pr_debug("Could not find gpu id 0x%x\n", args->gpu_id);
 		return -EINVAL;
 	}
 
@@ -278,7 +278,7 @@ static int kfd_ioctl_create_queue(struct file *filep, struct kfd_process *p,
 		goto err_bind_process;
 	}
 
-	pr_debug("kfd: creating queue for PASID %d on GPU 0x%x\n",
+	pr_debug("Creating queue for PASID %d on gpu 0x%x\n",
 			p->pasid,
 			dev->id);
 
@@ -296,15 +296,15 @@ static int kfd_ioctl_create_queue(struct file *filep, struct kfd_process *p,
 
 	mutex_unlock(&p->mutex);
 
-	pr_debug("kfd: queue id %d was created successfully\n", args->queue_id);
+	pr_debug("Queue id %d was created successfully\n", args->queue_id);
 
-	pr_debug("ring buffer address == 0x%016llX\n",
+	pr_debug("Ring buffer address == 0x%016llX\n",
 			args->ring_base_address);
 
-	pr_debug("read ptr address    == 0x%016llX\n",
+	pr_debug("Read ptr address    == 0x%016llX\n",
 			args->read_pointer_address);
 
-	pr_debug("write ptr address   == 0x%016llX\n",
+	pr_debug("Write ptr address   == 0x%016llX\n",
 			args->write_pointer_address);
 
 	return 0;
@@ -321,7 +321,7 @@ static int kfd_ioctl_destroy_queue(struct file *filp, struct kfd_process *p,
 	int retval;
 	struct kfd_ioctl_destroy_queue_args *args = data;
 
-	pr_debug("kfd: destroying queue id %d for PASID %d\n",
+	pr_debug("Destroying queue id %d for pasid %d\n",
 				args->queue_id,
 				p->pasid);
 
@@ -341,12 +341,12 @@ static int kfd_ioctl_update_queue(struct file *filp, struct kfd_process *p,
 	struct queue_properties properties;
 
 	if (args->queue_percentage > KFD_MAX_QUEUE_PERCENTAGE) {
-		pr_err("kfd: queue percentage must be between 0 to KFD_MAX_QUEUE_PERCENTAGE\n");
+		pr_err("Queue percentage must be between 0 to KFD_MAX_QUEUE_PERCENTAGE\n");
 		return -EINVAL;
 	}
 
 	if (args->queue_priority > KFD_MAX_QUEUE_PRIORITY) {
-		pr_err("kfd: queue priority must be between 0 to KFD_MAX_QUEUE_PRIORITY\n");
+		pr_err("Queue priority must be between 0 to KFD_MAX_QUEUE_PRIORITY\n");
 		return -EINVAL;
 	}
 
@@ -354,12 +354,12 @@ static int kfd_ioctl_update_queue(struct file *filp, struct kfd_process *p,
 		(!access_ok(VERIFY_WRITE,
 			(const void __user *) args->ring_base_address,
 			sizeof(uint64_t)))) {
-		pr_err("kfd: can't access ring base address\n");
+		pr_err("Can't access ring base address\n");
 		return -EFAULT;
 	}
 
 	if (!is_power_of_2(args->ring_size) && (args->ring_size != 0)) {
-		pr_err("kfd: ring size must be a power of 2 or 0\n");
+		pr_err("Ring size must be a power of 2 or 0\n");
 		return -EINVAL;
 	}
 
@@ -368,7 +368,7 @@ static int kfd_ioctl_update_queue(struct file *filp, struct kfd_process *p,
 	properties.queue_percent = args->queue_percentage;
 	properties.priority = args->queue_priority;
 
-	pr_debug("kfd: updating queue id %d for PASID %d\n",
+	pr_debug("Updating queue id %d for pasid %d\n",
 			args->queue_id, p->pasid);
 
 	mutex_lock(&p->mutex);
@@ -400,7 +400,7 @@ static int kfd_ioctl_set_memory_policy(struct file *filep,
 	}
 
 	dev = kfd_device_by_id(args->gpu_id);
-	if (dev == NULL)
+	if (!dev)
 		return -EINVAL;
 
 	mutex_lock(&p->mutex);
@@ -443,7 +443,7 @@ static int kfd_ioctl_dbg_register(struct file *filep,
 	long status = 0;
 
 	dev = kfd_device_by_id(args->gpu_id);
-	if (dev == NULL)
+	if (!dev)
 		return -EINVAL;
 
 	if (dev->device_info->asic_family == CHIP_CARRIZO) {
@@ -460,12 +460,11 @@ static int kfd_ioctl_dbg_register(struct file *filep,
 	 */
 	pdd = kfd_bind_process_to_device(dev, p);
 	if (IS_ERR(pdd)) {
-		mutex_unlock(&p->mutex);
-		mutex_unlock(kfd_get_dbgmgr_mutex());
-		return PTR_ERR(pdd);
+		status = PTR_ERR(pdd);
+		goto out;
 	}
 
-	if (dev->dbgmgr == NULL) {
+	if (!dev->dbgmgr) {
 		/* In case of a legal call, we have no dbgmgr yet */
 		create_ok = kfd_dbgmgr_create(&dbgmgr_ptr, dev);
 		if (create_ok) {
@@ -480,6 +479,7 @@ static int kfd_ioctl_dbg_register(struct file *filep,
 		status = -EINVAL;
 	}
 
+out:
 	mutex_unlock(&p->mutex);
 	mutex_unlock(kfd_get_dbgmgr_mutex());
 
@@ -494,7 +494,7 @@ static int kfd_ioctl_dbg_unregister(struct file *filep,
 	long status;
 
 	dev = kfd_device_by_id(args->gpu_id);
-	if (dev == NULL)
+	if (!dev)
 		return -EINVAL;
 
 	if (dev->device_info->asic_family == CHIP_CARRIZO) {
@@ -505,7 +505,7 @@ static int kfd_ioctl_dbg_unregister(struct file *filep,
 	mutex_lock(kfd_get_dbgmgr_mutex());
 
 	status = kfd_dbgmgr_unregister(dev->dbgmgr, p);
-	if (status == 0) {
+	if (!status) {
 		kfd_dbgmgr_destroy(dev->dbgmgr);
 		dev->dbgmgr = NULL;
 	}
@@ -539,7 +539,7 @@ static int kfd_ioctl_dbg_address_watch(struct file *filep,
 	memset((void *) &aw_info, 0, sizeof(struct dbg_address_watch_info));
 
 	dev = kfd_device_by_id(args->gpu_id);
-	if (dev == NULL)
+	if (!dev)
 		return -EINVAL;
 
 	if (dev->device_info->asic_family == CHIP_CARRIZO) {
@@ -580,8 +580,8 @@ static int kfd_ioctl_dbg_address_watch(struct file *filep,
 	args_idx += sizeof(aw_info.watch_address) * aw_info.num_watch_points;
 
 	if (args_idx >= args->buf_size_in_bytes - sizeof(*args)) {
-		kfree(args_buff);
-		return -EINVAL;
+		status = -EINVAL;
+		goto out;
 	}
 
 	watch_mask_value = (uint64_t) args_buff[args_idx];
@@ -604,8 +604,8 @@ static int kfd_ioctl_dbg_address_watch(struct file *filep,
 	}
 
 	if (args_idx >= args->buf_size_in_bytes - sizeof(args)) {
-		kfree(args_buff);
-		return -EINVAL;
+		status = -EINVAL;
+		goto out;
 	}
 
 	/* Currently HSA Event is not supported for DBG */
@@ -617,6 +617,7 @@ static int kfd_ioctl_dbg_address_watch(struct file *filep,
 
 	mutex_unlock(kfd_get_dbgmgr_mutex());
 
+out:
 	kfree(args_buff);
 
 	return status;
@@ -646,7 +647,7 @@ static int kfd_ioctl_dbg_wave_control(struct file *filep,
 				sizeof(wac_info.trapId);
 
 	dev = kfd_device_by_id(args->gpu_id);
-	if (dev == NULL)
+	if (!dev)
 		return -EINVAL;
 
 	if (dev->device_info->asic_family == CHIP_CARRIZO) {
@@ -782,8 +783,9 @@ static int kfd_ioctl_get_process_apertures(struct file *filp,
 				"scratch_limit %llX\n", pdd->scratch_limit);
 
 			args->num_of_nodes++;
-		} while ((pdd = kfd_get_next_process_device_data(p, pdd)) != NULL &&
-				(args->num_of_nodes < NUM_OF_SUPPORTED_GPUS));
+
+			pdd = kfd_get_next_process_device_data(p, pdd);
+		} while (pdd && (args->num_of_nodes < NUM_OF_SUPPORTED_GPUS));
 	}
 
 	mutex_unlock(&p->mutex);
@@ -846,9 +848,84 @@ static int kfd_ioctl_wait_events(struct file *filp, struct kfd_process *p,
 
 	return err;
 }
+static int kfd_ioctl_set_scratch_backing_va(struct file *filep,
+					struct kfd_process *p, void *data)
+{
+	struct kfd_ioctl_set_scratch_backing_va_args *args = data;
+	struct kfd_process_device *pdd;
+	struct kfd_dev *dev;
+	long err;
+
+	dev = kfd_device_by_id(args->gpu_id);
+	if (!dev)
+		return -EINVAL;
+
+	mutex_lock(&p->mutex);
+
+	pdd = kfd_bind_process_to_device(dev, p);
+	if (IS_ERR(pdd)) {
+		err = PTR_ERR(pdd);
+		goto bind_process_to_device_fail;
+	}
+
+	pdd->qpd.sh_hidden_private_base = args->va_addr;
+
+	mutex_unlock(&p->mutex);
+
+	if (sched_policy == KFD_SCHED_POLICY_NO_HWS && pdd->qpd.vmid != 0)
+		dev->kfd2kgd->set_scratch_backing_va(
+			dev->kgd, args->va_addr, pdd->qpd.vmid);
+
+	return 0;
+
+bind_process_to_device_fail:
+	mutex_unlock(&p->mutex);
+	return err;
+}
+
+static int kfd_ioctl_get_tile_config(struct file *filep,
+		struct kfd_process *p, void *data)
+{
+	struct kfd_ioctl_get_tile_config_args *args = data;
+	struct kfd_dev *dev;
+	struct tile_config config;
+	int err = 0;
+
+	dev = kfd_device_by_id(args->gpu_id);
+
+	dev->kfd2kgd->get_tile_config(dev->kgd, &config);
+
+	args->gb_addr_config = config.gb_addr_config;
+	args->num_banks = config.num_banks;
+	args->num_ranks = config.num_ranks;
+
+	if (args->num_tile_configs > config.num_tile_configs)
+		args->num_tile_configs = config.num_tile_configs;
+	err = copy_to_user((void __user *)args->tile_config_ptr,
+			config.tile_config_ptr,
+			args->num_tile_configs * sizeof(uint32_t));
+	if (err) {
+		args->num_tile_configs = 0;
+		return -EFAULT;
+	}
+
+	if (args->num_macro_tile_configs > config.num_macro_tile_configs)
+		args->num_macro_tile_configs =
+				config.num_macro_tile_configs;
+	err = copy_to_user((void __user *)args->macro_tile_config_ptr,
+			config.macro_tile_config_ptr,
+			args->num_macro_tile_configs * sizeof(uint32_t));
+	if (err) {
+		args->num_macro_tile_configs = 0;
+		return -EFAULT;
+	}
+
+	return 0;
+}
 
 #define AMDKFD_IOCTL_DEF(ioctl, _func, _flags) \
-	[_IOC_NR(ioctl)] = {.cmd = ioctl, .func = _func, .flags = _flags, .cmd_drv = 0, .name = #ioctl}
+	[_IOC_NR(ioctl)] = {.cmd = ioctl, .func = _func, .flags = _flags, \
+			    .cmd_drv = 0, .name = #ioctl}
 
 /** Ioctl table */
 static const struct amdkfd_ioctl_desc amdkfd_ioctls[] = {
@@ -899,6 +976,12 @@ static const struct amdkfd_ioctl_desc amdkfd_ioctls[] = {
 
 	AMDKFD_IOCTL_DEF(AMDKFD_IOC_DBG_WAVE_CONTROL,
 			kfd_ioctl_dbg_wave_control, 0),
+
+	AMDKFD_IOCTL_DEF(AMDKFD_IOC_SET_SCRATCH_BACKING_VA,
+			kfd_ioctl_set_scratch_backing_va, 0),
+
+	AMDKFD_IOCTL_DEF(AMDKFD_IOC_GET_TILE_CONFIG,
+			kfd_ioctl_get_tile_config, 0)
 };
 
 #define AMDKFD_CORE_IOCTL_COUNT	ARRAY_SIZE(amdkfd_ioctls)
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_dbgdev.c b/drivers/gpu/drm/amd/amdkfd/kfd_dbgdev.c
index d5e19b5fbbfb..0aa021aa0aa1 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_dbgdev.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_dbgdev.c
@@ -42,8 +42,6 @@
 
 static void dbgdev_address_watch_disable_nodiq(struct kfd_dev *dev)
 {
-	BUG_ON(!dev || !dev->kfd2kgd);
-
 	dev->kfd2kgd->address_watch_disable(dev->kgd);
 }
 
@@ -62,7 +60,8 @@ static int dbgdev_diq_submit_ib(struct kfd_dbgdev *dbgdev,
 	unsigned int *ib_packet_buff;
 	int status;
 
-	BUG_ON(!dbgdev || !dbgdev->kq || !packet_buff || !size_in_bytes);
+	if (WARN_ON(!size_in_bytes))
+		return -EINVAL;
 
 	kq = dbgdev->kq;
 
@@ -77,8 +76,8 @@ static int dbgdev_diq_submit_ib(struct kfd_dbgdev *dbgdev,
 	status = kq->ops.acquire_packet_buffer(kq,
 				pq_packets_size_in_bytes / sizeof(uint32_t),
 				&ib_packet_buff);
-	if (status != 0) {
-		pr_err("amdkfd: acquire_packet_buffer failed\n");
+	if (status) {
+		pr_err("acquire_packet_buffer failed\n");
 		return status;
 	}
 
@@ -115,8 +114,8 @@ static int dbgdev_diq_submit_ib(struct kfd_dbgdev *dbgdev,
 	status = kfd_gtt_sa_allocate(dbgdev->dev, sizeof(uint64_t),
 					&mem_obj);
 
-	if (status != 0) {
-		pr_err("amdkfd: Failed to allocate GART memory\n");
+	if (status) {
+		pr_err("Failed to allocate GART memory\n");
 		kq->ops.rollback_packet(kq);
 		return status;
 	}
@@ -168,8 +167,6 @@ static int dbgdev_diq_submit_ib(struct kfd_dbgdev *dbgdev,
 
 static int dbgdev_register_nodiq(struct kfd_dbgdev *dbgdev)
 {
-	BUG_ON(!dbgdev);
-
 	/*
 	 * no action is needed in this case,
 	 * just make sure diq will not be used
@@ -187,14 +184,12 @@ static int dbgdev_register_diq(struct kfd_dbgdev *dbgdev)
 	struct kernel_queue *kq = NULL;
 	int status;
 
-	BUG_ON(!dbgdev || !dbgdev->pqm || !dbgdev->dev);
-
 	status = pqm_create_queue(dbgdev->pqm, dbgdev->dev, NULL,
 				&properties, 0, KFD_QUEUE_TYPE_DIQ,
 				&qid);
 
 	if (status) {
-		pr_err("amdkfd: Failed to create DIQ\n");
+		pr_err("Failed to create DIQ\n");
 		return status;
 	}
 
@@ -202,8 +197,8 @@ static int dbgdev_register_diq(struct kfd_dbgdev *dbgdev)
 
 	kq = pqm_get_kernel_queue(dbgdev->pqm, qid);
 
-	if (kq == NULL) {
-		pr_err("amdkfd: Error getting DIQ\n");
+	if (!kq) {
+		pr_err("Error getting DIQ\n");
 		pqm_destroy_queue(dbgdev->pqm, qid);
 		return -EFAULT;
 	}
@@ -215,8 +210,6 @@ static int dbgdev_register_diq(struct kfd_dbgdev *dbgdev)
 
 static int dbgdev_unregister_nodiq(struct kfd_dbgdev *dbgdev)
 {
-	BUG_ON(!dbgdev || !dbgdev->dev);
-
 	/* disable watch address */
 	dbgdev_address_watch_disable_nodiq(dbgdev->dev);
 	return 0;
@@ -227,8 +220,6 @@ static int dbgdev_unregister_diq(struct kfd_dbgdev *dbgdev)
 	/* todo - disable address watch */
 	int status;
 
-	BUG_ON(!dbgdev || !dbgdev->pqm || !dbgdev->kq);
-
 	status = pqm_destroy_queue(dbgdev->pqm,
 			dbgdev->kq->queue->properties.queue_id);
 	dbgdev->kq = NULL;
@@ -245,14 +236,12 @@ static void dbgdev_address_watch_set_registers(
 {
 	union ULARGE_INTEGER addr;
 
-	BUG_ON(!adw_info || !addrHi || !addrLo || !cntl);
-
 	addr.quad_part = 0;
 	addrHi->u32All = 0;
 	addrLo->u32All = 0;
 	cntl->u32All = 0;
 
-	if (adw_info->watch_mask != NULL)
+	if (adw_info->watch_mask)
 		cntl->bitfields.mask =
 			(uint32_t) (adw_info->watch_mask[index] &
 					ADDRESS_WATCH_REG_CNTL_DEFAULT_MASK);
@@ -279,7 +268,7 @@ static void dbgdev_address_watch_set_registers(
 }
 
 static int dbgdev_address_watch_nodiq(struct kfd_dbgdev *dbgdev,
-					struct dbg_address_watch_info *adw_info)
+				      struct dbg_address_watch_info *adw_info)
 {
 	union TCP_WATCH_ADDR_H_BITS addrHi;
 	union TCP_WATCH_ADDR_L_BITS addrLo;
@@ -287,13 +276,11 @@ static int dbgdev_address_watch_nodiq(struct kfd_dbgdev *dbgdev,
 	struct kfd_process_device *pdd;
 	unsigned int i;
 
-	BUG_ON(!dbgdev || !dbgdev->dev || !adw_info);
-
 	/* taking the vmid for that process on the safe way using pdd */
 	pdd = kfd_get_process_device_data(dbgdev->dev,
 					adw_info->process);
 	if (!pdd) {
-		pr_err("amdkfd: Failed to get pdd for wave control no DIQ\n");
+		pr_err("Failed to get pdd for wave control no DIQ\n");
 		return -EFAULT;
 	}
 
@@ -303,17 +290,16 @@ static int dbgdev_address_watch_nodiq(struct kfd_dbgdev *dbgdev,
 
 	if ((adw_info->num_watch_points > MAX_WATCH_ADDRESSES) ||
 			(adw_info->num_watch_points == 0)) {
-		pr_err("amdkfd: num_watch_points is invalid\n");
+		pr_err("num_watch_points is invalid\n");
 		return -EINVAL;
 	}
 
-	if ((adw_info->watch_mode == NULL) ||
-		(adw_info->watch_address == NULL)) {
-		pr_err("amdkfd: adw_info fields are not valid\n");
+	if (!adw_info->watch_mode || !adw_info->watch_address) {
+		pr_err("adw_info fields are not valid\n");
 		return -EINVAL;
 	}
 
-	for (i = 0 ; i < adw_info->num_watch_points ; i++) {
+	for (i = 0; i < adw_info->num_watch_points; i++) {
 		dbgdev_address_watch_set_registers(adw_info, &addrHi, &addrLo,
 						&cntl, i, pdd->qpd.vmid);
 
@@ -348,7 +334,7 @@ static int dbgdev_address_watch_nodiq(struct kfd_dbgdev *dbgdev,
 }
 
 static int dbgdev_address_watch_diq(struct kfd_dbgdev *dbgdev,
-					struct dbg_address_watch_info *adw_info)
+				    struct dbg_address_watch_info *adw_info)
 {
 	struct pm4__set_config_reg *packets_vec;
 	union TCP_WATCH_ADDR_H_BITS addrHi;
@@ -363,28 +349,25 @@ static int dbgdev_address_watch_diq(struct kfd_dbgdev *dbgdev,
 	/* we do not control the vmid in DIQ mode, just a place holder */
 	unsigned int vmid = 0;
 
-	BUG_ON(!dbgdev || !dbgdev->dev || !adw_info);
-
 	addrHi.u32All = 0;
 	addrLo.u32All = 0;
 	cntl.u32All = 0;
 
 	if ((adw_info->num_watch_points > MAX_WATCH_ADDRESSES) ||
 			(adw_info->num_watch_points == 0)) {
-		pr_err("amdkfd: num_watch_points is invalid\n");
+		pr_err("num_watch_points is invalid\n");
 		return -EINVAL;
 	}
 
-	if ((NULL == adw_info->watch_mode) ||
-			(NULL == adw_info->watch_address)) {
-		pr_err("amdkfd: adw_info fields are not valid\n");
+	if (!adw_info->watch_mode || !adw_info->watch_address) {
+		pr_err("adw_info fields are not valid\n");
 		return -EINVAL;
 	}
 
 	status = kfd_gtt_sa_allocate(dbgdev->dev, ib_size, &mem_obj);
 
-	if (status != 0) {
-		pr_err("amdkfd: Failed to allocate GART memory\n");
+	if (status) {
+		pr_err("Failed to allocate GART memory\n");
 		return status;
 	}
 
@@ -442,8 +425,6 @@ static int dbgdev_address_watch_diq(struct kfd_dbgdev *dbgdev,
 					i,
 					ADDRESS_WATCH_REG_CNTL);
 
-		aw_reg_add_dword /= sizeof(uint32_t);
-
 		packets_vec[0].bitfields2.reg_offset =
 					aw_reg_add_dword - AMD_CONFIG_REG_BASE;
 
@@ -455,8 +436,6 @@ static int dbgdev_address_watch_diq(struct kfd_dbgdev *dbgdev,
 					i,
 					ADDRESS_WATCH_REG_ADDR_HI);
 
-		aw_reg_add_dword /= sizeof(uint32_t);
-
 		packets_vec[1].bitfields2.reg_offset =
 					aw_reg_add_dword - AMD_CONFIG_REG_BASE;
 		packets_vec[1].reg_data[0] = addrHi.u32All;
@@ -467,8 +446,6 @@ static int dbgdev_address_watch_diq(struct kfd_dbgdev *dbgdev,
 					i,
 					ADDRESS_WATCH_REG_ADDR_LO);
 
-		aw_reg_add_dword /= sizeof(uint32_t);
-
 		packets_vec[2].bitfields2.reg_offset =
 				aw_reg_add_dword - AMD_CONFIG_REG_BASE;
 		packets_vec[2].reg_data[0] = addrLo.u32All;
@@ -485,8 +462,6 @@ static int dbgdev_address_watch_diq(struct kfd_dbgdev *dbgdev,
 					i,
 					ADDRESS_WATCH_REG_CNTL);
 
-		aw_reg_add_dword /= sizeof(uint32_t);
-
 		packets_vec[3].bitfields2.reg_offset =
 					aw_reg_add_dword - AMD_CONFIG_REG_BASE;
 		packets_vec[3].reg_data[0] = cntl.u32All;
@@ -498,8 +473,8 @@ static int dbgdev_address_watch_diq(struct kfd_dbgdev *dbgdev,
 					packet_buff_uint,
 					ib_size);
 
-		if (status != 0) {
-			pr_err("amdkfd: Failed to submit IB to DIQ\n");
+		if (status) {
+			pr_err("Failed to submit IB to DIQ\n");
 			break;
 		}
 	}
@@ -518,8 +493,6 @@ static int dbgdev_wave_control_set_registers(
 	union GRBM_GFX_INDEX_BITS reg_gfx_index;
 	struct HsaDbgWaveMsgAMDGen2 *pMsg;
 
-	BUG_ON(!wac_info || !in_reg_sq_cmd || !in_reg_gfx_index);
-
 	reg_sq_cmd.u32All = 0;
 	reg_gfx_index.u32All = 0;
 	pMsg = &wac_info->dbgWave_msg.DbgWaveMsg.WaveMsgInfoGen2;
@@ -620,18 +593,16 @@ static int dbgdev_wave_control_diq(struct kfd_dbgdev *dbgdev,
 	struct pm4__set_config_reg *packets_vec;
 	size_t ib_size = sizeof(struct pm4__set_config_reg) * 3;
 
-	BUG_ON(!dbgdev || !wac_info);
-
 	reg_sq_cmd.u32All = 0;
 
 	status = dbgdev_wave_control_set_registers(wac_info, &reg_sq_cmd,
 							&reg_gfx_index);
 	if (status) {
-		pr_err("amdkfd: Failed to set wave control registers\n");
+		pr_err("Failed to set wave control registers\n");
 		return status;
 	}
 
-	/* we do not control the VMID in DIQ,so reset it to a known value */
+	/* we do not control the VMID in DIQ, so reset it to a known value */
 	reg_sq_cmd.bits.vm_id = 0;
 
 	pr_debug("\t\t %30s\n", "* * * * * * * * * * * * * * * * * *");
@@ -667,7 +638,7 @@ static int dbgdev_wave_control_diq(struct kfd_dbgdev *dbgdev,
 	status = kfd_gtt_sa_allocate(dbgdev->dev, ib_size, &mem_obj);
 
 	if (status != 0) {
-		pr_err("amdkfd: Failed to allocate GART memory\n");
+		pr_err("Failed to allocate GART memory\n");
 		return status;
 	}
 
@@ -719,8 +690,8 @@ static int dbgdev_wave_control_diq(struct kfd_dbgdev *dbgdev,
 			packet_buff_uint,
 			ib_size);
 
-	if (status != 0)
-		pr_err("amdkfd: Failed to submit IB to DIQ\n");
+	if (status)
+		pr_err("Failed to submit IB to DIQ\n");
 
 	kfd_gtt_sa_free(dbgdev->dev, mem_obj);
 
@@ -735,21 +706,19 @@ static int dbgdev_wave_control_nodiq(struct kfd_dbgdev *dbgdev,
 	union GRBM_GFX_INDEX_BITS reg_gfx_index;
 	struct kfd_process_device *pdd;
 
-	BUG_ON(!dbgdev || !dbgdev->dev || !wac_info);
-
 	reg_sq_cmd.u32All = 0;
 
 	/* taking the VMID for that process on the safe way using PDD */
 	pdd = kfd_get_process_device_data(dbgdev->dev, wac_info->process);
 
 	if (!pdd) {
-		pr_err("amdkfd: Failed to get pdd for wave control no DIQ\n");
+		pr_err("Failed to get pdd for wave control no DIQ\n");
 		return -EFAULT;
 	}
 	status = dbgdev_wave_control_set_registers(wac_info, &reg_sq_cmd,
 							&reg_gfx_index);
 	if (status) {
-		pr_err("amdkfd: Failed to set wave control registers\n");
+		pr_err("Failed to set wave control registers\n");
 		return status;
 	}
 
@@ -818,12 +787,13 @@ int dbgdev_wave_reset_wavefronts(struct kfd_dev *dev, struct kfd_process *p)
 
 	/* Scan all registers in the range ATC_VMID8_PASID_MAPPING ..
 	 * ATC_VMID15_PASID_MAPPING
-	 * to check which VMID the current process is mapped to. */
+	 * to check which VMID the current process is mapped to.
+	 */
 
 	for (vmid = first_vmid_to_scan; vmid <= last_vmid_to_scan; vmid++) {
 		if (dev->kfd2kgd->get_atc_vmid_pasid_mapping_valid
 				(dev->kgd, vmid)) {
-			if (dev->kfd2kgd->get_atc_vmid_pasid_mapping_valid
+			if (dev->kfd2kgd->get_atc_vmid_pasid_mapping_pasid
 					(dev->kgd, vmid) == p->pasid) {
 				pr_debug("Killing wave fronts of vmid %d and pasid %d\n",
 						vmid, p->pasid);
@@ -833,7 +803,7 @@ int dbgdev_wave_reset_wavefronts(struct kfd_dev *dev, struct kfd_process *p)
 	}
 
 	if (vmid > last_vmid_to_scan) {
-		pr_err("amdkfd: didn't found vmid for pasid (%d)\n", p->pasid);
+		pr_err("Didn't find vmid for pasid %d\n", p->pasid);
 		return -EFAULT;
 	}
 
@@ -860,8 +830,6 @@ int dbgdev_wave_reset_wavefronts(struct kfd_dev *dev, struct kfd_process *p)
 void kfd_dbgdev_init(struct kfd_dbgdev *pdbgdev, struct kfd_dev *pdev,
 			enum DBGDEV_TYPE type)
 {
-	BUG_ON(!pdbgdev || !pdev);
-
 	pdbgdev->dev = pdev;
 	pdbgdev->kq = NULL;
 	pdbgdev->type = type;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_dbgmgr.c b/drivers/gpu/drm/amd/amdkfd/kfd_dbgmgr.c
index 56d676396342..3da25f7bda6b 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_dbgmgr.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_dbgmgr.c
@@ -44,8 +44,6 @@ struct mutex *kfd_get_dbgmgr_mutex(void)
 
 static void kfd_dbgmgr_uninitialize(struct kfd_dbgmgr *pmgr)
 {
-	BUG_ON(!pmgr);
-
 	kfree(pmgr->dbgdev);
 
 	pmgr->dbgdev = NULL;
@@ -55,7 +53,7 @@ static void kfd_dbgmgr_uninitialize(struct kfd_dbgmgr *pmgr)
 
 void kfd_dbgmgr_destroy(struct kfd_dbgmgr *pmgr)
 {
-	if (pmgr != NULL) {
+	if (pmgr) {
 		kfd_dbgmgr_uninitialize(pmgr);
 		kfree(pmgr);
 	}
@@ -66,12 +64,12 @@ bool kfd_dbgmgr_create(struct kfd_dbgmgr **ppmgr, struct kfd_dev *pdev)
 	enum DBGDEV_TYPE type = DBGDEV_TYPE_DIQ;
 	struct kfd_dbgmgr *new_buff;
 
-	BUG_ON(pdev == NULL);
-	BUG_ON(!pdev->init_complete);
+	if (WARN_ON(!pdev->init_complete))
+		return false;
 
 	new_buff = kfd_alloc_struct(new_buff);
 	if (!new_buff) {
-		pr_err("amdkfd: Failed to allocate dbgmgr instance\n");
+		pr_err("Failed to allocate dbgmgr instance\n");
 		return false;
 	}
 
@@ -79,7 +77,7 @@ bool kfd_dbgmgr_create(struct kfd_dbgmgr **ppmgr, struct kfd_dev *pdev)
 	new_buff->dev = pdev;
 	new_buff->dbgdev = kfd_alloc_struct(new_buff->dbgdev);
 	if (!new_buff->dbgdev) {
-		pr_err("amdkfd: Failed to allocate dbgdev instance\n");
+		pr_err("Failed to allocate dbgdev instance\n");
 		kfree(new_buff);
 		return false;
 	}
@@ -96,8 +94,6 @@ bool kfd_dbgmgr_create(struct kfd_dbgmgr **ppmgr, struct kfd_dev *pdev)
 
 long kfd_dbgmgr_register(struct kfd_dbgmgr *pmgr, struct kfd_process *p)
 {
-	BUG_ON(!p || !pmgr || !pmgr->dbgdev);
-
 	if (pmgr->pasid != 0) {
 		pr_debug("H/W debugger is already active using pasid %d\n",
 				pmgr->pasid);
@@ -118,8 +114,6 @@ long kfd_dbgmgr_register(struct kfd_dbgmgr *pmgr, struct kfd_process *p)
 
 long kfd_dbgmgr_unregister(struct kfd_dbgmgr *pmgr, struct kfd_process *p)
 {
-	BUG_ON(!p || !pmgr || !pmgr->dbgdev);
-
 	/* Is the requests coming from the already registered process? */
 	if (pmgr->pasid != p->pasid) {
 		pr_debug("H/W debugger is not registered by calling pasid %d\n",
@@ -137,8 +131,6 @@ long kfd_dbgmgr_unregister(struct kfd_dbgmgr *pmgr, struct kfd_process *p)
 long kfd_dbgmgr_wave_control(struct kfd_dbgmgr *pmgr,
 				struct dbg_wave_control_info *wac_info)
 {
-	BUG_ON(!pmgr || !pmgr->dbgdev || !wac_info);
-
 	/* Is the requests coming from the already registered process? */
 	if (pmgr->pasid != wac_info->process->pasid) {
 		pr_debug("H/W debugger support was not registered for requester pasid %d\n",
@@ -152,9 +144,6 @@ long kfd_dbgmgr_wave_control(struct kfd_dbgmgr *pmgr,
 long kfd_dbgmgr_address_watch(struct kfd_dbgmgr *pmgr,
 				struct dbg_address_watch_info *adw_info)
 {
-	BUG_ON(!pmgr || !pmgr->dbgdev || !adw_info);
-
-
 	/* Is the requests coming from the already registered process? */
 	if (pmgr->pasid != adw_info->process->pasid) {
 		pr_debug("H/W debugger support was not registered for requester pasid %d\n",
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_dbgmgr.h b/drivers/gpu/drm/amd/amdkfd/kfd_dbgmgr.h
index 257a745ad0b5..a04a1fe1d0d9 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_dbgmgr.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_dbgmgr.h
@@ -30,13 +30,11 @@
 #pragma pack(push, 4)
 
 enum HSA_DBG_WAVEOP {
-	HSA_DBG_WAVEOP_HALT = 1,	/* Halts a wavefront		*/
-	HSA_DBG_WAVEOP_RESUME = 2,	/* Resumes a wavefront		*/
-	HSA_DBG_WAVEOP_KILL = 3,	/* Kills a wavefront		*/
-	HSA_DBG_WAVEOP_DEBUG = 4,	/* Causes wavefront to enter
-						debug mode		*/
-	HSA_DBG_WAVEOP_TRAP = 5,	/* Causes wavefront to take
-						a trap			*/
+	HSA_DBG_WAVEOP_HALT = 1,   /* Halts a wavefront */
+	HSA_DBG_WAVEOP_RESUME = 2, /* Resumes a wavefront */
+	HSA_DBG_WAVEOP_KILL = 3,   /* Kills a wavefront */
+	HSA_DBG_WAVEOP_DEBUG = 4,  /* Causes wavefront to enter dbg mode */
+	HSA_DBG_WAVEOP_TRAP = 5,   /* Causes wavefront to take a trap */
 	HSA_DBG_NUM_WAVEOP = 5,
 	HSA_DBG_MAX_WAVEOP = 0xFFFFFFFF
 };
@@ -81,15 +79,13 @@ struct HsaDbgWaveMsgAMDGen2 {
 			uint32_t UserData:8;	/* user data */
 			uint32_t ShaderArray:1;	/* Shader array */
 			uint32_t Priv:1;	/* Privileged */
-			uint32_t Reserved0:4;	/* This field is reserved,
-						   should be 0 */
+			uint32_t Reserved0:4;	/* Reserved, should be 0 */
 			uint32_t WaveId:4;	/* wave id */
 			uint32_t SIMD:2;	/* SIMD id */
 			uint32_t HSACU:4;	/* Compute unit */
 			uint32_t ShaderEngine:2;/* Shader engine */
 			uint32_t MessageType:2;	/* see HSA_DBG_WAVEMSG_TYPE */
-			uint32_t Reserved1:4;	/* This field is reserved,
-						   should be 0 */
+			uint32_t Reserved1:4;	/* Reserved, should be 0 */
 		} ui32;
 		uint32_t Value;
 	};
@@ -121,20 +117,23 @@ struct HsaDbgWaveMessage {
  * in the user mode instruction stream. The OS scheduler event is typically
  * associated and signaled by an interrupt issued by the GPU, but other HSA
  * system interrupt conditions from other HW (e.g. IOMMUv2) may be surfaced
- * by the KFD by this mechanism, too. */
+ * by the KFD by this mechanism, too.
+ */
 
 /* these are the new definitions for events */
 enum HSA_EVENTTYPE {
 	HSA_EVENTTYPE_SIGNAL = 0,	/* user-mode generated GPU signal */
 	HSA_EVENTTYPE_NODECHANGE = 1,	/* HSA node change (attach/detach) */
 	HSA_EVENTTYPE_DEVICESTATECHANGE = 2,	/* HSA device state change
-						   (start/stop) */
+						 * (start/stop)
+						 */
 	HSA_EVENTTYPE_HW_EXCEPTION = 3,	/* GPU shader exception event */
 	HSA_EVENTTYPE_SYSTEM_EVENT = 4,	/* GPU SYSCALL with parameter info */
 	HSA_EVENTTYPE_DEBUG_EVENT = 5,	/* GPU signal for debugging */
 	HSA_EVENTTYPE_PROFILE_EVENT = 6,/* GPU signal for profiling */
 	HSA_EVENTTYPE_QUEUE_EVENT = 7,	/* GPU signal queue idle state
-					   (EOP pm4) */
+					 * (EOP pm4)
+					 */
 	/* ...  */
 	HSA_EVENTTYPE_MAXID,
 	HSA_EVENTTYPE_TYPE_SIZE = 0xFFFFFFFF
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
index 3f95f7cb4019..61fff25b4ce7 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
@@ -26,7 +26,7 @@
 #include <linux/slab.h>
 #include "kfd_priv.h"
 #include "kfd_device_queue_manager.h"
-#include "kfd_pm4_headers.h"
+#include "kfd_pm4_headers_vi.h"
 
 #define MQD_SIZE_ALIGNED 768
 
@@ -98,11 +98,14 @@ static const struct kfd_device_info *lookup_device_info(unsigned short did)
 
 	for (i = 0; i < ARRAY_SIZE(supported_devices); i++) {
 		if (supported_devices[i].did == did) {
-			BUG_ON(supported_devices[i].device_info == NULL);
+			WARN_ON(!supported_devices[i].device_info);
 			return supported_devices[i].device_info;
 		}
 	}
 
+	dev_warn(kfd_device, "DID %04x is missing in supported_devices\n",
+		 did);
+
 	return NULL;
 }
 
@@ -114,8 +117,10 @@ struct kfd_dev *kgd2kfd_probe(struct kgd_dev *kgd,
 	const struct kfd_device_info *device_info =
 					lookup_device_info(pdev->device);
 
-	if (!device_info)
+	if (!device_info) {
+		dev_err(kfd_device, "kgd2kfd_probe failed\n");
 		return NULL;
+	}
 
 	kfd = kzalloc(sizeof(*kfd), GFP_KERNEL);
 	if (!kfd)
@@ -152,15 +157,16 @@ static bool device_iommu_pasid_init(struct kfd_dev *kfd)
 	}
 
 	if ((iommu_info.flags & required_iommu_flags) != required_iommu_flags) {
-		dev_err(kfd_device, "error required iommu flags ats(%i), pri(%i), pasid(%i)\n",
+		dev_err(kfd_device, "error required iommu flags ats %i, pri %i, pasid %i\n",
 		       (iommu_info.flags & AMD_IOMMU_DEVICE_FLAG_ATS_SUP) != 0,
 		       (iommu_info.flags & AMD_IOMMU_DEVICE_FLAG_PRI_SUP) != 0,
-		       (iommu_info.flags & AMD_IOMMU_DEVICE_FLAG_PASID_SUP) != 0);
+		       (iommu_info.flags & AMD_IOMMU_DEVICE_FLAG_PASID_SUP)
+									!= 0);
 		return false;
 	}
 
 	pasid_limit = min_t(unsigned int,
-			(unsigned int)1 << kfd->device_info->max_pasid_bits,
+			(unsigned int)(1 << kfd->device_info->max_pasid_bits),
 			iommu_info.max_pasids);
 	/*
 	 * last pasid is used for kernel queues doorbells
@@ -211,9 +217,8 @@ static int iommu_invalid_ppr_cb(struct pci_dev *pdev, int pasid,
 			flags);
 
 	dev = kfd_device_by_pci_dev(pdev);
-	BUG_ON(dev == NULL);
-
-	kfd_signal_iommu_event(dev, pasid, address,
+	if (!WARN_ON(!dev))
+		kfd_signal_iommu_event(dev, pasid, address,
 			flags & PPR_FAULT_WRITE, flags & PPR_FAULT_EXEC);
 
 	return AMD_IOMMU_INV_PRI_RSP_INVALID;
@@ -234,9 +239,9 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 	 * calculate max size of runlist packet.
 	 * There can be only 2 packets at once
 	 */
-	size += (KFD_MAX_NUM_OF_PROCESSES * sizeof(struct pm4_map_process) +
-		max_num_of_queues_per_device *
-		sizeof(struct pm4_map_queues) + sizeof(struct pm4_runlist)) * 2;
+	size += (KFD_MAX_NUM_OF_PROCESSES * sizeof(struct pm4_mes_map_process) +
+		max_num_of_queues_per_device * sizeof(struct pm4_mes_map_queues)
+		+ sizeof(struct pm4_mes_runlist)) * 2;
 
 	/* Add size of HIQ & DIQ */
 	size += KFD_KERNEL_QUEUE_SIZE * 2;
@@ -247,42 +252,37 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 	if (kfd->kfd2kgd->init_gtt_mem_allocation(
 			kfd->kgd, size, &kfd->gtt_mem,
 			&kfd->gtt_start_gpu_addr, &kfd->gtt_start_cpu_ptr)){
-		dev_err(kfd_device,
-			"Could not allocate %d bytes for device (%x:%x)\n",
-			size, kfd->pdev->vendor, kfd->pdev->device);
+		dev_err(kfd_device, "Could not allocate %d bytes\n", size);
 		goto out;
 	}
 
-	dev_info(kfd_device,
-		"Allocated %d bytes on gart for device(%x:%x)\n",
-		size, kfd->pdev->vendor, kfd->pdev->device);
+	dev_info(kfd_device, "Allocated %d bytes on gart\n", size);
 
 	/* Initialize GTT sa with 512 byte chunk size */
 	if (kfd_gtt_sa_init(kfd, size, 512) != 0) {
-		dev_err(kfd_device,
-			"Error initializing gtt sub-allocator\n");
+		dev_err(kfd_device, "Error initializing gtt sub-allocator\n");
 		goto kfd_gtt_sa_init_error;
 	}
 
-	kfd_doorbell_init(kfd);
-
-	if (kfd_topology_add_device(kfd) != 0) {
+	if (kfd_doorbell_init(kfd)) {
 		dev_err(kfd_device,
-			"Error adding device (%x:%x) to topology\n",
-			kfd->pdev->vendor, kfd->pdev->device);
+			"Error initializing doorbell aperture\n");
+		goto kfd_doorbell_error;
+	}
+
+	if (kfd_topology_add_device(kfd)) {
+		dev_err(kfd_device, "Error adding device to topology\n");
 		goto kfd_topology_add_device_error;
 	}
 
 	if (kfd_interrupt_init(kfd)) {
-		dev_err(kfd_device,
-			"Error initializing interrupts for device (%x:%x)\n",
-			kfd->pdev->vendor, kfd->pdev->device);
+		dev_err(kfd_device, "Error initializing interrupts\n");
 		goto kfd_interrupt_error;
 	}
 
 	if (!device_iommu_pasid_init(kfd)) {
 		dev_err(kfd_device,
-			"Error initializing iommuv2 for device (%x:%x)\n",
+			"Error initializing iommuv2 for device %x:%x\n",
 			kfd->pdev->vendor, kfd->pdev->device);
 		goto device_iommu_pasid_error;
 	}
@@ -292,15 +292,13 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 
 	kfd->dqm = device_queue_manager_init(kfd);
 	if (!kfd->dqm) {
-		dev_err(kfd_device,
-			"Error initializing queue manager for device (%x:%x)\n",
-			kfd->pdev->vendor, kfd->pdev->device);
+		dev_err(kfd_device, "Error initializing queue manager\n");
 		goto device_queue_manager_error;
 	}
 
-	if (kfd->dqm->ops.start(kfd->dqm) != 0) {
+	if (kfd->dqm->ops.start(kfd->dqm)) {
 		dev_err(kfd_device,
-			"Error starting queuen manager for device (%x:%x)\n",
+			"Error starting queue manager for device %x:%x\n",
 			kfd->pdev->vendor, kfd->pdev->device);
 		goto dqm_start_error;
 	}
@@ -308,10 +306,10 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 	kfd->dbgmgr = NULL;
 
 	kfd->init_complete = true;
-	dev_info(kfd_device, "added device (%x:%x)\n", kfd->pdev->vendor,
+	dev_info(kfd_device, "added device %x:%x\n", kfd->pdev->vendor,
 		 kfd->pdev->device);
 
-	pr_debug("kfd: Starting kfd with the following scheduling policy %d\n",
+	pr_debug("Starting kfd with the following scheduling policy %d\n",
 		sched_policy);
 
 	goto out;
@@ -325,11 +323,13 @@ device_iommu_pasid_error:
 kfd_interrupt_error:
 	kfd_topology_remove_device(kfd);
 kfd_topology_add_device_error:
+	kfd_doorbell_fini(kfd);
+kfd_doorbell_error:
 	kfd_gtt_sa_fini(kfd);
 kfd_gtt_sa_init_error:
 	kfd->kfd2kgd->free_gtt_mem(kfd->kgd, kfd->gtt_mem);
 	dev_err(kfd_device,
-		"device (%x:%x) NOT added due to errors\n",
+		"device %x:%x NOT added due to errors\n",
 		kfd->pdev->vendor, kfd->pdev->device);
 out:
 	return kfd->init_complete;
@@ -342,6 +342,7 @@ void kgd2kfd_device_exit(struct kfd_dev *kfd)
 		amd_iommu_free_device(kfd->pdev);
 		kfd_interrupt_exit(kfd);
 		kfd_topology_remove_device(kfd);
+		kfd_doorbell_fini(kfd);
 		kfd_gtt_sa_fini(kfd);
 		kfd->kfd2kgd->free_gtt_mem(kfd->kgd, kfd->gtt_mem);
 	}
@@ -351,8 +352,6 @@ void kgd2kfd_device_exit(struct kfd_dev *kfd)
 
 void kgd2kfd_suspend(struct kfd_dev *kfd)
 {
-	BUG_ON(kfd == NULL);
-
 	if (kfd->init_complete) {
 		kfd->dqm->ops.stop(kfd->dqm);
 		amd_iommu_set_invalidate_ctx_cb(kfd->pdev, NULL);
@@ -366,14 +365,15 @@ int kgd2kfd_resume(struct kfd_dev *kfd)
 	unsigned int pasid_limit;
 	int err;
 
-	BUG_ON(kfd == NULL);
-
 	pasid_limit = kfd_get_pasid_limit();
 
 	if (kfd->init_complete) {
 		err = amd_iommu_init_device(kfd->pdev, pasid_limit);
-		if (err < 0)
+		if (err < 0) {
+			dev_err(kfd_device, "failed to initialize iommu\n");
 			return -ENXIO;
+		}
+
 		amd_iommu_set_invalidate_ctx_cb(kfd->pdev,
 						iommu_pasid_shutdown_callback);
 		amd_iommu_set_invalid_ppr_cb(kfd->pdev, iommu_invalid_ppr_cb);
@@ -402,26 +402,27 @@ void kgd2kfd_interrupt(struct kfd_dev *kfd, const void *ih_ring_entry)
 static int kfd_gtt_sa_init(struct kfd_dev *kfd, unsigned int buf_size,
 				unsigned int chunk_size)
 {
-	unsigned int num_of_bits;
+	unsigned int num_of_longs;
 
-	BUG_ON(!kfd);
-	BUG_ON(!kfd->gtt_mem);
-	BUG_ON(buf_size < chunk_size);
-	BUG_ON(buf_size == 0);
-	BUG_ON(chunk_size == 0);
+	if (WARN_ON(buf_size < chunk_size))
+		return -EINVAL;
+	if (WARN_ON(buf_size == 0))
+		return -EINVAL;
+	if (WARN_ON(chunk_size == 0))
+		return -EINVAL;
 
 	kfd->gtt_sa_chunk_size = chunk_size;
 	kfd->gtt_sa_num_of_chunks = buf_size / chunk_size;
 
-	num_of_bits = kfd->gtt_sa_num_of_chunks / BITS_PER_BYTE;
-	BUG_ON(num_of_bits == 0);
+	num_of_longs = (kfd->gtt_sa_num_of_chunks + BITS_PER_LONG - 1) /
+		BITS_PER_LONG;
 
-	kfd->gtt_sa_bitmap = kzalloc(num_of_bits, GFP_KERNEL);
+	kfd->gtt_sa_bitmap = kcalloc(num_of_longs, sizeof(long), GFP_KERNEL);
 
 	if (!kfd->gtt_sa_bitmap)
 		return -ENOMEM;
 
-	pr_debug("kfd: gtt_sa_num_of_chunks = %d, gtt_sa_bitmap = %p\n",
+	pr_debug("gtt_sa_num_of_chunks = %d, gtt_sa_bitmap = %p\n",
 			kfd->gtt_sa_num_of_chunks, kfd->gtt_sa_bitmap);
 
 	mutex_init(&kfd->gtt_sa_lock);
@@ -455,8 +456,6 @@ int kfd_gtt_sa_allocate(struct kfd_dev *kfd, unsigned int size,
 {
 	unsigned int found, start_search, cur_size;
 
-	BUG_ON(!kfd);
-
 	if (size == 0)
 		return -EINVAL;
 
@@ -467,7 +466,7 @@ int kfd_gtt_sa_allocate(struct kfd_dev *kfd, unsigned int size,
 	if ((*mem_obj) == NULL)
 		return -ENOMEM;
 
-	pr_debug("kfd: allocated mem_obj = %p for size = %d\n", *mem_obj, size);
+	pr_debug("Allocated mem_obj = %p for size = %d\n", *mem_obj, size);
 
 	start_search = 0;
 
@@ -479,7 +478,7 @@ kfd_gtt_restart_search:
 					kfd->gtt_sa_num_of_chunks,
 					start_search);
 
-	pr_debug("kfd: found = %d\n", found);
+	pr_debug("Found = %d\n", found);
 
 	/* If there wasn't any free chunk, bail out */
 	if (found == kfd->gtt_sa_num_of_chunks)
@@ -497,12 +496,12 @@ kfd_gtt_restart_search:
 					found,
 					kfd->gtt_sa_chunk_size);
 
-	pr_debug("kfd: gpu_addr = %p, cpu_addr = %p\n",
+	pr_debug("gpu_addr = %p, cpu_addr = %p\n",
 			(uint64_t *) (*mem_obj)->gpu_addr, (*mem_obj)->cpu_ptr);
 
 	/* If we need only one chunk, mark it as allocated and get out */
 	if (size <= kfd->gtt_sa_chunk_size) {
-		pr_debug("kfd: single bit\n");
+		pr_debug("Single bit\n");
 		set_bit(found, kfd->gtt_sa_bitmap);
 		goto kfd_gtt_out;
 	}
@@ -537,7 +536,7 @@ kfd_gtt_restart_search:
 
 	} while (cur_size > 0);
 
-	pr_debug("kfd: range_start = %d, range_end = %d\n",
+	pr_debug("range_start = %d, range_end = %d\n",
 		(*mem_obj)->range_start, (*mem_obj)->range_end);
 
 	/* Mark the chunks as allocated */
@@ -551,7 +550,7 @@ kfd_gtt_out:
 	return 0;
 
 kfd_gtt_no_free_chunk:
-	pr_debug("kfd: allocation failed with mem_obj = %p\n", mem_obj);
+	pr_debug("Allocation failed with mem_obj = %p\n", mem_obj);
 	mutex_unlock(&kfd->gtt_sa_lock);
 	kfree(mem_obj);
 	return -ENOMEM;
@@ -561,13 +560,11 @@ int kfd_gtt_sa_free(struct kfd_dev *kfd, struct kfd_mem_obj *mem_obj)
 {
 	unsigned int bit;
 
-	BUG_ON(!kfd);
-
 	/* Act like kfree when trying to free a NULL object */
 	if (!mem_obj)
 		return 0;
 
-	pr_debug("kfd: free mem_obj = %p, range_start = %d, range_end = %d\n",
+	pr_debug("Free mem_obj = %p, range_start = %d, range_end = %d\n",
 			mem_obj, mem_obj->range_start, mem_obj->range_end);
 
 	mutex_lock(&kfd->gtt_sa_lock);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
index 42de22bbe14c..53a66e821624 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
@@ -79,20 +79,17 @@ static bool is_pipe_enabled(struct device_queue_manager *dqm, int mec, int pipe)
 
 unsigned int get_queues_num(struct device_queue_manager *dqm)
 {
-	BUG_ON(!dqm || !dqm->dev);
 	return bitmap_weight(dqm->dev->shared_resources.queue_bitmap,
 				KGD_MAX_QUEUES);
 }
 
 unsigned int get_queues_per_pipe(struct device_queue_manager *dqm)
 {
-	BUG_ON(!dqm || !dqm->dev);
 	return dqm->dev->shared_resources.num_queue_per_pipe;
 }
 
 unsigned int get_pipes_per_mec(struct device_queue_manager *dqm)
 {
-	BUG_ON(!dqm || !dqm->dev);
 	return dqm->dev->shared_resources.num_pipe_per_mec;
 }
 
@@ -121,7 +118,7 @@ static int allocate_vmid(struct device_queue_manager *dqm,
 
 	/* Kaveri kfd vmid's starts from vmid 8 */
 	allocated_vmid = bit + KFD_VMID_START_OFFSET;
-	pr_debug("kfd: vmid allocation %d\n", allocated_vmid);
+	pr_debug("vmid allocation %d\n", allocated_vmid);
 	qpd->vmid = allocated_vmid;
 	q->properties.vmid = allocated_vmid;
 
@@ -152,42 +149,38 @@ static int create_queue_nocpsch(struct device_queue_manager *dqm,
 {
 	int retval;
 
-	BUG_ON(!dqm || !q || !qpd || !allocated_vmid);
-
-	pr_debug("kfd: In func %s\n", __func__);
 	print_queue(q);
 
 	mutex_lock(&dqm->lock);
 
 	if (dqm->total_queue_count >= max_num_of_queues_per_device) {
-		pr_warn("amdkfd: Can't create new usermode queue because %d queues were already created\n",
+		pr_warn("Can't create new usermode queue because %d queues were already created\n",
 				dqm->total_queue_count);
-		mutex_unlock(&dqm->lock);
-		return -EPERM;
+		retval = -EPERM;
+		goto out_unlock;
 	}
 
 	if (list_empty(&qpd->queues_list)) {
 		retval = allocate_vmid(dqm, qpd, q);
-		if (retval != 0) {
-			mutex_unlock(&dqm->lock);
-			return retval;
-		}
+		if (retval)
+			goto out_unlock;
 	}
 	*allocated_vmid = qpd->vmid;
 	q->properties.vmid = qpd->vmid;
 
 	if (q->properties.type == KFD_QUEUE_TYPE_COMPUTE)
 		retval = create_compute_queue_nocpsch(dqm, q, qpd);
-	if (q->properties.type == KFD_QUEUE_TYPE_SDMA)
+	else if (q->properties.type == KFD_QUEUE_TYPE_SDMA)
 		retval = create_sdma_queue_nocpsch(dqm, q, qpd);
+	else
+		retval = -EINVAL;
 
-	if (retval != 0) {
+	if (retval) {
 		if (list_empty(&qpd->queues_list)) {
 			deallocate_vmid(dqm, qpd, q);
 			*allocated_vmid = 0;
 		}
-		mutex_unlock(&dqm->lock);
-		return retval;
+		goto out_unlock;
 	}
 
 	list_add(&q->list, &qpd->queues_list);
@@ -205,8 +198,9 @@ static int create_queue_nocpsch(struct device_queue_manager *dqm,
 	pr_debug("Total of %d queues are accountable so far\n",
 			dqm->total_queue_count);
 
+out_unlock:
 	mutex_unlock(&dqm->lock);
-	return 0;
+	return retval;
 }
 
 static int allocate_hqd(struct device_queue_manager *dqm, struct queue *q)
@@ -216,7 +210,8 @@ static int allocate_hqd(struct device_queue_manager *dqm, struct queue *q)
 
 	set = false;
 
-	for (pipe = dqm->next_pipe_to_allocate, i = 0; i < get_pipes_per_mec(dqm);
+	for (pipe = dqm->next_pipe_to_allocate, i = 0;
+			i < get_pipes_per_mec(dqm);
 			pipe = ((pipe + 1) % get_pipes_per_mec(dqm)), ++i) {
 
 		if (!is_pipe_enabled(dqm, 0, pipe))
@@ -239,8 +234,7 @@ static int allocate_hqd(struct device_queue_manager *dqm, struct queue *q)
 	if (!set)
 		return -EBUSY;
 
-	pr_debug("kfd: DQM %s hqd slot - pipe (%d) queue(%d)\n",
-				__func__, q->pipe, q->queue);
+	pr_debug("hqd slot - pipe %d, queue %d\n", q->pipe, q->queue);
 	/* horizontal hqd allocation */
 	dqm->next_pipe_to_allocate = (pipe + 1) % get_pipes_per_mec(dqm);
 
@@ -260,36 +254,38 @@ static int create_compute_queue_nocpsch(struct device_queue_manager *dqm,
 	int retval;
 	struct mqd_manager *mqd;
 
-	BUG_ON(!dqm || !q || !qpd);
-
 	mqd = dqm->ops.get_mqd_manager(dqm, KFD_MQD_TYPE_COMPUTE);
-	if (mqd == NULL)
+	if (!mqd)
 		return -ENOMEM;
 
 	retval = allocate_hqd(dqm, q);
-	if (retval != 0)
+	if (retval)
 		return retval;
 
 	retval = mqd->init_mqd(mqd, &q->mqd, &q->mqd_mem_obj,
 				&q->gart_mqd_addr, &q->properties);
-	if (retval != 0) {
-		deallocate_hqd(dqm, q);
-		return retval;
-	}
+	if (retval)
+		goto out_deallocate_hqd;
 
-	pr_debug("kfd: loading mqd to hqd on pipe (%d) queue (%d)\n",
-			q->pipe,
-			q->queue);
+	pr_debug("Loading mqd to hqd on pipe %d, queue %d\n",
+			q->pipe, q->queue);
 
-	retval = mqd->load_mqd(mqd, q->mqd, q->pipe,
-			q->queue, (uint32_t __user *) q->properties.write_ptr);
-	if (retval != 0) {
-		deallocate_hqd(dqm, q);
-		mqd->uninit_mqd(mqd, q->mqd, q->mqd_mem_obj);
-		return retval;
-	}
+	dqm->dev->kfd2kgd->set_scratch_backing_va(
+			dqm->dev->kgd, qpd->sh_hidden_private_base, qpd->vmid);
+
+	retval = mqd->load_mqd(mqd, q->mqd, q->pipe, q->queue, &q->properties,
+			       q->process->mm);
+	if (retval)
+		goto out_uninit_mqd;
 
 	return 0;
+
+out_uninit_mqd:
+	mqd->uninit_mqd(mqd, q->mqd, q->mqd_mem_obj);
+out_deallocate_hqd:
+	deallocate_hqd(dqm, q);
+
+	return retval;
 }
 
 static int destroy_queue_nocpsch(struct device_queue_manager *dqm,
@@ -299,12 +295,8 @@ static int destroy_queue_nocpsch(struct device_queue_manager *dqm,
 	int retval;
 	struct mqd_manager *mqd;
 
-	BUG_ON(!dqm || !q || !q->mqd || !qpd);
-
 	retval = 0;
 
-	pr_debug("kfd: In Func %s\n", __func__);
-
 	mutex_lock(&dqm->lock);
 
 	if (q->properties.type == KFD_QUEUE_TYPE_COMPUTE) {
@@ -323,7 +315,7 @@ static int destroy_queue_nocpsch(struct device_queue_manager *dqm,
 		dqm->sdma_queue_count--;
 		deallocate_sdma_queue(dqm, q->sdma_id);
 	} else {
-		pr_debug("q->properties.type is invalid (%d)\n",
+		pr_debug("q->properties.type %d is invalid\n",
 				q->properties.type);
 		retval = -EINVAL;
 		goto out;
@@ -334,7 +326,7 @@ static int destroy_queue_nocpsch(struct device_queue_manager *dqm,
 				QUEUE_PREEMPT_DEFAULT_TIMEOUT_MS,
 				q->pipe, q->queue);
 
-	if (retval != 0)
+	if (retval)
 		goto out;
 
 	mqd->uninit_mqd(mqd, q->mqd, q->mqd_mem_obj);
@@ -364,14 +356,12 @@ static int update_queue(struct device_queue_manager *dqm, struct queue *q)
 	struct mqd_manager *mqd;
 	bool prev_active = false;
 
-	BUG_ON(!dqm || !q || !q->mqd);
-
 	mutex_lock(&dqm->lock);
 	mqd = dqm->ops.get_mqd_manager(dqm,
 			get_mqd_type_from_queue_type(q->properties.type));
-	if (mqd == NULL) {
-		mutex_unlock(&dqm->lock);
-		return -ENOMEM;
+	if (!mqd) {
+		retval = -ENOMEM;
+		goto out_unlock;
 	}
 
 	if (q->properties.is_active)
@@ -385,12 +375,13 @@ static int update_queue(struct device_queue_manager *dqm, struct queue *q)
 	retval = mqd->update_mqd(mqd, q->mqd, &q->properties);
 	if ((q->properties.is_active) && (!prev_active))
 		dqm->queue_count++;
-	else if ((!q->properties.is_active) && (prev_active))
+	else if (!q->properties.is_active && prev_active)
 		dqm->queue_count--;
 
 	if (sched_policy != KFD_SCHED_POLICY_NO_HWS)
 		retval = execute_queues_cpsch(dqm, false);
 
+out_unlock:
 	mutex_unlock(&dqm->lock);
 	return retval;
 }
@@ -400,15 +391,16 @@ static struct mqd_manager *get_mqd_manager_nocpsch(
 {
 	struct mqd_manager *mqd;
 
-	BUG_ON(!dqm || type >= KFD_MQD_TYPE_MAX);
+	if (WARN_ON(type >= KFD_MQD_TYPE_MAX))
+		return NULL;
 
-	pr_debug("kfd: In func %s mqd type %d\n", __func__, type);
+	pr_debug("mqd type %d\n", type);
 
 	mqd = dqm->mqds[type];
 	if (!mqd) {
 		mqd = mqd_manager_init(type, dqm->dev);
-		if (mqd == NULL)
-			pr_err("kfd: mqd manager is NULL");
+		if (!mqd)
+			pr_err("mqd manager is NULL");
 		dqm->mqds[type] = mqd;
 	}
 
@@ -421,11 +413,7 @@ static int register_process_nocpsch(struct device_queue_manager *dqm,
 	struct device_process_node *n;
 	int retval;
 
-	BUG_ON(!dqm || !qpd);
-
-	pr_debug("kfd: In func %s\n", __func__);
-
-	n = kzalloc(sizeof(struct device_process_node), GFP_KERNEL);
+	n = kzalloc(sizeof(*n), GFP_KERNEL);
 	if (!n)
 		return -ENOMEM;
 
@@ -449,10 +437,6 @@ static int unregister_process_nocpsch(struct device_queue_manager *dqm,
 	int retval;
 	struct device_process_node *cur, *next;
 
-	BUG_ON(!dqm || !qpd);
-
-	pr_debug("In func %s\n", __func__);
-
 	pr_debug("qpd->queues_list is %s\n",
 			list_empty(&qpd->queues_list) ? "empty" : "not empty");
 
@@ -493,51 +477,39 @@ static void init_interrupts(struct device_queue_manager *dqm)
 {
 	unsigned int i;
 
-	BUG_ON(dqm == NULL);
-
 	for (i = 0 ; i < get_pipes_per_mec(dqm) ; i++)
 		if (is_pipe_enabled(dqm, 0, i))
 			dqm->dev->kfd2kgd->init_interrupts(dqm->dev->kgd, i);
 }
 
-static int init_scheduler(struct device_queue_manager *dqm)
-{
-	int retval = 0;
-
-	BUG_ON(!dqm);
-
-	pr_debug("kfd: In %s\n", __func__);
-
-	return retval;
-}
-
 static int initialize_nocpsch(struct device_queue_manager *dqm)
 {
-	int i;
+	int pipe, queue;
 
-	BUG_ON(!dqm);
+	pr_debug("num of pipes: %d\n", get_pipes_per_mec(dqm));
 
-	pr_debug("kfd: In func %s num of pipes: %d\n",
-			__func__, get_pipes_per_mec(dqm));
+	dqm->allocated_queues = kcalloc(get_pipes_per_mec(dqm),
+					sizeof(unsigned int), GFP_KERNEL);
+	if (!dqm->allocated_queues)
+		return -ENOMEM;
 
 	mutex_init(&dqm->lock);
 	INIT_LIST_HEAD(&dqm->queues);
 	dqm->queue_count = dqm->next_pipe_to_allocate = 0;
 	dqm->sdma_queue_count = 0;
-	dqm->allocated_queues = kcalloc(get_pipes_per_mec(dqm),
-					sizeof(unsigned int), GFP_KERNEL);
-	if (!dqm->allocated_queues) {
-		mutex_destroy(&dqm->lock);
-		return -ENOMEM;
-	}
 
-	for (i = 0; i < get_pipes_per_mec(dqm); i++)
-		dqm->allocated_queues[i] = (1 << get_queues_per_pipe(dqm)) - 1;
+	for (pipe = 0; pipe < get_pipes_per_mec(dqm); pipe++) {
+		int pipe_offset = pipe * get_queues_per_pipe(dqm);
+
+		for (queue = 0; queue < get_queues_per_pipe(dqm); queue++)
+			if (test_bit(pipe_offset + queue,
+				     dqm->dev->shared_resources.queue_bitmap))
+				dqm->allocated_queues[pipe] |= 1 << queue;
+	}
 
 	dqm->vmid_bitmap = (1 << VMID_PER_DEVICE) - 1;
 	dqm->sdma_bitmap = (1 << CIK_SDMA_QUEUES) - 1;
 
-	init_scheduler(dqm);
 	return 0;
 }
 
@@ -545,9 +517,7 @@ static void uninitialize_nocpsch(struct device_queue_manager *dqm)
 {
 	int i;
 
-	BUG_ON(!dqm);
-
-	BUG_ON(dqm->queue_count > 0 || dqm->processes_count > 0);
+	WARN_ON(dqm->queue_count > 0 || dqm->processes_count > 0);
 
 	kfree(dqm->allocated_queues);
 	for (i = 0 ; i < KFD_MQD_TYPE_MAX ; i++)
@@ -604,33 +574,34 @@ static int create_sdma_queue_nocpsch(struct device_queue_manager *dqm,
 		return -ENOMEM;
 
 	retval = allocate_sdma_queue(dqm, &q->sdma_id);
-	if (retval != 0)
+	if (retval)
 		return retval;
 
 	q->properties.sdma_queue_id = q->sdma_id % CIK_SDMA_QUEUES_PER_ENGINE;
 	q->properties.sdma_engine_id = q->sdma_id / CIK_SDMA_ENGINE_NUM;
 
-	pr_debug("kfd: sdma id is:    %d\n", q->sdma_id);
-	pr_debug("     sdma queue id: %d\n", q->properties.sdma_queue_id);
-	pr_debug("     sdma engine id: %d\n", q->properties.sdma_engine_id);
+	pr_debug("SDMA id is:    %d\n", q->sdma_id);
+	pr_debug("SDMA queue id: %d\n", q->properties.sdma_queue_id);
+	pr_debug("SDMA engine id: %d\n", q->properties.sdma_engine_id);
 
 	dqm->ops_asic_specific.init_sdma_vm(dqm, q, qpd);
 	retval = mqd->init_mqd(mqd, &q->mqd, &q->mqd_mem_obj,
 				&q->gart_mqd_addr, &q->properties);
-	if (retval != 0) {
-		deallocate_sdma_queue(dqm, q->sdma_id);
-		return retval;
-	}
+	if (retval)
+		goto out_deallocate_sdma_queue;
 
-	retval = mqd->load_mqd(mqd, q->mqd, 0,
-				0, NULL);
-	if (retval != 0) {
-		deallocate_sdma_queue(dqm, q->sdma_id);
-		mqd->uninit_mqd(mqd, q->mqd, q->mqd_mem_obj);
-		return retval;
-	}
+	retval = mqd->load_mqd(mqd, q->mqd, 0, 0, &q->properties, NULL);
+	if (retval)
+		goto out_uninit_mqd;
 
 	return 0;
+
+out_uninit_mqd:
+	mqd->uninit_mqd(mqd, q->mqd, q->mqd_mem_obj);
+out_deallocate_sdma_queue:
+	deallocate_sdma_queue(dqm, q->sdma_id);
+
+	return retval;
 }
 
 /*
@@ -642,10 +613,6 @@ static int set_sched_resources(struct device_queue_manager *dqm)
 	int i, mec;
 	struct scheduling_resources res;
 
-	BUG_ON(!dqm);
-
-	pr_debug("kfd: In func %s\n", __func__);
-
 	res.vmid_mask = (1 << VMID_PER_DEVICE) - 1;
 	res.vmid_mask <<= KFD_VMID_START_OFFSET;
 
@@ -663,7 +630,8 @@ static int set_sched_resources(struct device_queue_manager *dqm)
 
 		/* This situation may be hit in the future if a new HW
 		 * generation exposes more than 64 queues. If so, the
-		 * definition of res.queue_mask needs updating */
+		 * definition of res.queue_mask needs updating
+		 */
 		if (WARN_ON(i >= (sizeof(res.queue_mask)*8))) {
 			pr_err("Invalid queue enabled by amdgpu: %d\n", i);
 			break;
@@ -674,9 +642,9 @@ static int set_sched_resources(struct device_queue_manager *dqm)
 	res.gws_mask = res.oac_mask = res.gds_heap_base =
 						res.gds_heap_size = 0;
 
-	pr_debug("kfd: scheduling resources:\n"
-			"      vmid mask: 0x%8X\n"
-			"      queue mask: 0x%8llX\n",
+	pr_debug("Scheduling resources:\n"
+			"vmid mask: 0x%8X\n"
+			"queue mask: 0x%8llX\n",
 			res.vmid_mask, res.queue_mask);
 
 	return pm_send_set_resources(&dqm->packets, &res);
@@ -686,10 +654,7 @@ static int initialize_cpsch(struct device_queue_manager *dqm)
 {
 	int retval;
 
-	BUG_ON(!dqm);
-
-	pr_debug("kfd: In func %s num of pipes: %d\n",
-			__func__, get_pipes_per_mec(dqm));
+	pr_debug("num of pipes: %d\n", get_pipes_per_mec(dqm));
 
 	mutex_init(&dqm->lock);
 	INIT_LIST_HEAD(&dqm->queues);
@@ -697,13 +662,9 @@ static int initialize_cpsch(struct device_queue_manager *dqm)
 	dqm->sdma_queue_count = 0;
 	dqm->active_runlist = false;
 	retval = dqm->ops_asic_specific.initialize(dqm);
-	if (retval != 0)
-		goto fail_init_pipelines;
-
-	return 0;
+	if (retval)
+		mutex_destroy(&dqm->lock);
 
-fail_init_pipelines:
-	mutex_destroy(&dqm->lock);
 	return retval;
 }
 
@@ -712,25 +673,23 @@ static int start_cpsch(struct device_queue_manager *dqm)
 	struct device_process_node *node;
 	int retval;
 
-	BUG_ON(!dqm);
-
 	retval = 0;
 
 	retval = pm_init(&dqm->packets, dqm);
-	if (retval != 0)
+	if (retval)
 		goto fail_packet_manager_init;
 
 	retval = set_sched_resources(dqm);
-	if (retval != 0)
+	if (retval)
 		goto fail_set_sched_resources;
 
-	pr_debug("kfd: allocating fence memory\n");
+	pr_debug("Allocating fence memory\n");
 
 	/* allocate fence memory on the gart */
 	retval = kfd_gtt_sa_allocate(dqm->dev, sizeof(*dqm->fence_addr),
 					&dqm->fence_mem);
 
-	if (retval != 0)
+	if (retval)
 		goto fail_allocate_vidmem;
 
 	dqm->fence_addr = dqm->fence_mem->cpu_ptr;
@@ -758,8 +717,6 @@ static int stop_cpsch(struct device_queue_manager *dqm)
 	struct device_process_node *node;
 	struct kfd_process_device *pdd;
 
-	BUG_ON(!dqm);
-
 	destroy_queues_cpsch(dqm, true, true);
 
 	list_for_each_entry(node, &dqm->queues, list) {
@@ -776,13 +733,9 @@ static int create_kernel_queue_cpsch(struct device_queue_manager *dqm,
 					struct kernel_queue *kq,
 					struct qcm_process_device *qpd)
 {
-	BUG_ON(!dqm || !kq || !qpd);
-
-	pr_debug("kfd: In func %s\n", __func__);
-
 	mutex_lock(&dqm->lock);
 	if (dqm->total_queue_count >= max_num_of_queues_per_device) {
-		pr_warn("amdkfd: Can't create new kernel queue because %d queues were already created\n",
+		pr_warn("Can't create new kernel queue because %d queues were already created\n",
 				dqm->total_queue_count);
 		mutex_unlock(&dqm->lock);
 		return -EPERM;
@@ -809,10 +762,6 @@ static void destroy_kernel_queue_cpsch(struct device_queue_manager *dqm,
 					struct kernel_queue *kq,
 					struct qcm_process_device *qpd)
 {
-	BUG_ON(!dqm || !kq);
-
-	pr_debug("kfd: In %s\n", __func__);
-
 	mutex_lock(&dqm->lock);
 	/* here we actually preempt the DIQ */
 	destroy_queues_cpsch(dqm, true, false);
@@ -844,8 +793,6 @@ static int create_queue_cpsch(struct device_queue_manager *dqm, struct queue *q,
 	int retval;
 	struct mqd_manager *mqd;
 
-	BUG_ON(!dqm || !q || !qpd);
-
 	retval = 0;
 
 	if (allocate_vmid)
@@ -854,7 +801,7 @@ static int create_queue_cpsch(struct device_queue_manager *dqm, struct queue *q,
 	mutex_lock(&dqm->lock);
 
 	if (dqm->total_queue_count >= max_num_of_queues_per_device) {
-		pr_warn("amdkfd: Can't create new usermode queue because %d queues were already created\n",
+		pr_warn("Can't create new usermode queue because %d queues were already created\n",
 				dqm->total_queue_count);
 		retval = -EPERM;
 		goto out;
@@ -866,15 +813,15 @@ static int create_queue_cpsch(struct device_queue_manager *dqm, struct queue *q,
 	mqd = dqm->ops.get_mqd_manager(dqm,
 			get_mqd_type_from_queue_type(q->properties.type));
 
-	if (mqd == NULL) {
-		mutex_unlock(&dqm->lock);
-		return -ENOMEM;
+	if (!mqd) {
+		retval = -ENOMEM;
+		goto out;
 	}
 
 	dqm->ops_asic_specific.init_sdma_vm(dqm, q, qpd);
 	retval = mqd->init_mqd(mqd, &q->mqd, &q->mqd_mem_obj,
 				&q->gart_mqd_addr, &q->properties);
-	if (retval != 0)
+	if (retval)
 		goto out;
 
 	list_add(&q->list, &qpd->queues_list);
@@ -884,7 +831,7 @@ static int create_queue_cpsch(struct device_queue_manager *dqm, struct queue *q,
 	}
 
 	if (q->properties.type == KFD_QUEUE_TYPE_SDMA)
-			dqm->sdma_queue_count++;
+		dqm->sdma_queue_count++;
 	/*
 	 * Unconditionally increment this counter, regardless of the queue's
 	 * type or whether the queue is active.
@@ -903,12 +850,11 @@ int amdkfd_fence_wait_timeout(unsigned int *fence_addr,
 				unsigned int fence_value,
 				unsigned long timeout)
 {
-	BUG_ON(!fence_addr);
 	timeout += jiffies;
 
 	while (*fence_addr != fence_value) {
 		if (time_after(jiffies, timeout)) {
-			pr_err("kfd: qcm fence wait loop timeout expired\n");
+			pr_err("qcm fence wait loop timeout expired\n");
 			return -ETIME;
 		}
 		schedule();
@@ -932,8 +878,6 @@ static int destroy_queues_cpsch(struct device_queue_manager *dqm,
 	enum kfd_preempt_type_filter preempt_type;
 	struct kfd_process_device *pdd;
 
-	BUG_ON(!dqm);
-
 	retval = 0;
 
 	if (lock)
@@ -941,7 +885,7 @@ static int destroy_queues_cpsch(struct device_queue_manager *dqm,
 	if (!dqm->active_runlist)
 		goto out;
 
-	pr_debug("kfd: Before destroying queues, sdma queue count is : %u\n",
+	pr_debug("Before destroying queues, sdma queue count is : %u\n",
 		dqm->sdma_queue_count);
 
 	if (dqm->sdma_queue_count > 0) {
@@ -955,7 +899,7 @@ static int destroy_queues_cpsch(struct device_queue_manager *dqm,
 
 	retval = pm_send_unmap_queue(&dqm->packets, KFD_QUEUE_TYPE_COMPUTE,
 			preempt_type, 0, false, 0);
-	if (retval != 0)
+	if (retval)
 		goto out;
 
 	*dqm->fence_addr = KFD_FENCE_INIT;
@@ -964,7 +908,7 @@ static int destroy_queues_cpsch(struct device_queue_manager *dqm,
 	/* should be timed out */
 	retval = amdkfd_fence_wait_timeout(dqm->fence_addr, KFD_FENCE_COMPLETED,
 				QUEUE_PREEMPT_DEFAULT_TIMEOUT_MS);
-	if (retval != 0) {
+	if (retval) {
 		pdd = kfd_get_process_device_data(dqm->dev,
 				kfd_get_process(current));
 		pdd->reset_wavefronts = true;
@@ -983,14 +927,12 @@ static int execute_queues_cpsch(struct device_queue_manager *dqm, bool lock)
 {
 	int retval;
 
-	BUG_ON(!dqm);
-
 	if (lock)
 		mutex_lock(&dqm->lock);
 
 	retval = destroy_queues_cpsch(dqm, false, false);
-	if (retval != 0) {
-		pr_err("kfd: the cp might be in an unrecoverable state due to an unsuccessful queues preemption");
+	if (retval) {
+		pr_err("The cp might be in an unrecoverable state due to an unsuccessful queues preemption");
 		goto out;
 	}
 
@@ -1005,8 +947,8 @@ static int execute_queues_cpsch(struct device_queue_manager *dqm, bool lock)
 	}
 
 	retval = pm_send_runlist(&dqm->packets, &dqm->queues);
-	if (retval != 0) {
-		pr_err("kfd: failed to execute runlist");
+	if (retval) {
+		pr_err("failed to execute runlist");
 		goto out;
 	}
 	dqm->active_runlist = true;
@@ -1025,8 +967,6 @@ static int destroy_queue_cpsch(struct device_queue_manager *dqm,
 	struct mqd_manager *mqd;
 	bool preempt_all_queues;
 
-	BUG_ON(!dqm || !qpd || !q);
-
 	preempt_all_queues = false;
 
 	retval = 0;
@@ -1098,8 +1038,6 @@ static bool set_cache_memory_policy(struct device_queue_manager *dqm,
 {
 	bool retval;
 
-	pr_debug("kfd: In func %s\n", __func__);
-
 	mutex_lock(&dqm->lock);
 
 	if (alternate_aperture_size == 0) {
@@ -1120,14 +1058,11 @@ static bool set_cache_memory_policy(struct device_queue_manager *dqm,
 		uint64_t base = (uintptr_t)alternate_aperture_base;
 		uint64_t limit = base + alternate_aperture_size - 1;
 
-		if (limit <= base)
-			goto out;
-
-		if ((base & APE1_FIXED_BITS_MASK) != 0)
-			goto out;
-
-		if ((limit & APE1_FIXED_BITS_MASK) != APE1_LIMIT_ALIGNMENT)
+		if (limit <= base || (base & APE1_FIXED_BITS_MASK) != 0 ||
+		   (limit & APE1_FIXED_BITS_MASK) != APE1_LIMIT_ALIGNMENT) {
+			retval = false;
 			goto out;
+		}
 
 		qpd->sh_mem_ape1_base = base >> 16;
 		qpd->sh_mem_ape1_limit = limit >> 16;
@@ -1144,27 +1079,22 @@ static bool set_cache_memory_policy(struct device_queue_manager *dqm,
 	if ((sched_policy == KFD_SCHED_POLICY_NO_HWS) && (qpd->vmid != 0))
 		program_sh_mem_settings(dqm, qpd);
 
-	pr_debug("kfd: sh_mem_config: 0x%x, ape1_base: 0x%x, ape1_limit: 0x%x\n",
+	pr_debug("sh_mem_config: 0x%x, ape1_base: 0x%x, ape1_limit: 0x%x\n",
 		qpd->sh_mem_config, qpd->sh_mem_ape1_base,
 		qpd->sh_mem_ape1_limit);
 
-	mutex_unlock(&dqm->lock);
-	return retval;
-
 out:
 	mutex_unlock(&dqm->lock);
-	return false;
+	return retval;
 }
 
 struct device_queue_manager *device_queue_manager_init(struct kfd_dev *dev)
 {
 	struct device_queue_manager *dqm;
 
-	BUG_ON(!dev);
+	pr_debug("Loading device queue manager\n");
 
-	pr_debug("kfd: loading device queue manager\n");
-
-	dqm = kzalloc(sizeof(struct device_queue_manager), GFP_KERNEL);
+	dqm = kzalloc(sizeof(*dqm), GFP_KERNEL);
 	if (!dqm)
 		return NULL;
 
@@ -1202,8 +1132,8 @@ struct device_queue_manager *device_queue_manager_init(struct kfd_dev *dev)
 		dqm->ops.set_cache_memory_policy = set_cache_memory_policy;
 		break;
 	default:
-		BUG();
-		break;
+		pr_err("Invalid scheduling policy %d\n", sched_policy);
+		goto out_free;
 	}
 
 	switch (dev->device_info->asic_family) {
@@ -1216,18 +1146,16 @@ struct device_queue_manager *device_queue_manager_init(struct kfd_dev *dev)
 		break;
 	}
 
-	if (dqm->ops.initialize(dqm) != 0) {
-		kfree(dqm);
-		return NULL;
-	}
+	if (!dqm->ops.initialize(dqm))
+		return dqm;
 
-	return dqm;
+out_free:
+	kfree(dqm);
+	return NULL;
 }
 
 void device_queue_manager_uninit(struct device_queue_manager *dqm)
 {
-	BUG_ON(!dqm);
-
 	dqm->ops.uninitialize(dqm);
 	kfree(dqm);
 }
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_cik.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_cik.c
index 48dc0561b402..72c3cbabc0a7 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_cik.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_cik.c
@@ -24,6 +24,7 @@
 #include "kfd_device_queue_manager.h"
 #include "cik_regs.h"
 #include "oss/oss_2_4_sh_mask.h"
+#include "gca/gfx_7_2_sh_mask.h"
 
 static bool set_cache_memory_policy_cik(struct device_queue_manager *dqm,
 				   struct qcm_process_device *qpd,
@@ -65,7 +66,7 @@ static uint32_t compute_sh_mem_bases_64bit(unsigned int top_address_nybble)
 	 * for LDS/Scratch and GPUVM.
 	 */
 
-	BUG_ON((top_address_nybble & 1) || top_address_nybble > 0xE ||
+	WARN_ON((top_address_nybble & 1) || top_address_nybble > 0xE ||
 		top_address_nybble == 0);
 
 	return PRIVATE_BASE(top_address_nybble << 12) |
@@ -104,8 +105,6 @@ static int register_process_cik(struct device_queue_manager *dqm,
 	struct kfd_process_device *pdd;
 	unsigned int temp;
 
-	BUG_ON(!dqm || !qpd);
-
 	pdd = qpd_to_pdd(qpd);
 
 	/* check if sh_mem_config register already configured */
@@ -125,9 +124,10 @@ static int register_process_cik(struct device_queue_manager *dqm,
 	} else {
 		temp = get_sh_mem_bases_nybble_64(pdd);
 		qpd->sh_mem_bases = compute_sh_mem_bases_64bit(temp);
+		qpd->sh_mem_config |= 1  << SH_MEM_CONFIG__PRIVATE_ATC__SHIFT;
 	}
 
-	pr_debug("kfd: is32bit process: %d sh_mem_bases nybble: 0x%X and register 0x%X\n",
+	pr_debug("is32bit process: %d sh_mem_bases nybble: 0x%X and register 0x%X\n",
 		qpd->pqm->process->is_32bit_user_mode, temp, qpd->sh_mem_bases);
 
 	return 0;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_vi.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_vi.c
index 7e9cae9d349b..40e9ddd096cd 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_vi.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_vi.c
@@ -67,7 +67,7 @@ static uint32_t compute_sh_mem_bases_64bit(unsigned int top_address_nybble)
 	 * for LDS/Scratch and GPUVM.
 	 */
 
-	BUG_ON((top_address_nybble & 1) || top_address_nybble > 0xE ||
+	WARN_ON((top_address_nybble & 1) || top_address_nybble > 0xE ||
 		top_address_nybble == 0);
 
 	return top_address_nybble << 12 |
@@ -110,8 +110,6 @@ static int register_process_vi(struct device_queue_manager *dqm,
 	struct kfd_process_device *pdd;
 	unsigned int temp;
 
-	BUG_ON(!dqm || !qpd);
-
 	pdd = qpd_to_pdd(qpd);
 
 	/* check if sh_mem_config register already configured */
@@ -137,9 +135,11 @@ static int register_process_vi(struct device_queue_manager *dqm,
 		qpd->sh_mem_bases = compute_sh_mem_bases_64bit(temp);
 		qpd->sh_mem_config |= SH_MEM_ADDRESS_MODE_HSA64 <<
 			SH_MEM_CONFIG__ADDRESS_MODE__SHIFT;
+		qpd->sh_mem_config |= 1  <<
+			SH_MEM_CONFIG__PRIVATE_ATC__SHIFT;
 	}
 
-	pr_debug("kfd: is32bit process: %d sh_mem_bases nybble: 0x%X and register 0x%X\n",
+	pr_debug("is32bit process: %d sh_mem_bases nybble: 0x%X and register 0x%X\n",
 		qpd->pqm->process->is_32bit_user_mode, temp, qpd->sh_mem_bases);
 
 	return 0;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_doorbell.c b/drivers/gpu/drm/amd/amdkfd/kfd_doorbell.c
index 453c5d66e5c3..acf4d2a977ad 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_doorbell.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_doorbell.c
@@ -59,7 +59,7 @@ static inline size_t doorbell_process_allocation(void)
 }
 
 /* Doorbell calculations for device init. */
-void kfd_doorbell_init(struct kfd_dev *kfd)
+int kfd_doorbell_init(struct kfd_dev *kfd)
 {
 	size_t doorbell_start_offset;
 	size_t doorbell_aperture_size;
@@ -95,26 +95,35 @@ void kfd_doorbell_init(struct kfd_dev *kfd)
 	kfd->doorbell_kernel_ptr = ioremap(kfd->doorbell_base,
 						doorbell_process_allocation());
 
-	BUG_ON(!kfd->doorbell_kernel_ptr);
+	if (!kfd->doorbell_kernel_ptr)
+		return -ENOMEM;
 
-	pr_debug("kfd: doorbell initialization:\n");
-	pr_debug("kfd: doorbell base           == 0x%08lX\n",
+	pr_debug("Doorbell initialization:\n");
+	pr_debug("doorbell base           == 0x%08lX\n",
 			(uintptr_t)kfd->doorbell_base);
 
-	pr_debug("kfd: doorbell_id_offset      == 0x%08lX\n",
+	pr_debug("doorbell_id_offset      == 0x%08lX\n",
 			kfd->doorbell_id_offset);
 
-	pr_debug("kfd: doorbell_process_limit  == 0x%08lX\n",
+	pr_debug("doorbell_process_limit  == 0x%08lX\n",
 			doorbell_process_limit);
 
-	pr_debug("kfd: doorbell_kernel_offset  == 0x%08lX\n",
+	pr_debug("doorbell_kernel_offset  == 0x%08lX\n",
 			(uintptr_t)kfd->doorbell_base);
 
-	pr_debug("kfd: doorbell aperture size  == 0x%08lX\n",
+	pr_debug("doorbell aperture size  == 0x%08lX\n",
 			kfd->shared_resources.doorbell_aperture_size);
 
-	pr_debug("kfd: doorbell kernel address == 0x%08lX\n",
+	pr_debug("doorbell kernel address == 0x%08lX\n",
 			(uintptr_t)kfd->doorbell_kernel_ptr);
+
+	return 0;
+}
+
+void kfd_doorbell_fini(struct kfd_dev *kfd)
+{
+	if (kfd->doorbell_kernel_ptr)
+		iounmap(kfd->doorbell_kernel_ptr);
 }
 
 int kfd_doorbell_mmap(struct kfd_process *process, struct vm_area_struct *vma)
@@ -131,7 +140,7 @@ int kfd_doorbell_mmap(struct kfd_process *process, struct vm_area_struct *vma)
 
 	/* Find kfd device according to gpu id */
 	dev = kfd_device_by_id(vma->vm_pgoff);
-	if (dev == NULL)
+	if (!dev)
 		return -EINVAL;
 
 	/* Calculate physical address of doorbell */
@@ -142,12 +151,11 @@ int kfd_doorbell_mmap(struct kfd_process *process, struct vm_area_struct *vma)
 
 	vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot);
 
-	pr_debug("kfd: mapping doorbell page in %s\n"
+	pr_debug("Mapping doorbell page\n"
 		 "     target user address == 0x%08llX\n"
 		 "     physical address    == 0x%08llX\n"
 		 "     vm_flags            == 0x%04lX\n"
 		 "     size                == 0x%04lX\n",
-		 __func__,
 		 (unsigned long long) vma->vm_start, address, vma->vm_flags,
 		 doorbell_process_allocation());
 
@@ -166,8 +174,6 @@ u32 __iomem *kfd_get_kernel_doorbell(struct kfd_dev *kfd,
 {
 	u32 inx;
 
-	BUG_ON(!kfd || !doorbell_off);
-
 	mutex_lock(&kfd->doorbell_mutex);
 	inx = find_first_zero_bit(kfd->doorbell_available_index,
 					KFD_MAX_NUM_OF_QUEUES_PER_PROCESS);
@@ -185,7 +191,7 @@ u32 __iomem *kfd_get_kernel_doorbell(struct kfd_dev *kfd,
 	*doorbell_off = KERNEL_DOORBELL_PASID * (doorbell_process_allocation() /
 							sizeof(u32)) + inx;
 
-	pr_debug("kfd: get kernel queue doorbell\n"
+	pr_debug("Get kernel queue doorbell\n"
 			 "     doorbell offset   == 0x%08X\n"
 			 "     kernel address    == 0x%08lX\n",
 		*doorbell_off, (uintptr_t)(kfd->doorbell_kernel_ptr + inx));
@@ -197,8 +203,6 @@ void kfd_release_kernel_doorbell(struct kfd_dev *kfd, u32 __iomem *db_addr)
 {
 	unsigned int inx;
 
-	BUG_ON(!kfd || !db_addr);
-
 	inx = (unsigned int)(db_addr - kfd->doorbell_kernel_ptr);
 
 	mutex_lock(&kfd->doorbell_mutex);
@@ -210,7 +214,7 @@ inline void write_kernel_doorbell(u32 __iomem *db, u32 value)
 {
 	if (db) {
 		writel(value, db);
-		pr_debug("writing %d to doorbell address 0x%p\n", value, db);
+		pr_debug("Writing %d to doorbell address 0x%p\n", value, db);
 	}
 }
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_events.c b/drivers/gpu/drm/amd/amdkfd/kfd_events.c
index d1ce83d73a87..5979158c3f7b 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_events.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_events.c
@@ -110,7 +110,7 @@ static bool allocate_free_slot(struct kfd_process *process,
 			*out_page = page;
 			*out_slot_index = slot;
 
-			pr_debug("allocated event signal slot in page %p, slot %d\n",
+			pr_debug("Allocated event signal slot in page %p, slot %d\n",
 					page, slot);
 
 			return true;
@@ -155,9 +155,9 @@ static bool allocate_signal_page(struct file *devkfd, struct kfd_process *p)
 						   struct signal_page,
 						   event_pages)->page_index + 1;
 
-	pr_debug("allocated new event signal page at %p, for process %p\n",
+	pr_debug("Allocated new event signal page at %p, for process %p\n",
 			page, p);
-	pr_debug("page index is %d\n", page->page_index);
+	pr_debug("Page index is %d\n", page->page_index);
 
 	list_add(&page->event_pages, &p->signal_event_pages);
 
@@ -194,7 +194,8 @@ static void release_event_notification_slot(struct signal_page *page,
 	page->free_slots++;
 
 	/* We don't free signal pages, they are retained by the process
-	 * and reused until it exits. */
+	 * and reused until it exits.
+	 */
 }
 
 static struct signal_page *lookup_signal_page_by_index(struct kfd_process *p,
@@ -246,7 +247,7 @@ static u32 make_nonsignal_event_id(struct kfd_process *p)
 
 	for (id = p->next_nonsignal_event_id;
 		id < KFD_LAST_NONSIGNAL_EVENT_ID &&
-		lookup_event_by_id(p, id) != NULL;
+		lookup_event_by_id(p, id);
 		id++)
 		;
 
@@ -265,7 +266,7 @@ static u32 make_nonsignal_event_id(struct kfd_process *p)
 
 	for (id = KFD_FIRST_NONSIGNAL_EVENT_ID;
 		id < KFD_LAST_NONSIGNAL_EVENT_ID &&
-		lookup_event_by_id(p, id) != NULL;
+		lookup_event_by_id(p, id);
 		id++)
 		;
 
@@ -291,13 +292,13 @@ static int create_signal_event(struct file *devkfd,
 				struct kfd_event *ev)
 {
 	if (p->signal_event_count == KFD_SIGNAL_EVENT_LIMIT) {
-		pr_warn("amdkfd: Signal event wasn't created because limit was reached\n");
+		pr_warn("Signal event wasn't created because limit was reached\n");
 		return -ENOMEM;
 	}
 
 	if (!allocate_event_notification_slot(devkfd, p, &ev->signal_page,
 						&ev->signal_slot_index)) {
-		pr_warn("amdkfd: Signal event wasn't created because out of kernel memory\n");
+		pr_warn("Signal event wasn't created because out of kernel memory\n");
 		return -ENOMEM;
 	}
 
@@ -309,11 +310,7 @@ static int create_signal_event(struct file *devkfd,
 	ev->event_id = make_signal_event_id(ev->signal_page,
 						ev->signal_slot_index);
 
-	pr_debug("signal event number %zu created with id %d, address %p\n",
-			p->signal_event_count, ev->event_id,
-			ev->user_signal_address);
-
-	pr_debug("signal event number %zu created with id %d, address %p\n",
+	pr_debug("Signal event number %zu created with id %d, address %p\n",
 			p->signal_event_count, ev->event_id,
 			ev->user_signal_address);
 
@@ -345,7 +342,7 @@ void kfd_event_init_process(struct kfd_process *p)
 
 static void destroy_event(struct kfd_process *p, struct kfd_event *ev)
 {
-	if (ev->signal_page != NULL) {
+	if (ev->signal_page) {
 		release_event_notification_slot(ev->signal_page,
 						ev->signal_slot_index);
 		p->signal_event_count--;
@@ -584,7 +581,7 @@ void kfd_signal_event_interrupt(unsigned int pasid, uint32_t partial_id,
 		 * search faster.
 		 */
 		struct signal_page *page;
-		unsigned i;
+		unsigned int i;
 
 		list_for_each_entry(page, &p->signal_event_pages, event_pages)
 			for (i = 0; i < SLOTS_PER_PAGE; i++)
@@ -816,7 +813,7 @@ int kfd_event_mmap(struct kfd_process *p, struct vm_area_struct *vma)
 	/* check required size is logical */
 	if (get_order(KFD_SIGNAL_EVENT_LIMIT * 8) !=
 			get_order(vma->vm_end - vma->vm_start)) {
-		pr_err("amdkfd: event page mmap requested illegal size\n");
+		pr_err("Event page mmap requested illegal size\n");
 		return -EINVAL;
 	}
 
@@ -825,7 +822,7 @@ int kfd_event_mmap(struct kfd_process *p, struct vm_area_struct *vma)
 	page = lookup_signal_page_by_index(p, page_index);
 	if (!page) {
 		/* Probably KFD bug, but mmap is user-accessible. */
-		pr_debug("signal page could not be found for page_index %u\n",
+		pr_debug("Signal page could not be found for page_index %u\n",
 				page_index);
 		return -EINVAL;
 	}
@@ -836,7 +833,7 @@ int kfd_event_mmap(struct kfd_process *p, struct vm_area_struct *vma)
 	vma->vm_flags |= VM_IO | VM_DONTCOPY | VM_DONTEXPAND | VM_NORESERVE
 		       | VM_DONTDUMP | VM_PFNMAP;
 
-	pr_debug("mapping signal page\n");
+	pr_debug("Mapping signal page\n");
 	pr_debug("     start user address  == 0x%08lx\n", vma->vm_start);
 	pr_debug("     end user address    == 0x%08lx\n", vma->vm_end);
 	pr_debug("     pfn                 == 0x%016lX\n", pfn);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c b/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c
index 2b655103ba79..c59384bbbc5f 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c
@@ -304,7 +304,7 @@ int kfd_init_apertures(struct kfd_process *process)
 		id < NUM_OF_SUPPORTED_GPUS) {
 
 		pdd = kfd_create_process_device_data(dev, process);
-		if (pdd == NULL) {
+		if (!pdd) {
 			pr_err("Failed to create process device data\n");
 			return -1;
 		}
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_interrupt.c b/drivers/gpu/drm/amd/amdkfd/kfd_interrupt.c
index 7f134aa9bfd3..70b3a99cffc2 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_interrupt.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_interrupt.c
@@ -179,7 +179,7 @@ static void interrupt_wq(struct work_struct *work)
 bool interrupt_is_wanted(struct kfd_dev *dev, const uint32_t *ih_ring_entry)
 {
 	/* integer and bitwise OR so there is no boolean short-circuiting */
-	unsigned wanted = 0;
+	unsigned int wanted = 0;
 
 	wanted |= dev->device_info->event_interrupt_class->interrupt_isr(dev,
 								ih_ring_entry);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c b/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c
index d135cd002a95..681b639f5133 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c
@@ -41,11 +41,11 @@ static bool initialize(struct kernel_queue *kq, struct kfd_dev *dev,
 	int retval;
 	union PM4_MES_TYPE_3_HEADER nop;
 
-	BUG_ON(!kq || !dev);
-	BUG_ON(type != KFD_QUEUE_TYPE_DIQ && type != KFD_QUEUE_TYPE_HIQ);
+	if (WARN_ON(type != KFD_QUEUE_TYPE_DIQ && type != KFD_QUEUE_TYPE_HIQ))
+		return false;
 
-	pr_debug("amdkfd: In func %s initializing queue type %d size %d\n",
-			__func__, KFD_QUEUE_TYPE_HIQ, queue_size);
+	pr_debug("Initializing queue type %d size %d\n", KFD_QUEUE_TYPE_HIQ,
+			queue_size);
 
 	memset(&prop, 0, sizeof(prop));
 	memset(&nop, 0, sizeof(nop));
@@ -63,23 +63,23 @@ static bool initialize(struct kernel_queue *kq, struct kfd_dev *dev,
 						KFD_MQD_TYPE_HIQ);
 		break;
 	default:
-		BUG();
-		break;
+		pr_err("Invalid queue type %d\n", type);
+		return false;
 	}
 
-	if (kq->mqd == NULL)
+	if (!kq->mqd)
 		return false;
 
 	prop.doorbell_ptr = kfd_get_kernel_doorbell(dev, &prop.doorbell_off);
 
-	if (prop.doorbell_ptr == NULL) {
-		pr_err("amdkfd: error init doorbell");
+	if (!prop.doorbell_ptr) {
+		pr_err("Failed to initialize doorbell");
 		goto err_get_kernel_doorbell;
 	}
 
 	retval = kfd_gtt_sa_allocate(dev, queue_size, &kq->pq);
 	if (retval != 0) {
-		pr_err("amdkfd: error init pq queues size (%d)\n", queue_size);
+		pr_err("Failed to init pq queues size %d\n", queue_size);
 		goto err_pq_allocate_vidmem;
 	}
 
@@ -87,7 +87,7 @@ static bool initialize(struct kernel_queue *kq, struct kfd_dev *dev,
 	kq->pq_gpu_addr = kq->pq->gpu_addr;
 
 	retval = kq->ops_asic_specific.initialize(kq, dev, type, queue_size);
-	if (retval == false)
+	if (!retval)
 		goto err_eop_allocate_vidmem;
 
 	retval = kfd_gtt_sa_allocate(dev, sizeof(*kq->rptr_kernel),
@@ -139,11 +139,12 @@ static bool initialize(struct kernel_queue *kq, struct kfd_dev *dev,
 
 	/* assign HIQ to HQD */
 	if (type == KFD_QUEUE_TYPE_HIQ) {
-		pr_debug("assigning hiq to hqd\n");
+		pr_debug("Assigning hiq to hqd\n");
 		kq->queue->pipe = KFD_CIK_HIQ_PIPE;
 		kq->queue->queue = KFD_CIK_HIQ_QUEUE;
 		kq->mqd->load_mqd(kq->mqd, kq->queue->mqd, kq->queue->pipe,
-					kq->queue->queue, NULL);
+				  kq->queue->queue, &kq->queue->properties,
+				  NULL);
 	} else {
 		/* allocate fence for DIQ */
 
@@ -180,8 +181,6 @@ err_get_kernel_doorbell:
 
 static void uninitialize(struct kernel_queue *kq)
 {
-	BUG_ON(!kq);
-
 	if (kq->queue->properties.type == KFD_QUEUE_TYPE_HIQ)
 		kq->mqd->destroy_mqd(kq->mqd,
 					NULL,
@@ -211,8 +210,6 @@ static int acquire_packet_buffer(struct kernel_queue *kq,
 	uint32_t wptr, rptr;
 	unsigned int *queue_address;
 
-	BUG_ON(!kq || !buffer_ptr);
-
 	rptr = *kq->rptr_kernel;
 	wptr = *kq->wptr_kernel;
 	queue_address = (unsigned int *)kq->pq_kernel_addr;
@@ -252,11 +249,7 @@ static void submit_packet(struct kernel_queue *kq)
 {
 #ifdef DEBUG
 	int i;
-#endif
-
-	BUG_ON(!kq);
 
-#ifdef DEBUG
 	for (i = *kq->wptr_kernel; i < kq->pending_wptr; i++) {
 		pr_debug("0x%2X ", kq->pq_kernel_addr[i]);
 		if (i % 15 == 0)
@@ -272,7 +265,6 @@ static void submit_packet(struct kernel_queue *kq)
 
 static void rollback_packet(struct kernel_queue *kq)
 {
-	BUG_ON(!kq);
 	kq->pending_wptr = *kq->queue->properties.write_ptr;
 }
 
@@ -281,9 +273,7 @@ struct kernel_queue *kernel_queue_init(struct kfd_dev *dev,
 {
 	struct kernel_queue *kq;
 
-	BUG_ON(!dev);
-
-	kq = kzalloc(sizeof(struct kernel_queue), GFP_KERNEL);
+	kq = kzalloc(sizeof(*kq), GFP_KERNEL);
 	if (!kq)
 		return NULL;
 
@@ -304,7 +294,7 @@ struct kernel_queue *kernel_queue_init(struct kfd_dev *dev,
 	}
 
 	if (!kq->ops.initialize(kq, dev, type, KFD_KERNEL_QUEUE_SIZE)) {
-		pr_err("amdkfd: failed to init kernel queue\n");
+		pr_err("Failed to init kernel queue\n");
 		kfree(kq);
 		return NULL;
 	}
@@ -313,32 +303,37 @@ struct kernel_queue *kernel_queue_init(struct kfd_dev *dev,
 
 void kernel_queue_uninit(struct kernel_queue *kq)
 {
-	BUG_ON(!kq);
-
 	kq->ops.uninitialize(kq);
 	kfree(kq);
 }
 
+/* FIXME: Can this test be removed? */
 static __attribute__((unused)) void test_kq(struct kfd_dev *dev)
 {
 	struct kernel_queue *kq;
 	uint32_t *buffer, i;
 	int retval;
 
-	BUG_ON(!dev);
-
-	pr_err("amdkfd: starting kernel queue test\n");
+	pr_err("Starting kernel queue test\n");
 
 	kq = kernel_queue_init(dev, KFD_QUEUE_TYPE_HIQ);
-	BUG_ON(!kq);
+	if (unlikely(!kq)) {
+		pr_err("  Failed to initialize HIQ\n");
+		pr_err("Kernel queue test failed\n");
+		return;
+	}
 
 	retval = kq->ops.acquire_packet_buffer(kq, 5, &buffer);
-	BUG_ON(retval != 0);
+	if (unlikely(retval != 0)) {
+		pr_err("  Failed to acquire packet buffer\n");
+		pr_err("Kernel queue test failed\n");
+		return;
+	}
 	for (i = 0; i < 5; i++)
 		buffer[i] = kq->nop_packet;
 	kq->ops.submit_packet(kq);
 
-	pr_err("amdkfd: ending kernel queue test\n");
+	pr_err("Ending kernel queue test\n");
 }
 
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_module.c b/drivers/gpu/drm/amd/amdkfd/kfd_module.c
index 850a5623661f..0d73bea22c45 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_module.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_module.c
@@ -61,7 +61,8 @@ MODULE_PARM_DESC(send_sigterm,
 
 static int amdkfd_init_completed;
 
-int kgd2kfd_init(unsigned interface_version, const struct kgd2kfd_calls **g2f)
+int kgd2kfd_init(unsigned int interface_version,
+		const struct kgd2kfd_calls **g2f)
 {
 	if (!amdkfd_init_completed)
 		return -EPROBE_DEFER;
@@ -90,7 +91,7 @@ static int __init kfd_module_init(void)
 	/* Verify module parameters */
 	if ((sched_policy < KFD_SCHED_POLICY_HWS) ||
 		(sched_policy > KFD_SCHED_POLICY_NO_HWS)) {
-		pr_err("kfd: sched_policy has invalid value\n");
+		pr_err("sched_policy has invalid value\n");
 		return -1;
 	}
 
@@ -98,13 +99,13 @@ static int __init kfd_module_init(void)
 	if ((max_num_of_queues_per_device < 1) ||
 		(max_num_of_queues_per_device >
 			KFD_MAX_NUM_OF_QUEUES_PER_DEVICE)) {
-		pr_err("kfd: max_num_of_queues_per_device must be between 1 to KFD_MAX_NUM_OF_QUEUES_PER_DEVICE\n");
+		pr_err("max_num_of_queues_per_device must be between 1 to KFD_MAX_NUM_OF_QUEUES_PER_DEVICE\n");
 		return -1;
 	}
 
 	err = kfd_pasid_init();
 	if (err < 0)
-		goto err_pasid;
+		return err;
 
 	err = kfd_chardev_init();
 	if (err < 0)
@@ -126,7 +127,6 @@ err_topology:
 	kfd_chardev_exit();
 err_ioctl:
 	kfd_pasid_exit();
-err_pasid:
 	return err;
 }
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.h b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.h
index 213a71e0b6c7..1f3a6ba7eed2 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.h
@@ -67,7 +67,8 @@ struct mqd_manager {
 
 	int	(*load_mqd)(struct mqd_manager *mm, void *mqd,
 				uint32_t pipe_id, uint32_t queue_id,
-				uint32_t __user *wptr);
+				struct queue_properties *p,
+				struct mm_struct *mms);
 
 	int	(*update_mqd)(struct mqd_manager *mm, void *mqd,
 				struct queue_properties *q);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_cik.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_cik.c
index 6acc4313363e..44ffd23348fc 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_cik.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_cik.c
@@ -44,10 +44,6 @@ static int init_mqd(struct mqd_manager *mm, void **mqd,
 	struct cik_mqd *m;
 	int retval;
 
-	BUG_ON(!mm || !q || !mqd);
-
-	pr_debug("kfd: In func %s\n", __func__);
-
 	retval = kfd_gtt_sa_allocate(mm->dev, sizeof(struct cik_mqd),
 					mqd_mem_obj);
 
@@ -101,7 +97,7 @@ static int init_mqd(struct mqd_manager *mm, void **mqd,
 		m->cp_hqd_iq_rptr = AQL_ENABLE;
 
 	*mqd = m;
-	if (gart_addr != NULL)
+	if (gart_addr)
 		*gart_addr = addr;
 	retval = mm->update_mqd(mm, m, q);
 
@@ -115,8 +111,6 @@ static int init_mqd_sdma(struct mqd_manager *mm, void **mqd,
 	int retval;
 	struct cik_sdma_rlc_registers *m;
 
-	BUG_ON(!mm || !mqd || !mqd_mem_obj);
-
 	retval = kfd_gtt_sa_allocate(mm->dev,
 					sizeof(struct cik_sdma_rlc_registers),
 					mqd_mem_obj);
@@ -129,7 +123,7 @@ static int init_mqd_sdma(struct mqd_manager *mm, void **mqd,
 	memset(m, 0, sizeof(struct cik_sdma_rlc_registers));
 
 	*mqd = m;
-	if (gart_addr != NULL)
+	if (gart_addr)
 		*gart_addr = (*mqd_mem_obj)->gpu_addr;
 
 	retval = mm->update_mqd(mm, m, q);
@@ -140,27 +134,31 @@ static int init_mqd_sdma(struct mqd_manager *mm, void **mqd,
 static void uninit_mqd(struct mqd_manager *mm, void *mqd,
 			struct kfd_mem_obj *mqd_mem_obj)
 {
-	BUG_ON(!mm || !mqd);
 	kfd_gtt_sa_free(mm->dev, mqd_mem_obj);
 }
 
 static void uninit_mqd_sdma(struct mqd_manager *mm, void *mqd,
 				struct kfd_mem_obj *mqd_mem_obj)
 {
-	BUG_ON(!mm || !mqd);
 	kfd_gtt_sa_free(mm->dev, mqd_mem_obj);
 }
 
 static int load_mqd(struct mqd_manager *mm, void *mqd, uint32_t pipe_id,
-			uint32_t queue_id, uint32_t __user *wptr)
+		    uint32_t queue_id, struct queue_properties *p,
+		    struct mm_struct *mms)
 {
-	return mm->dev->kfd2kgd->hqd_load
-		(mm->dev->kgd, mqd, pipe_id, queue_id, wptr);
+	/* AQL write pointer counts in 64B packets, PM4/CP counts in dwords. */
+	uint32_t wptr_shift = (p->format == KFD_QUEUE_FORMAT_AQL ? 4 : 0);
+	uint32_t wptr_mask = (uint32_t)((p->queue_size / sizeof(uint32_t)) - 1);
+
+	return mm->dev->kfd2kgd->hqd_load(mm->dev->kgd, mqd, pipe_id, queue_id,
+					  (uint32_t __user *)p->write_ptr,
+					  wptr_shift, wptr_mask, mms);
 }
 
 static int load_mqd_sdma(struct mqd_manager *mm, void *mqd,
-			uint32_t pipe_id, uint32_t queue_id,
-			uint32_t __user *wptr)
+			 uint32_t pipe_id, uint32_t queue_id,
+			 struct queue_properties *p, struct mm_struct *mms)
 {
 	return mm->dev->kfd2kgd->hqd_sdma_load(mm->dev->kgd, mqd);
 }
@@ -170,10 +168,6 @@ static int update_mqd(struct mqd_manager *mm, void *mqd,
 {
 	struct cik_mqd *m;
 
-	BUG_ON(!mm || !q || !mqd);
-
-	pr_debug("kfd: In func %s\n", __func__);
-
 	m = get_mqd(mqd);
 	m->cp_hqd_pq_control = DEFAULT_RPTR_BLOCK_SIZE |
 				DEFAULT_MIN_AVAIL_SIZE | PQ_ATC_EN;
@@ -188,21 +182,17 @@ static int update_mqd(struct mqd_manager *mm, void *mqd,
 	m->cp_hqd_pq_base_hi = upper_32_bits((uint64_t)q->queue_address >> 8);
 	m->cp_hqd_pq_rptr_report_addr_lo = lower_32_bits((uint64_t)q->read_ptr);
 	m->cp_hqd_pq_rptr_report_addr_hi = upper_32_bits((uint64_t)q->read_ptr);
-	m->cp_hqd_pq_doorbell_control = DOORBELL_EN |
-					DOORBELL_OFFSET(q->doorbell_off);
+	m->cp_hqd_pq_doorbell_control = DOORBELL_OFFSET(q->doorbell_off);
 
 	m->cp_hqd_vmid = q->vmid;
 
-	if (q->format == KFD_QUEUE_FORMAT_AQL) {
+	if (q->format == KFD_QUEUE_FORMAT_AQL)
 		m->cp_hqd_pq_control |= NO_UPDATE_RPTR;
-	}
 
-	m->cp_hqd_active = 0;
 	q->is_active = false;
 	if (q->queue_size > 0 &&
 			q->queue_address != 0 &&
 			q->queue_percent > 0) {
-		m->cp_hqd_active = 1;
 		q->is_active = true;
 	}
 
@@ -214,8 +204,6 @@ static int update_mqd_sdma(struct mqd_manager *mm, void *mqd,
 {
 	struct cik_sdma_rlc_registers *m;
 
-	BUG_ON(!mm || !mqd || !q);
-
 	m = get_sdma_mqd(mqd);
 	m->sdma_rlc_rb_cntl = ffs(q->queue_size / sizeof(unsigned int)) <<
 			SDMA0_RLC0_RB_CNTL__RB_SIZE__SHIFT |
@@ -254,7 +242,7 @@ static int destroy_mqd(struct mqd_manager *mm, void *mqd,
 			unsigned int timeout, uint32_t pipe_id,
 			uint32_t queue_id)
 {
-	return mm->dev->kfd2kgd->hqd_destroy(mm->dev->kgd, type, timeout,
+	return mm->dev->kfd2kgd->hqd_destroy(mm->dev->kgd, mqd, type, timeout,
 					pipe_id, queue_id);
 }
 
@@ -301,10 +289,6 @@ static int init_mqd_hiq(struct mqd_manager *mm, void **mqd,
 	struct cik_mqd *m;
 	int retval;
 
-	BUG_ON(!mm || !q || !mqd || !mqd_mem_obj);
-
-	pr_debug("kfd: In func %s\n", __func__);
-
 	retval = kfd_gtt_sa_allocate(mm->dev, sizeof(struct cik_mqd),
 					mqd_mem_obj);
 
@@ -359,10 +343,6 @@ static int update_mqd_hiq(struct mqd_manager *mm, void *mqd,
 {
 	struct cik_mqd *m;
 
-	BUG_ON(!mm || !q || !mqd);
-
-	pr_debug("kfd: In func %s\n", __func__);
-
 	m = get_mqd(mqd);
 	m->cp_hqd_pq_control = DEFAULT_RPTR_BLOCK_SIZE |
 				DEFAULT_MIN_AVAIL_SIZE |
@@ -400,8 +380,6 @@ struct cik_sdma_rlc_registers *get_sdma_mqd(void *mqd)
 {
 	struct cik_sdma_rlc_registers *m;
 
-	BUG_ON(!mqd);
-
 	m = (struct cik_sdma_rlc_registers *)mqd;
 
 	return m;
@@ -412,12 +390,10 @@ struct mqd_manager *mqd_manager_init_cik(enum KFD_MQD_TYPE type,
 {
 	struct mqd_manager *mqd;
 
-	BUG_ON(!dev);
-	BUG_ON(type >= KFD_MQD_TYPE_MAX);
-
-	pr_debug("kfd: In func %s\n", __func__);
+	if (WARN_ON(type >= KFD_MQD_TYPE_MAX))
+		return NULL;
 
-	mqd = kzalloc(sizeof(struct mqd_manager), GFP_KERNEL);
+	mqd = kzalloc(sizeof(*mqd), GFP_KERNEL);
 	if (!mqd)
 		return NULL;
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_vi.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_vi.c
index a9b9882a9a77..73cbfe186dd2 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_vi.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_vi.c
@@ -85,7 +85,7 @@ static int init_mqd(struct mqd_manager *mm, void **mqd,
 		m->cp_hqd_iq_rptr = 1;
 
 	*mqd = m;
-	if (gart_addr != NULL)
+	if (gart_addr)
 		*gart_addr = addr;
 	retval = mm->update_mqd(mm, m, q);
 
@@ -94,10 +94,15 @@ static int init_mqd(struct mqd_manager *mm, void **mqd,
 
 static int load_mqd(struct mqd_manager *mm, void *mqd,
 			uint32_t pipe_id, uint32_t queue_id,
-			uint32_t __user *wptr)
+			struct queue_properties *p, struct mm_struct *mms)
 {
-	return mm->dev->kfd2kgd->hqd_load
-		(mm->dev->kgd, mqd, pipe_id, queue_id, wptr);
+	/* AQL write pointer counts in 64B packets, PM4/CP counts in dwords. */
+	uint32_t wptr_shift = (p->format == KFD_QUEUE_FORMAT_AQL ? 4 : 0);
+	uint32_t wptr_mask = (uint32_t)((p->queue_size / sizeof(uint32_t)) - 1);
+
+	return mm->dev->kfd2kgd->hqd_load(mm->dev->kgd, mqd, pipe_id, queue_id,
+					  (uint32_t __user *)p->write_ptr,
+					  wptr_shift, wptr_mask, mms);
 }
 
 static int __update_mqd(struct mqd_manager *mm, void *mqd,
@@ -106,10 +111,6 @@ static int __update_mqd(struct mqd_manager *mm, void *mqd,
 {
 	struct vi_mqd *m;
 
-	BUG_ON(!mm || !q || !mqd);
-
-	pr_debug("kfd: In func %s\n", __func__);
-
 	m = get_mqd(mqd);
 
 	m->cp_hqd_pq_control = 5 << CP_HQD_PQ_CONTROL__RPTR_BLOCK_SIZE__SHIFT |
@@ -117,7 +118,7 @@ static int __update_mqd(struct mqd_manager *mm, void *mqd,
 			mtype << CP_HQD_PQ_CONTROL__MTYPE__SHIFT;
 	m->cp_hqd_pq_control |=
 			ffs(q->queue_size / sizeof(unsigned int)) - 1 - 1;
-	pr_debug("kfd: cp_hqd_pq_control 0x%x\n", m->cp_hqd_pq_control);
+	pr_debug("cp_hqd_pq_control 0x%x\n", m->cp_hqd_pq_control);
 
 	m->cp_hqd_pq_base_lo = lower_32_bits((uint64_t)q->queue_address >> 8);
 	m->cp_hqd_pq_base_hi = upper_32_bits((uint64_t)q->queue_address >> 8);
@@ -126,10 +127,9 @@ static int __update_mqd(struct mqd_manager *mm, void *mqd,
 	m->cp_hqd_pq_rptr_report_addr_hi = upper_32_bits((uint64_t)q->read_ptr);
 
 	m->cp_hqd_pq_doorbell_control =
-		1 << CP_HQD_PQ_DOORBELL_CONTROL__DOORBELL_EN__SHIFT |
 		q->doorbell_off <<
 			CP_HQD_PQ_DOORBELL_CONTROL__DOORBELL_OFFSET__SHIFT;
-	pr_debug("kfd: cp_hqd_pq_doorbell_control 0x%x\n",
+	pr_debug("cp_hqd_pq_doorbell_control 0x%x\n",
 			m->cp_hqd_pq_doorbell_control);
 
 	m->cp_hqd_eop_control = atc_bit << CP_HQD_EOP_CONTROL__EOP_ATC__SHIFT |
@@ -139,8 +139,15 @@ static int __update_mqd(struct mqd_manager *mm, void *mqd,
 			3 << CP_HQD_IB_CONTROL__MIN_IB_AVAIL_SIZE__SHIFT |
 			mtype << CP_HQD_IB_CONTROL__MTYPE__SHIFT;
 
-	m->cp_hqd_eop_control |=
-		ffs(q->eop_ring_buffer_size / sizeof(unsigned int)) - 1 - 1;
+	/*
+	 * HW does not clamp this field correctly. Maximum EOP queue size
+	 * is constrained by per-SE EOP done signal count, which is 8-bit.
+	 * Limit is 0xFF EOP entries (= 0x7F8 dwords). CP will not submit
+	 * more than (EOP entry count - 1) so a queue size of 0x800 dwords
+	 * is safe, giving a maximum field value of 0xA.
+	 */
+	m->cp_hqd_eop_control |= min(0xA,
+		ffs(q->eop_ring_buffer_size / sizeof(unsigned int)) - 1 - 1);
 	m->cp_hqd_eop_base_addr_lo =
 			lower_32_bits(q->eop_ring_buffer_address >> 8);
 	m->cp_hqd_eop_base_addr_hi =
@@ -156,12 +163,10 @@ static int __update_mqd(struct mqd_manager *mm, void *mqd,
 				2 << CP_HQD_PQ_CONTROL__SLOT_BASED_WPTR__SHIFT;
 	}
 
-	m->cp_hqd_active = 0;
 	q->is_active = false;
 	if (q->queue_size > 0 &&
 			q->queue_address != 0 &&
 			q->queue_percent > 0) {
-		m->cp_hqd_active = 1;
 		q->is_active = true;
 	}
 
@@ -181,14 +186,13 @@ static int destroy_mqd(struct mqd_manager *mm, void *mqd,
 			uint32_t queue_id)
 {
 	return mm->dev->kfd2kgd->hqd_destroy
-		(mm->dev->kgd, type, timeout,
+		(mm->dev->kgd, mqd, type, timeout,
 		pipe_id, queue_id);
 }
 
 static void uninit_mqd(struct mqd_manager *mm, void *mqd,
 			struct kfd_mem_obj *mqd_mem_obj)
 {
-	BUG_ON(!mm || !mqd);
 	kfd_gtt_sa_free(mm->dev, mqd_mem_obj);
 }
 
@@ -238,12 +242,10 @@ struct mqd_manager *mqd_manager_init_vi(enum KFD_MQD_TYPE type,
 {
 	struct mqd_manager *mqd;
 
-	BUG_ON(!dev);
-	BUG_ON(type >= KFD_MQD_TYPE_MAX);
-
-	pr_debug("kfd: In func %s\n", __func__);
+	if (WARN_ON(type >= KFD_MQD_TYPE_MAX))
+		return NULL;
 
-	mqd = kzalloc(sizeof(struct mqd_manager), GFP_KERNEL);
+	mqd = kzalloc(sizeof(*mqd), GFP_KERNEL);
 	if (!mqd)
 		return NULL;
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager.c
index 7131998848d7..1d312603de9f 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager.c
@@ -26,7 +26,6 @@
 #include "kfd_device_queue_manager.h"
 #include "kfd_kernel_queue.h"
 #include "kfd_priv.h"
-#include "kfd_pm4_headers.h"
 #include "kfd_pm4_headers_vi.h"
 #include "kfd_pm4_opcodes.h"
 
@@ -35,7 +34,8 @@ static inline void inc_wptr(unsigned int *wptr, unsigned int increment_bytes,
 {
 	unsigned int temp = *wptr + increment_bytes / sizeof(uint32_t);
 
-	BUG_ON((temp * sizeof(uint32_t)) > buffer_size_bytes);
+	WARN((temp * sizeof(uint32_t)) > buffer_size_bytes,
+	     "Runlist IB overflow");
 	*wptr = temp;
 }
 
@@ -43,12 +43,12 @@ static unsigned int build_pm4_header(unsigned int opcode, size_t packet_size)
 {
 	union PM4_MES_TYPE_3_HEADER header;
 
-	header.u32all = 0;
+	header.u32All = 0;
 	header.opcode = opcode;
 	header.count = packet_size/sizeof(uint32_t) - 2;
 	header.type = PM4_TYPE_3;
 
-	return header.u32all;
+	return header.u32All;
 }
 
 static void pm_calc_rlib_size(struct packet_manager *pm,
@@ -58,8 +58,6 @@ static void pm_calc_rlib_size(struct packet_manager *pm,
 	unsigned int process_count, queue_count;
 	unsigned int map_queue_size;
 
-	BUG_ON(!pm || !rlib_size || !over_subscription);
-
 	process_count = pm->dqm->processes_count;
 	queue_count = pm->dqm->queue_count;
 
@@ -67,15 +65,12 @@ static void pm_calc_rlib_size(struct packet_manager *pm,
 	*over_subscription = false;
 	if ((process_count > 1) || queue_count > get_queues_num(pm->dqm)) {
 		*over_subscription = true;
-		pr_debug("kfd: over subscribed runlist\n");
+		pr_debug("Over subscribed runlist\n");
 	}
 
-	map_queue_size =
-		(pm->dqm->dev->device_info->asic_family == CHIP_CARRIZO) ?
-		sizeof(struct pm4_mes_map_queues) :
-		sizeof(struct pm4_map_queues);
+	map_queue_size = sizeof(struct pm4_mes_map_queues);
 	/* calculate run list ib allocation size */
-	*rlib_size = process_count * sizeof(struct pm4_map_process) +
+	*rlib_size = process_count * sizeof(struct pm4_mes_map_process) +
 		     queue_count * map_queue_size;
 
 	/*
@@ -83,9 +78,9 @@ static void pm_calc_rlib_size(struct packet_manager *pm,
 	 * when over subscription
 	 */
 	if (*over_subscription)
-		*rlib_size += sizeof(struct pm4_runlist);
+		*rlib_size += sizeof(struct pm4_mes_runlist);
 
-	pr_debug("kfd: runlist ib size %d\n", *rlib_size);
+	pr_debug("runlist ib size %d\n", *rlib_size);
 }
 
 static int pm_allocate_runlist_ib(struct packet_manager *pm,
@@ -96,17 +91,16 @@ static int pm_allocate_runlist_ib(struct packet_manager *pm,
 {
 	int retval;
 
-	BUG_ON(!pm);
-	BUG_ON(pm->allocated);
-	BUG_ON(is_over_subscription == NULL);
+	if (WARN_ON(pm->allocated))
+		return -EINVAL;
 
 	pm_calc_rlib_size(pm, rl_buffer_size, is_over_subscription);
 
 	retval = kfd_gtt_sa_allocate(pm->dqm->dev, *rl_buffer_size,
 					&pm->ib_buffer_obj);
 
-	if (retval != 0) {
-		pr_err("kfd: failed to allocate runlist IB\n");
+	if (retval) {
+		pr_err("Failed to allocate runlist IB\n");
 		return retval;
 	}
 
@@ -121,15 +115,16 @@ static int pm_allocate_runlist_ib(struct packet_manager *pm,
 static int pm_create_runlist(struct packet_manager *pm, uint32_t *buffer,
 			uint64_t ib, size_t ib_size_in_dwords, bool chain)
 {
-	struct pm4_runlist *packet;
+	struct pm4_mes_runlist *packet;
 
-	BUG_ON(!pm || !buffer || !ib);
+	if (WARN_ON(!ib))
+		return -EFAULT;
 
-	packet = (struct pm4_runlist *)buffer;
+	packet = (struct pm4_mes_runlist *)buffer;
 
-	memset(buffer, 0, sizeof(struct pm4_runlist));
-	packet->header.u32all = build_pm4_header(IT_RUN_LIST,
-						sizeof(struct pm4_runlist));
+	memset(buffer, 0, sizeof(struct pm4_mes_runlist));
+	packet->header.u32All = build_pm4_header(IT_RUN_LIST,
+						sizeof(struct pm4_mes_runlist));
 
 	packet->bitfields4.ib_size = ib_size_in_dwords;
 	packet->bitfields4.chain = chain ? 1 : 0;
@@ -144,20 +139,16 @@ static int pm_create_runlist(struct packet_manager *pm, uint32_t *buffer,
 static int pm_create_map_process(struct packet_manager *pm, uint32_t *buffer,
 				struct qcm_process_device *qpd)
 {
-	struct pm4_map_process *packet;
+	struct pm4_mes_map_process *packet;
 	struct queue *cur;
 	uint32_t num_queues;
 
-	BUG_ON(!pm || !buffer || !qpd);
-
-	packet = (struct pm4_map_process *)buffer;
-
-	pr_debug("kfd: In func %s\n", __func__);
+	packet = (struct pm4_mes_map_process *)buffer;
 
-	memset(buffer, 0, sizeof(struct pm4_map_process));
+	memset(buffer, 0, sizeof(struct pm4_mes_map_process));
 
-	packet->header.u32all = build_pm4_header(IT_MAP_PROCESS,
-					sizeof(struct pm4_map_process));
+	packet->header.u32All = build_pm4_header(IT_MAP_PROCESS,
+					sizeof(struct pm4_mes_map_process));
 	packet->bitfields2.diq_enable = (qpd->is_debug) ? 1 : 0;
 	packet->bitfields2.process_quantum = 1;
 	packet->bitfields2.pasid = qpd->pqm->process->pasid;
@@ -175,27 +166,26 @@ static int pm_create_map_process(struct packet_manager *pm, uint32_t *buffer,
 	packet->sh_mem_ape1_base = qpd->sh_mem_ape1_base;
 	packet->sh_mem_ape1_limit = qpd->sh_mem_ape1_limit;
 
+	/* TODO: scratch support */
+	packet->sh_hidden_private_base_vmid = 0;
+
 	packet->gds_addr_lo = lower_32_bits(qpd->gds_context_area);
 	packet->gds_addr_hi = upper_32_bits(qpd->gds_context_area);
 
 	return 0;
 }
 
-static int pm_create_map_queue_vi(struct packet_manager *pm, uint32_t *buffer,
+static int pm_create_map_queue(struct packet_manager *pm, uint32_t *buffer,
 		struct queue *q, bool is_static)
 {
 	struct pm4_mes_map_queues *packet;
 	bool use_static = is_static;
 
-	BUG_ON(!pm || !buffer || !q);
-
-	pr_debug("kfd: In func %s\n", __func__);
-
 	packet = (struct pm4_mes_map_queues *)buffer;
-	memset(buffer, 0, sizeof(struct pm4_map_queues));
+	memset(buffer, 0, sizeof(struct pm4_mes_map_queues));
 
-	packet->header.u32all = build_pm4_header(IT_MAP_QUEUES,
-						sizeof(struct pm4_map_queues));
+	packet->header.u32All = build_pm4_header(IT_MAP_QUEUES,
+						sizeof(struct pm4_mes_map_queues));
 	packet->bitfields2.alloc_format =
 		alloc_format__mes_map_queues__one_per_pipe_vi;
 	packet->bitfields2.num_queues = 1;
@@ -223,10 +213,8 @@ static int pm_create_map_queue_vi(struct packet_manager *pm, uint32_t *buffer,
 		use_static = false; /* no static queues under SDMA */
 		break;
 	default:
-		pr_err("kfd: in %s queue type %d\n", __func__,
-				q->properties.type);
-		BUG();
-		break;
+		WARN(1, "queue type %d", q->properties.type);
+		return -EINVAL;
 	}
 	packet->bitfields3.doorbell_offset =
 			q->properties.doorbell_off;
@@ -246,68 +234,6 @@ static int pm_create_map_queue_vi(struct packet_manager *pm, uint32_t *buffer,
 	return 0;
 }
 
-static int pm_create_map_queue(struct packet_manager *pm, uint32_t *buffer,
-				struct queue *q, bool is_static)
-{
-	struct pm4_map_queues *packet;
-	bool use_static = is_static;
-
-	BUG_ON(!pm || !buffer || !q);
-
-	pr_debug("kfd: In func %s\n", __func__);
-
-	packet = (struct pm4_map_queues *)buffer;
-	memset(buffer, 0, sizeof(struct pm4_map_queues));
-
-	packet->header.u32all = build_pm4_header(IT_MAP_QUEUES,
-						sizeof(struct pm4_map_queues));
-	packet->bitfields2.alloc_format =
-				alloc_format__mes_map_queues__one_per_pipe;
-	packet->bitfields2.num_queues = 1;
-	packet->bitfields2.queue_sel =
-		queue_sel__mes_map_queues__map_to_hws_determined_queue_slots;
-
-	packet->bitfields2.vidmem = (q->properties.is_interop) ?
-			vidmem__mes_map_queues__uses_video_memory :
-			vidmem__mes_map_queues__uses_no_video_memory;
-
-	switch (q->properties.type) {
-	case KFD_QUEUE_TYPE_COMPUTE:
-	case KFD_QUEUE_TYPE_DIQ:
-		packet->bitfields2.engine_sel =
-				engine_sel__mes_map_queues__compute;
-		break;
-	case KFD_QUEUE_TYPE_SDMA:
-		packet->bitfields2.engine_sel =
-				engine_sel__mes_map_queues__sdma0;
-		use_static = false; /* no static queues under SDMA */
-		break;
-	default:
-		BUG();
-		break;
-	}
-
-	packet->mes_map_queues_ordinals[0].bitfields3.doorbell_offset =
-			q->properties.doorbell_off;
-
-	packet->mes_map_queues_ordinals[0].bitfields3.is_static =
-			(use_static) ? 1 : 0;
-
-	packet->mes_map_queues_ordinals[0].mqd_addr_lo =
-			lower_32_bits(q->gart_mqd_addr);
-
-	packet->mes_map_queues_ordinals[0].mqd_addr_hi =
-			upper_32_bits(q->gart_mqd_addr);
-
-	packet->mes_map_queues_ordinals[0].wptr_addr_lo =
-			lower_32_bits((uint64_t)q->properties.write_ptr);
-
-	packet->mes_map_queues_ordinals[0].wptr_addr_hi =
-			upper_32_bits((uint64_t)q->properties.write_ptr);
-
-	return 0;
-}
-
 static int pm_create_runlist_ib(struct packet_manager *pm,
 				struct list_head *queues,
 				uint64_t *rl_gpu_addr,
@@ -322,19 +248,16 @@ static int pm_create_runlist_ib(struct packet_manager *pm,
 	struct kernel_queue *kq;
 	bool is_over_subscription;
 
-	BUG_ON(!pm || !queues || !rl_size_bytes || !rl_gpu_addr);
-
 	rl_wptr = retval = proccesses_mapped = 0;
 
 	retval = pm_allocate_runlist_ib(pm, &rl_buffer, rl_gpu_addr,
 				&alloc_size_bytes, &is_over_subscription);
-	if (retval != 0)
+	if (retval)
 		return retval;
 
 	*rl_size_bytes = alloc_size_bytes;
 
-	pr_debug("kfd: In func %s\n", __func__);
-	pr_debug("kfd: building runlist ib process count: %d queues count %d\n",
+	pr_debug("Building runlist ib process count: %d queues count %d\n",
 		pm->dqm->processes_count, pm->dqm->queue_count);
 
 	/* build the run list ib packet */
@@ -342,42 +265,35 @@ static int pm_create_runlist_ib(struct packet_manager *pm,
 		qpd = cur->qpd;
 		/* build map process packet */
 		if (proccesses_mapped >= pm->dqm->processes_count) {
-			pr_debug("kfd: not enough space left in runlist IB\n");
+			pr_debug("Not enough space left in runlist IB\n");
 			pm_release_ib(pm);
 			return -ENOMEM;
 		}
 
 		retval = pm_create_map_process(pm, &rl_buffer[rl_wptr], qpd);
-		if (retval != 0)
+		if (retval)
 			return retval;
 
 		proccesses_mapped++;
-		inc_wptr(&rl_wptr, sizeof(struct pm4_map_process),
+		inc_wptr(&rl_wptr, sizeof(struct pm4_mes_map_process),
 				alloc_size_bytes);
 
 		list_for_each_entry(kq, &qpd->priv_queue_list, list) {
 			if (!kq->queue->properties.is_active)
 				continue;
 
-			pr_debug("kfd: static_queue, mapping kernel q %d, is debug status %d\n",
+			pr_debug("static_queue, mapping kernel q %d, is debug status %d\n",
 				kq->queue->queue, qpd->is_debug);
 
-			if (pm->dqm->dev->device_info->asic_family ==
-					CHIP_CARRIZO)
-				retval = pm_create_map_queue_vi(pm,
-						&rl_buffer[rl_wptr],
-						kq->queue,
-						qpd->is_debug);
-			else
-				retval = pm_create_map_queue(pm,
+			retval = pm_create_map_queue(pm,
 						&rl_buffer[rl_wptr],
 						kq->queue,
 						qpd->is_debug);
-			if (retval != 0)
+			if (retval)
 				return retval;
 
 			inc_wptr(&rl_wptr,
-				sizeof(struct pm4_map_queues),
+				sizeof(struct pm4_mes_map_queues),
 				alloc_size_bytes);
 		}
 
@@ -385,51 +301,44 @@ static int pm_create_runlist_ib(struct packet_manager *pm,
 			if (!q->properties.is_active)
 				continue;
 
-			pr_debug("kfd: static_queue, mapping user queue %d, is debug status %d\n",
+			pr_debug("static_queue, mapping user queue %d, is debug status %d\n",
 				q->queue, qpd->is_debug);
 
-			if (pm->dqm->dev->device_info->asic_family ==
-					CHIP_CARRIZO)
-				retval = pm_create_map_queue_vi(pm,
-						&rl_buffer[rl_wptr],
-						q,
-						qpd->is_debug);
-			else
-				retval = pm_create_map_queue(pm,
+			retval = pm_create_map_queue(pm,
 						&rl_buffer[rl_wptr],
 						q,
 						qpd->is_debug);
 
-			if (retval != 0)
+			if (retval)
 				return retval;
 
 			inc_wptr(&rl_wptr,
-				sizeof(struct pm4_map_queues),
+				sizeof(struct pm4_mes_map_queues),
 				alloc_size_bytes);
 		}
 	}
 
-	pr_debug("kfd: finished map process and queues to runlist\n");
+	pr_debug("Finished map process and queues to runlist\n");
 
 	if (is_over_subscription)
-		pm_create_runlist(pm, &rl_buffer[rl_wptr], *rl_gpu_addr,
-				alloc_size_bytes / sizeof(uint32_t), true);
+		retval = pm_create_runlist(pm, &rl_buffer[rl_wptr],
+					*rl_gpu_addr,
+					alloc_size_bytes / sizeof(uint32_t),
+					true);
 
 	for (i = 0; i < alloc_size_bytes / sizeof(uint32_t); i++)
 		pr_debug("0x%2X ", rl_buffer[i]);
 	pr_debug("\n");
 
-	return 0;
+	return retval;
 }
 
 int pm_init(struct packet_manager *pm, struct device_queue_manager *dqm)
 {
-	BUG_ON(!dqm);
-
 	pm->dqm = dqm;
 	mutex_init(&pm->lock);
 	pm->priv_queue = kernel_queue_init(dqm->dev, KFD_QUEUE_TYPE_HIQ);
-	if (pm->priv_queue == NULL) {
+	if (!pm->priv_queue) {
 		mutex_destroy(&pm->lock);
 		return -ENOMEM;
 	}
@@ -440,8 +349,6 @@ int pm_init(struct packet_manager *pm, struct device_queue_manager *dqm)
 
 void pm_uninit(struct packet_manager *pm)
 {
-	BUG_ON(!pm);
-
 	mutex_destroy(&pm->lock);
 	kernel_queue_uninit(pm->priv_queue);
 }
@@ -449,25 +356,22 @@ void pm_uninit(struct packet_manager *pm)
 int pm_send_set_resources(struct packet_manager *pm,
 				struct scheduling_resources *res)
 {
-	struct pm4_set_resources *packet;
-
-	BUG_ON(!pm || !res);
-
-	pr_debug("kfd: In func %s\n", __func__);
+	struct pm4_mes_set_resources *packet;
+	int retval = 0;
 
 	mutex_lock(&pm->lock);
 	pm->priv_queue->ops.acquire_packet_buffer(pm->priv_queue,
 					sizeof(*packet) / sizeof(uint32_t),
-			(unsigned int **)&packet);
-	if (packet == NULL) {
-		mutex_unlock(&pm->lock);
-		pr_err("kfd: failed to allocate buffer on kernel queue\n");
-		return -ENOMEM;
+					(unsigned int **)&packet);
+	if (!packet) {
+		pr_err("Failed to allocate buffer on kernel queue\n");
+		retval = -ENOMEM;
+		goto out;
 	}
 
-	memset(packet, 0, sizeof(struct pm4_set_resources));
-	packet->header.u32all = build_pm4_header(IT_SET_RESOURCES,
-					sizeof(struct pm4_set_resources));
+	memset(packet, 0, sizeof(struct pm4_mes_set_resources));
+	packet->header.u32All = build_pm4_header(IT_SET_RESOURCES,
+					sizeof(struct pm4_mes_set_resources));
 
 	packet->bitfields2.queue_type =
 			queue_type__mes_set_resources__hsa_interface_queue_hiq;
@@ -485,9 +389,10 @@ int pm_send_set_resources(struct packet_manager *pm,
 
 	pm->priv_queue->ops.submit_packet(pm->priv_queue);
 
+out:
 	mutex_unlock(&pm->lock);
 
-	return 0;
+	return retval;
 }
 
 int pm_send_runlist(struct packet_manager *pm, struct list_head *dqm_queues)
@@ -497,26 +402,24 @@ int pm_send_runlist(struct packet_manager *pm, struct list_head *dqm_queues)
 	size_t rl_ib_size, packet_size_dwords;
 	int retval;
 
-	BUG_ON(!pm || !dqm_queues);
-
 	retval = pm_create_runlist_ib(pm, dqm_queues, &rl_gpu_ib_addr,
 					&rl_ib_size);
-	if (retval != 0)
+	if (retval)
 		goto fail_create_runlist_ib;
 
-	pr_debug("kfd: runlist IB address: 0x%llX\n", rl_gpu_ib_addr);
+	pr_debug("runlist IB address: 0x%llX\n", rl_gpu_ib_addr);
 
-	packet_size_dwords = sizeof(struct pm4_runlist) / sizeof(uint32_t);
+	packet_size_dwords = sizeof(struct pm4_mes_runlist) / sizeof(uint32_t);
 	mutex_lock(&pm->lock);
 
 	retval = pm->priv_queue->ops.acquire_packet_buffer(pm->priv_queue,
 					packet_size_dwords, &rl_buffer);
-	if (retval != 0)
+	if (retval)
 		goto fail_acquire_packet_buffer;
 
 	retval = pm_create_runlist(pm, rl_buffer, rl_gpu_ib_addr,
 					rl_ib_size / sizeof(uint32_t), false);
-	if (retval != 0)
+	if (retval)
 		goto fail_create_runlist;
 
 	pm->priv_queue->ops.submit_packet(pm->priv_queue);
@@ -530,8 +433,7 @@ fail_create_runlist:
 fail_acquire_packet_buffer:
 	mutex_unlock(&pm->lock);
 fail_create_runlist_ib:
-	if (pm->allocated)
-		pm_release_ib(pm);
+	pm_release_ib(pm);
 	return retval;
 }
 
@@ -539,20 +441,21 @@ int pm_send_query_status(struct packet_manager *pm, uint64_t fence_address,
 			uint32_t fence_value)
 {
 	int retval;
-	struct pm4_query_status *packet;
+	struct pm4_mes_query_status *packet;
 
-	BUG_ON(!pm || !fence_address);
+	if (WARN_ON(!fence_address))
+		return -EFAULT;
 
 	mutex_lock(&pm->lock);
 	retval = pm->priv_queue->ops.acquire_packet_buffer(
 			pm->priv_queue,
-			sizeof(struct pm4_query_status) / sizeof(uint32_t),
+			sizeof(struct pm4_mes_query_status) / sizeof(uint32_t),
 			(unsigned int **)&packet);
-	if (retval != 0)
+	if (retval)
 		goto fail_acquire_packet_buffer;
 
-	packet->header.u32all = build_pm4_header(IT_QUERY_STATUS,
-					sizeof(struct pm4_query_status));
+	packet->header.u32All = build_pm4_header(IT_QUERY_STATUS,
+					sizeof(struct pm4_mes_query_status));
 
 	packet->bitfields2.context_id = 0;
 	packet->bitfields2.interrupt_sel =
@@ -566,9 +469,6 @@ int pm_send_query_status(struct packet_manager *pm, uint64_t fence_address,
 	packet->data_lo = lower_32_bits((uint64_t)fence_value);
 
 	pm->priv_queue->ops.submit_packet(pm->priv_queue);
-	mutex_unlock(&pm->lock);
-
-	return 0;
 
 fail_acquire_packet_buffer:
 	mutex_unlock(&pm->lock);
@@ -582,24 +482,22 @@ int pm_send_unmap_queue(struct packet_manager *pm, enum kfd_queue_type type,
 {
 	int retval;
 	uint32_t *buffer;
-	struct pm4_unmap_queues *packet;
-
-	BUG_ON(!pm);
+	struct pm4_mes_unmap_queues *packet;
 
 	mutex_lock(&pm->lock);
 	retval = pm->priv_queue->ops.acquire_packet_buffer(
 			pm->priv_queue,
-			sizeof(struct pm4_unmap_queues) / sizeof(uint32_t),
+			sizeof(struct pm4_mes_unmap_queues) / sizeof(uint32_t),
 			&buffer);
-	if (retval != 0)
+	if (retval)
 		goto err_acquire_packet_buffer;
 
-	packet = (struct pm4_unmap_queues *)buffer;
-	memset(buffer, 0, sizeof(struct pm4_unmap_queues));
-	pr_debug("kfd: static_queue: unmapping queues: mode is %d , reset is %d , type is %d\n",
+	packet = (struct pm4_mes_unmap_queues *)buffer;
+	memset(buffer, 0, sizeof(struct pm4_mes_unmap_queues));
+	pr_debug("static_queue: unmapping queues: mode is %d , reset is %d , type is %d\n",
 		mode, reset, type);
-	packet->header.u32all = build_pm4_header(IT_UNMAP_QUEUES,
-					sizeof(struct pm4_unmap_queues));
+	packet->header.u32All = build_pm4_header(IT_UNMAP_QUEUES,
+					sizeof(struct pm4_mes_unmap_queues));
 	switch (type) {
 	case KFD_QUEUE_TYPE_COMPUTE:
 	case KFD_QUEUE_TYPE_DIQ:
@@ -611,8 +509,9 @@ int pm_send_unmap_queue(struct packet_manager *pm, enum kfd_queue_type type,
 			engine_sel__mes_unmap_queues__sdma0 + sdma_engine;
 		break;
 	default:
-		BUG();
-		break;
+		WARN(1, "queue type %d", type);
+		retval = -EINVAL;
+		goto err_invalid;
 	}
 
 	if (reset)
@@ -636,16 +535,17 @@ int pm_send_unmap_queue(struct packet_manager *pm, enum kfd_queue_type type,
 		break;
 	case KFD_PREEMPT_TYPE_FILTER_ALL_QUEUES:
 		packet->bitfields2.queue_sel =
-				queue_sel__mes_unmap_queues__perform_request_on_all_active_queues;
+				queue_sel__mes_unmap_queues__unmap_all_queues;
 		break;
 	case KFD_PREEMPT_TYPE_FILTER_DYNAMIC_QUEUES:
 		/* in this case, we do not preempt static queues */
 		packet->bitfields2.queue_sel =
-				queue_sel__mes_unmap_queues__perform_request_on_dynamic_queues_only;
+				queue_sel__mes_unmap_queues__unmap_all_non_static_queues;
 		break;
 	default:
-		BUG();
-		break;
+		WARN(1, "filter %d", mode);
+		retval = -EINVAL;
+		goto err_invalid;
 	}
 
 	pm->priv_queue->ops.submit_packet(pm->priv_queue);
@@ -653,6 +553,8 @@ int pm_send_unmap_queue(struct packet_manager *pm, enum kfd_queue_type type,
 	mutex_unlock(&pm->lock);
 	return 0;
 
+err_invalid:
+	pm->priv_queue->ops.rollback_packet(pm->priv_queue);
 err_acquire_packet_buffer:
 	mutex_unlock(&pm->lock);
 	return retval;
@@ -660,8 +562,6 @@ err_acquire_packet_buffer:
 
 void pm_release_ib(struct packet_manager *pm)
 {
-	BUG_ON(!pm);
-
 	mutex_lock(&pm->lock);
 	if (pm->allocated) {
 		kfd_gtt_sa_free(pm->dqm->dev, pm->ib_buffer_obj);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_pasid.c b/drivers/gpu/drm/amd/amdkfd/kfd_pasid.c
index 6cfe7f1f18cf..1e06de0bc673 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_pasid.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_pasid.c
@@ -32,7 +32,8 @@ int kfd_pasid_init(void)
 {
 	pasid_limit = KFD_MAX_NUM_OF_PROCESSES;
 
-	pasid_bitmap = kcalloc(BITS_TO_LONGS(pasid_limit), sizeof(long), GFP_KERNEL);
+	pasid_bitmap = kcalloc(BITS_TO_LONGS(pasid_limit), sizeof(long),
+				GFP_KERNEL);
 	if (!pasid_bitmap)
 		return -ENOMEM;
 
@@ -91,6 +92,6 @@ unsigned int kfd_pasid_alloc(void)
 
 void kfd_pasid_free(unsigned int pasid)
 {
-	BUG_ON(pasid == 0 || pasid >= pasid_limit);
-	clear_bit(pasid, pasid_bitmap);
+	if (!WARN_ON(pasid == 0 || pasid >= pasid_limit))
+		clear_bit(pasid, pasid_bitmap);
 }
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_pm4_headers.h b/drivers/gpu/drm/amd/amdkfd/kfd_pm4_headers.h
index 5b393f3e34a9..e50f73d25de6 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_pm4_headers.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_pm4_headers.h
@@ -28,112 +28,19 @@
 #define PM4_MES_HEADER_DEFINED
 union PM4_MES_TYPE_3_HEADER {
 	struct {
-		uint32_t reserved1:8;	/* < reserved */
-		uint32_t opcode:8;	/* < IT opcode */
-		uint32_t count:14;	/* < number of DWORDs - 1
-					 * in the information body.
-					 */
-		uint32_t type:2;	/* < packet identifier.
-					 * It should be 3 for type 3 packets
-					 */
+		/* reserved */
+		uint32_t reserved1:8;
+		/* IT opcode */
+		uint32_t opcode:8;
+		/* number of DWORDs - 1 in the information body */
+		uint32_t count:14;
+		/* packet identifier. It should be 3 for type 3 packets */
+		uint32_t type:2;
 	};
 	uint32_t u32all;
 };
 #endif /* PM4_MES_HEADER_DEFINED */
 
-/* --------------------MES_SET_RESOURCES-------------------- */
-
-#ifndef PM4_MES_SET_RESOURCES_DEFINED
-#define PM4_MES_SET_RESOURCES_DEFINED
-enum set_resources_queue_type_enum {
-	queue_type__mes_set_resources__kernel_interface_queue_kiq = 0,
-	queue_type__mes_set_resources__hsa_interface_queue_hiq = 1,
-	queue_type__mes_set_resources__hsa_debug_interface_queue = 4
-};
-
-struct pm4_set_resources {
-	union {
-		union PM4_MES_TYPE_3_HEADER header;	/* header */
-		uint32_t ordinal1;
-	};
-
-	union {
-		struct {
-			uint32_t vmid_mask:16;
-			uint32_t unmap_latency:8;
-			uint32_t reserved1:5;
-			enum set_resources_queue_type_enum queue_type:3;
-		} bitfields2;
-		uint32_t ordinal2;
-	};
-
-	uint32_t queue_mask_lo;
-	uint32_t queue_mask_hi;
-	uint32_t gws_mask_lo;
-	uint32_t gws_mask_hi;
-
-	union {
-		struct {
-			uint32_t oac_mask:16;
-			uint32_t reserved2:16;
-		} bitfields7;
-		uint32_t ordinal7;
-	};
-
-	union {
-		struct {
-			uint32_t gds_heap_base:6;
-			uint32_t reserved3:5;
-			uint32_t gds_heap_size:6;
-			uint32_t reserved4:15;
-		} bitfields8;
-		uint32_t ordinal8;
-	};
-
-};
-#endif
-
-/*--------------------MES_RUN_LIST-------------------- */
-
-#ifndef PM4_MES_RUN_LIST_DEFINED
-#define PM4_MES_RUN_LIST_DEFINED
-
-struct pm4_runlist {
-	union {
-		union PM4_MES_TYPE_3_HEADER header;	/* header */
-		uint32_t ordinal1;
-	};
-
-	union {
-		struct {
-			uint32_t reserved1:2;
-			uint32_t ib_base_lo:30;
-		} bitfields2;
-		uint32_t ordinal2;
-	};
-
-	union {
-		struct {
-			uint32_t ib_base_hi:16;
-			uint32_t reserved2:16;
-		} bitfields3;
-		uint32_t ordinal3;
-	};
-
-	union {
-		struct {
-			uint32_t ib_size:20;
-			uint32_t chain:1;
-			uint32_t offload_polling:1;
-			uint32_t reserved3:1;
-			uint32_t valid:1;
-			uint32_t reserved4:8;
-		} bitfields4;
-		uint32_t ordinal4;
-	};
-
-};
-#endif
 
 /*--------------------MES_MAP_PROCESS-------------------- */
 
@@ -186,217 +93,58 @@ struct pm4_map_process {
 };
 #endif
 
-/*--------------------MES_MAP_QUEUES--------------------*/
-
-#ifndef PM4_MES_MAP_QUEUES_DEFINED
-#define PM4_MES_MAP_QUEUES_DEFINED
-enum map_queues_queue_sel_enum {
-	queue_sel__mes_map_queues__map_to_specified_queue_slots = 0,
-	queue_sel__mes_map_queues__map_to_hws_determined_queue_slots = 1,
-	queue_sel__mes_map_queues__enable_process_queues = 2
-};
+#ifndef PM4_MES_MAP_PROCESS_DEFINED_KV_SCRATCH
+#define PM4_MES_MAP_PROCESS_DEFINED_KV_SCRATCH
 
-enum map_queues_vidmem_enum {
-	vidmem__mes_map_queues__uses_no_video_memory = 0,
-	vidmem__mes_map_queues__uses_video_memory = 1
-};
-
-enum map_queues_alloc_format_enum {
-	alloc_format__mes_map_queues__one_per_pipe = 0,
-	alloc_format__mes_map_queues__all_on_one_pipe = 1
-};
-
-enum map_queues_engine_sel_enum {
-	engine_sel__mes_map_queues__compute = 0,
-	engine_sel__mes_map_queues__sdma0 = 2,
-	engine_sel__mes_map_queues__sdma1 = 3
-};
-
-struct pm4_map_queues {
+struct pm4_map_process_scratch_kv {
 	union {
-		union PM4_MES_TYPE_3_HEADER header;	/* header */
-		uint32_t ordinal1;
-	};
-
-	union {
-		struct {
-			uint32_t reserved1:4;
-			enum map_queues_queue_sel_enum queue_sel:2;
-			uint32_t reserved2:2;
-			uint32_t vmid:4;
-			uint32_t reserved3:4;
-			enum map_queues_vidmem_enum vidmem:2;
-			uint32_t reserved4:6;
-			enum map_queues_alloc_format_enum alloc_format:2;
-			enum map_queues_engine_sel_enum engine_sel:3;
-			uint32_t num_queues:3;
-		} bitfields2;
-		uint32_t ordinal2;
-	};
-
-	struct {
-		union {
-			struct {
-				uint32_t is_static:1;
-				uint32_t reserved5:1;
-				uint32_t doorbell_offset:21;
-				uint32_t reserved6:3;
-				uint32_t queue:6;
-			} bitfields3;
-			uint32_t ordinal3;
-		};
-
-		uint32_t mqd_addr_lo;
-		uint32_t mqd_addr_hi;
-		uint32_t wptr_addr_lo;
-		uint32_t wptr_addr_hi;
-
-	} mes_map_queues_ordinals[1];	/* 1..N of these ordinal groups */
-
-};
-#endif
-
-/*--------------------MES_QUERY_STATUS--------------------*/
-
-#ifndef PM4_MES_QUERY_STATUS_DEFINED
-#define PM4_MES_QUERY_STATUS_DEFINED
-enum query_status_interrupt_sel_enum {
-	interrupt_sel__mes_query_status__completion_status = 0,
-	interrupt_sel__mes_query_status__process_status = 1,
-	interrupt_sel__mes_query_status__queue_status = 2
-};
-
-enum query_status_command_enum {
-	command__mes_query_status__interrupt_only = 0,
-	command__mes_query_status__fence_only_immediate = 1,
-	command__mes_query_status__fence_only_after_write_ack = 2,
-	command__mes_query_status__fence_wait_for_write_ack_send_interrupt = 3
-};
-
-enum query_status_engine_sel_enum {
-	engine_sel__mes_query_status__compute = 0,
-	engine_sel__mes_query_status__sdma0_queue = 2,
-	engine_sel__mes_query_status__sdma1_queue = 3
-};
-
-struct pm4_query_status {
-	union {
-		union PM4_MES_TYPE_3_HEADER header;	/* header */
-		uint32_t ordinal1;
-	};
-
-	union {
-		struct {
-			uint32_t context_id:28;
-			enum query_status_interrupt_sel_enum interrupt_sel:2;
-			enum query_status_command_enum command:2;
-		} bitfields2;
-		uint32_t ordinal2;
+		union PM4_MES_TYPE_3_HEADER   header; /* header */
+		uint32_t            ordinal1;
 	};
 
 	union {
 		struct {
 			uint32_t pasid:16;
-			uint32_t reserved1:16;
-		} bitfields3a;
-		struct {
-			uint32_t reserved2:2;
-			uint32_t doorbell_offset:21;
-			uint32_t reserved3:3;
-			enum query_status_engine_sel_enum engine_sel:3;
-			uint32_t reserved4:3;
-		} bitfields3b;
-		uint32_t ordinal3;
-	};
-
-	uint32_t addr_lo;
-	uint32_t addr_hi;
-	uint32_t data_lo;
-	uint32_t data_hi;
-};
-#endif
-
-/*--------------------MES_UNMAP_QUEUES--------------------*/
-
-#ifndef PM4_MES_UNMAP_QUEUES_DEFINED
-#define PM4_MES_UNMAP_QUEUES_DEFINED
-enum unmap_queues_action_enum {
-	action__mes_unmap_queues__preempt_queues = 0,
-	action__mes_unmap_queues__reset_queues = 1,
-	action__mes_unmap_queues__disable_process_queues = 2
-};
-
-enum unmap_queues_queue_sel_enum {
-	queue_sel__mes_unmap_queues__perform_request_on_specified_queues = 0,
-	queue_sel__mes_unmap_queues__perform_request_on_pasid_queues = 1,
-	queue_sel__mes_unmap_queues__perform_request_on_all_active_queues = 2,
-	queue_sel__mes_unmap_queues__perform_request_on_dynamic_queues_only = 3
-};
-
-enum unmap_queues_engine_sel_enum {
-	engine_sel__mes_unmap_queues__compute = 0,
-	engine_sel__mes_unmap_queues__sdma0 = 2,
-	engine_sel__mes_unmap_queues__sdma1 = 3
-};
-
-struct pm4_unmap_queues {
-	union {
-		union PM4_MES_TYPE_3_HEADER header;	/* header */
-		uint32_t ordinal1;
-	};
-
-	union {
-		struct {
-			enum unmap_queues_action_enum action:2;
-			uint32_t reserved1:2;
-			enum unmap_queues_queue_sel_enum queue_sel:2;
-			uint32_t reserved2:20;
-			enum unmap_queues_engine_sel_enum engine_sel:3;
-			uint32_t num_queues:3;
+			uint32_t reserved1:8;
+			uint32_t diq_enable:1;
+			uint32_t process_quantum:7;
 		} bitfields2;
 		uint32_t ordinal2;
 	};
 
 	union {
 		struct {
-			uint32_t pasid:16;
-			uint32_t reserved3:16;
-		} bitfields3a;
-		struct {
-			uint32_t reserved4:2;
-			uint32_t doorbell_offset0:21;
-			uint32_t reserved5:9;
-		} bitfields3b;
+			uint32_t page_table_base:28;
+			uint32_t reserved2:4;
+		} bitfields3;
 		uint32_t ordinal3;
 	};
 
-	union {
-		struct {
-			uint32_t reserved6:2;
-			uint32_t doorbell_offset1:21;
-			uint32_t reserved7:9;
-		} bitfields4;
-		uint32_t ordinal4;
-	};
-
-	union {
-		struct {
-			uint32_t reserved8:2;
-			uint32_t doorbell_offset2:21;
-			uint32_t reserved9:9;
-		} bitfields5;
-		uint32_t ordinal5;
-	};
+	uint32_t reserved3;
+	uint32_t sh_mem_bases;
+	uint32_t sh_mem_config;
+	uint32_t sh_mem_ape1_base;
+	uint32_t sh_mem_ape1_limit;
+	uint32_t sh_hidden_private_base_vmid;
+	uint32_t reserved4;
+	uint32_t reserved5;
+	uint32_t gds_addr_lo;
+	uint32_t gds_addr_hi;
 
 	union {
 		struct {
-			uint32_t reserved10:2;
-			uint32_t doorbell_offset3:21;
-			uint32_t reserved11:9;
-		} bitfields6;
-		uint32_t ordinal6;
+			uint32_t num_gws:6;
+			uint32_t reserved6:2;
+			uint32_t num_oac:4;
+			uint32_t reserved7:4;
+			uint32_t gds_size:6;
+			uint32_t num_queues:10;
+		} bitfields14;
+		uint32_t ordinal14;
 	};
 
+	uint32_t completion_signal_lo32;
+uint32_t completion_signal_hi32;
 };
 #endif
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_pm4_headers_vi.h b/drivers/gpu/drm/amd/amdkfd/kfd_pm4_headers_vi.h
index 08c721922812..7c8d9b357749 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_pm4_headers_vi.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_pm4_headers_vi.h
@@ -30,10 +30,12 @@ union PM4_MES_TYPE_3_HEADER {
 	struct {
 		uint32_t reserved1 : 8; /* < reserved */
 		uint32_t opcode    : 8; /* < IT opcode */
-		uint32_t count     : 14;/* < number of DWORDs - 1 in the
-		information body. */
-		uint32_t type      : 2; /* < packet identifier.
-					It should be 3 for type 3 packets */
+		uint32_t count     : 14;/* < Number of DWORDS - 1 in the
+					 *   information body
+					 */
+		uint32_t type      : 2; /* < packet identifier
+					 *   It should be 3 for type 3 packets
+					 */
 	};
 	uint32_t u32All;
 };
@@ -124,9 +126,10 @@ struct pm4_mes_runlist {
 			uint32_t ib_size:20;
 			uint32_t chain:1;
 			uint32_t offload_polling:1;
-			uint32_t reserved3:1;
+			uint32_t reserved2:1;
 			uint32_t valid:1;
-			uint32_t reserved4:8;
+			uint32_t process_cnt:4;
+			uint32_t reserved3:4;
 		} bitfields4;
 		uint32_t ordinal4;
 	};
@@ -141,8 +144,8 @@ struct pm4_mes_runlist {
 
 struct pm4_mes_map_process {
 	union {
-		union PM4_MES_TYPE_3_HEADER   header;            /* header */
-		uint32_t            ordinal1;
+		union PM4_MES_TYPE_3_HEADER header;	/* header */
+		uint32_t ordinal1;
 	};
 
 	union {
@@ -153,36 +156,48 @@ struct pm4_mes_map_process {
 			uint32_t process_quantum:7;
 		} bitfields2;
 		uint32_t ordinal2;
-};
+	};
 
 	union {
 		struct {
 			uint32_t page_table_base:28;
-			uint32_t reserved2:4;
+			uint32_t reserved3:4;
 		} bitfields3;
 		uint32_t ordinal3;
 	};
 
+	uint32_t reserved;
+
 	uint32_t sh_mem_bases;
+	uint32_t sh_mem_config;
 	uint32_t sh_mem_ape1_base;
 	uint32_t sh_mem_ape1_limit;
-	uint32_t sh_mem_config;
+
+	uint32_t sh_hidden_private_base_vmid;
+
+	uint32_t reserved2;
+	uint32_t reserved3;
+
 	uint32_t gds_addr_lo;
 	uint32_t gds_addr_hi;
 
 	union {
 		struct {
 			uint32_t num_gws:6;
-			uint32_t reserved3:2;
+			uint32_t reserved4:2;
 			uint32_t num_oac:4;
-			uint32_t reserved4:4;
+			uint32_t reserved5:4;
 			uint32_t gds_size:6;
 			uint32_t num_queues:10;
 		} bitfields10;
 		uint32_t ordinal10;
 	};
 
+	uint32_t completion_signal_lo;
+	uint32_t completion_signal_hi;
+
 };
+
 #endif
 
 /*--------------------MES_MAP_QUEUES--------------------*/
@@ -335,7 +350,7 @@ enum mes_unmap_queues_engine_sel_enum {
 	engine_sel__mes_unmap_queues__sdmal = 3
 };
 
-struct PM4_MES_UNMAP_QUEUES {
+struct pm4_mes_unmap_queues {
 	union {
 		union PM4_MES_TYPE_3_HEADER   header;            /* header */
 		uint32_t            ordinal1;
@@ -395,4 +410,101 @@ struct PM4_MES_UNMAP_QUEUES {
 };
 #endif
 
+#ifndef PM4_MEC_RELEASE_MEM_DEFINED
+#define PM4_MEC_RELEASE_MEM_DEFINED
+enum RELEASE_MEM_event_index_enum {
+	event_index___release_mem__end_of_pipe = 5,
+	event_index___release_mem__shader_done = 6
+};
+
+enum RELEASE_MEM_cache_policy_enum {
+	cache_policy___release_mem__lru = 0,
+	cache_policy___release_mem__stream = 1,
+	cache_policy___release_mem__bypass = 2
+};
+
+enum RELEASE_MEM_dst_sel_enum {
+	dst_sel___release_mem__memory_controller = 0,
+	dst_sel___release_mem__tc_l2 = 1,
+	dst_sel___release_mem__queue_write_pointer_register = 2,
+	dst_sel___release_mem__queue_write_pointer_poll_mask_bit = 3
+};
+
+enum RELEASE_MEM_int_sel_enum {
+	int_sel___release_mem__none = 0,
+	int_sel___release_mem__send_interrupt_only = 1,
+	int_sel___release_mem__send_interrupt_after_write_confirm = 2,
+	int_sel___release_mem__send_data_after_write_confirm = 3
+};
+
+enum RELEASE_MEM_data_sel_enum {
+	data_sel___release_mem__none = 0,
+	data_sel___release_mem__send_32_bit_low = 1,
+	data_sel___release_mem__send_64_bit_data = 2,
+	data_sel___release_mem__send_gpu_clock_counter = 3,
+	data_sel___release_mem__send_cp_perfcounter_hi_lo = 4,
+	data_sel___release_mem__store_gds_data_to_memory = 5
+};
+
+struct pm4_mec_release_mem {
+	union {
+		union PM4_MES_TYPE_3_HEADER header;     /*header */
+		unsigned int ordinal1;
+	};
+
+	union {
+		struct {
+			unsigned int event_type:6;
+			unsigned int reserved1:2;
+			enum RELEASE_MEM_event_index_enum event_index:4;
+			unsigned int tcl1_vol_action_ena:1;
+			unsigned int tc_vol_action_ena:1;
+			unsigned int reserved2:1;
+			unsigned int tc_wb_action_ena:1;
+			unsigned int tcl1_action_ena:1;
+			unsigned int tc_action_ena:1;
+			unsigned int reserved3:6;
+			unsigned int atc:1;
+			enum RELEASE_MEM_cache_policy_enum cache_policy:2;
+			unsigned int reserved4:5;
+		} bitfields2;
+		unsigned int ordinal2;
+	};
+
+	union {
+		struct {
+			unsigned int reserved5:16;
+			enum RELEASE_MEM_dst_sel_enum dst_sel:2;
+			unsigned int reserved6:6;
+			enum RELEASE_MEM_int_sel_enum int_sel:3;
+			unsigned int reserved7:2;
+			enum RELEASE_MEM_data_sel_enum data_sel:3;
+		} bitfields3;
+		unsigned int ordinal3;
+	};
+
+	union {
+		struct {
+			unsigned int reserved8:2;
+			unsigned int address_lo_32b:30;
+		} bitfields4;
+		struct {
+			unsigned int reserved9:3;
+			unsigned int address_lo_64b:29;
+		} bitfields5;
+		unsigned int ordinal4;
+	};
+
+	unsigned int address_hi;
+
+	unsigned int data_lo;
+
+	unsigned int data_hi;
+};
+#endif
+
+enum {
+	CACHE_FLUSH_AND_INV_TS_EVENT = 0x00000014
+};
+
 #endif
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
index 4750cabe4252..b397ec726400 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
@@ -239,11 +239,6 @@ enum kfd_preempt_type_filter {
 	KFD_PREEMPT_TYPE_FILTER_BY_PASID
 };
 
-enum kfd_preempt_type {
-	KFD_PREEMPT_TYPE_WAVEFRONT,
-	KFD_PREEMPT_TYPE_WAVEFRONT_RESET
-};
-
 /**
  * enum kfd_queue_type
  *
@@ -294,13 +289,13 @@ enum kfd_queue_format {
  * @write_ptr: Defines the number of dwords written to the ring buffer.
  *
  * @doorbell_ptr: This field aim is to notify the H/W of new packet written to
- * the queue ring buffer. This field should be similar to write_ptr and the user
- * should update this field after he updated the write_ptr.
+ * the queue ring buffer. This field should be similar to write_ptr and the
+ * user should update this field after he updated the write_ptr.
  *
  * @doorbell_off: The doorbell offset in the doorbell pci-bar.
  *
- * @is_interop: Defines if this is a interop queue. Interop queue means that the
- * queue can access both graphics and compute resources.
+ * @is_interop: Defines if this is a interop queue. Interop queue means that
+ * the queue can access both graphics and compute resources.
  *
  * @is_active: Defines if the queue is active or not.
  *
@@ -352,9 +347,10 @@ struct queue_properties {
  * @properties: The queue properties.
  *
  * @mec: Used only in no cp scheduling mode and identifies to micro engine id
- * that the queue should be execute on.
+ *	 that the queue should be execute on.
  *
- * @pipe: Used only in no cp scheduling mode and identifies the queue's pipe id.
+ * @pipe: Used only in no cp scheduling mode and identifies the queue's pipe
+ *	  id.
  *
  * @queue: Used only in no cp scheduliong mode and identifies the queue's slot.
  *
@@ -436,6 +432,7 @@ struct qcm_process_device {
 	uint32_t gds_size;
 	uint32_t num_gws;
 	uint32_t num_oac;
+	uint32_t sh_hidden_private_base;
 };
 
 /* Data that is per-process-per device. */
@@ -520,8 +517,8 @@ struct kfd_process {
 	struct mutex event_mutex;
 	/* All events in process hashed by ID, linked on kfd_event.events. */
 	DECLARE_HASHTABLE(events, 4);
-	struct list_head signal_event_pages;	/* struct slot_page_header.
-								event_pages */
+	/* struct slot_page_header.event_pages */
+	struct list_head signal_event_pages;
 	u32 next_nonsignal_event_id;
 	size_t signal_event_count;
 };
@@ -559,8 +556,10 @@ struct kfd_process_device *kfd_create_process_device_data(struct kfd_dev *dev,
 							struct kfd_process *p);
 
 /* Process device data iterator */
-struct kfd_process_device *kfd_get_first_process_device_data(struct kfd_process *p);
-struct kfd_process_device *kfd_get_next_process_device_data(struct kfd_process *p,
+struct kfd_process_device *kfd_get_first_process_device_data(
+							struct kfd_process *p);
+struct kfd_process_device *kfd_get_next_process_device_data(
+						struct kfd_process *p,
 						struct kfd_process_device *pdd);
 bool kfd_has_process_device_data(struct kfd_process *p);
 
@@ -573,7 +572,8 @@ unsigned int kfd_pasid_alloc(void);
 void kfd_pasid_free(unsigned int pasid);
 
 /* Doorbells */
-void kfd_doorbell_init(struct kfd_dev *kfd);
+int kfd_doorbell_init(struct kfd_dev *kfd);
+void kfd_doorbell_fini(struct kfd_dev *kfd);
 int kfd_doorbell_mmap(struct kfd_process *process, struct vm_area_struct *vma);
 u32 __iomem *kfd_get_kernel_doorbell(struct kfd_dev *kfd,
 					unsigned int *doorbell_off);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
index 035bbc98a63d..c74cf22a1ed9 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_process.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
@@ -79,9 +79,7 @@ struct kfd_process *kfd_create_process(const struct task_struct *thread)
 {
 	struct kfd_process *process;
 
-	BUG_ON(!kfd_process_wq);
-
-	if (thread->mm == NULL)
+	if (!thread->mm)
 		return ERR_PTR(-EINVAL);
 
 	/* Only the pthreads threading model is supported. */
@@ -101,7 +99,7 @@ struct kfd_process *kfd_create_process(const struct task_struct *thread)
 	/* A prior open of /dev/kfd could have already created the process. */
 	process = find_process(thread);
 	if (process)
-		pr_debug("kfd: process already found\n");
+		pr_debug("Process already found\n");
 
 	if (!process)
 		process = create_process(thread);
@@ -117,7 +115,7 @@ struct kfd_process *kfd_get_process(const struct task_struct *thread)
 {
 	struct kfd_process *process;
 
-	if (thread->mm == NULL)
+	if (!thread->mm)
 		return ERR_PTR(-EINVAL);
 
 	/* Only the pthreads threading model is supported. */
@@ -202,10 +200,8 @@ static void kfd_process_destroy_delayed(struct rcu_head *rcu)
 	struct kfd_process_release_work *work;
 	struct kfd_process *p;
 
-	BUG_ON(!kfd_process_wq);
-
 	p = container_of(rcu, struct kfd_process, rcu);
-	BUG_ON(atomic_read(&p->mm->mm_count) <= 0);
+	WARN_ON(atomic_read(&p->mm->mm_count) <= 0);
 
 	mmdrop(p->mm);
 
@@ -229,7 +225,8 @@ static void kfd_process_notifier_release(struct mmu_notifier *mn,
 	 * mmu_notifier srcu is read locked
 	 */
 	p = container_of(mn, struct kfd_process, mmu_notifier);
-	BUG_ON(p->mm != mm);
+	if (WARN_ON(p->mm != mm))
+		return;
 
 	mutex_lock(&kfd_processes_mutex);
 	hash_del_rcu(&p->kfd_processes);
@@ -250,7 +247,7 @@ static void kfd_process_notifier_release(struct mmu_notifier *mn,
 			kfd_dbgmgr_destroy(pdd->dev->dbgmgr);
 
 		if (pdd->reset_wavefronts) {
-			pr_warn("amdkfd: Resetting all wave fronts\n");
+			pr_warn("Resetting all wave fronts\n");
 			dbgdev_wave_reset_wavefronts(pdd->dev, p);
 			pdd->reset_wavefronts = false;
 		}
@@ -407,8 +404,6 @@ void kfd_unbind_process_from_device(struct kfd_dev *dev, unsigned int pasid)
 	struct kfd_process *p;
 	struct kfd_process_device *pdd;
 
-	BUG_ON(dev == NULL);
-
 	/*
 	 * Look for the process that matches the pasid. If there is no such
 	 * process, we either released it in amdkfd's own notifier, or there
@@ -449,14 +444,16 @@ void kfd_unbind_process_from_device(struct kfd_dev *dev, unsigned int pasid)
 	mutex_unlock(&p->mutex);
 }
 
-struct kfd_process_device *kfd_get_first_process_device_data(struct kfd_process *p)
+struct kfd_process_device *kfd_get_first_process_device_data(
+						struct kfd_process *p)
 {
 	return list_first_entry(&p->per_device_data,
 				struct kfd_process_device,
 				per_device_list);
 }
 
-struct kfd_process_device *kfd_get_next_process_device_data(struct kfd_process *p,
+struct kfd_process_device *kfd_get_next_process_device_data(
+						struct kfd_process *p,
 						struct kfd_process_device *pdd)
 {
 	if (list_is_last(&pdd->per_device_list, &p->per_device_data))
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c
index 32cdf2b483db..1cae95e2b13a 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c
@@ -32,12 +32,9 @@ static inline struct process_queue_node *get_queue_by_qid(
 {
 	struct process_queue_node *pqn;
 
-	BUG_ON(!pqm);
-
 	list_for_each_entry(pqn, &pqm->queues, process_queue_list) {
-		if (pqn->q && pqn->q->properties.queue_id == qid)
-			return pqn;
-		if (pqn->kq && pqn->kq->queue->properties.queue_id == qid)
+		if ((pqn->q && pqn->q->properties.queue_id == qid) ||
+		    (pqn->kq && pqn->kq->queue->properties.queue_id == qid))
 			return pqn;
 	}
 
@@ -49,17 +46,13 @@ static int find_available_queue_slot(struct process_queue_manager *pqm,
 {
 	unsigned long found;
 
-	BUG_ON(!pqm || !qid);
-
-	pr_debug("kfd: in %s\n", __func__);
-
 	found = find_first_zero_bit(pqm->queue_slot_bitmap,
 			KFD_MAX_NUM_OF_QUEUES_PER_PROCESS);
 
-	pr_debug("kfd: the new slot id %lu\n", found);
+	pr_debug("The new slot id %lu\n", found);
 
 	if (found >= KFD_MAX_NUM_OF_QUEUES_PER_PROCESS) {
-		pr_info("amdkfd: Can not open more queues for process with pasid %d\n",
+		pr_info("Cannot open more queues for process with pasid %d\n",
 				pqm->process->pasid);
 		return -ENOMEM;
 	}
@@ -72,13 +65,11 @@ static int find_available_queue_slot(struct process_queue_manager *pqm,
 
 int pqm_init(struct process_queue_manager *pqm, struct kfd_process *p)
 {
-	BUG_ON(!pqm);
-
 	INIT_LIST_HEAD(&pqm->queues);
 	pqm->queue_slot_bitmap =
 			kzalloc(DIV_ROUND_UP(KFD_MAX_NUM_OF_QUEUES_PER_PROCESS,
 					BITS_PER_BYTE), GFP_KERNEL);
-	if (pqm->queue_slot_bitmap == NULL)
+	if (!pqm->queue_slot_bitmap)
 		return -ENOMEM;
 	pqm->process = p;
 
@@ -90,10 +81,6 @@ void pqm_uninit(struct process_queue_manager *pqm)
 	int retval;
 	struct process_queue_node *pqn, *next;
 
-	BUG_ON(!pqm);
-
-	pr_debug("In func %s\n", __func__);
-
 	list_for_each_entry_safe(pqn, next, &pqm->queues, process_queue_list) {
 		retval = pqm_destroy_queue(
 				pqm,
@@ -102,7 +89,7 @@ void pqm_uninit(struct process_queue_manager *pqm)
 					pqn->kq->queue->properties.queue_id);
 
 		if (retval != 0) {
-			pr_err("kfd: failed to destroy queue\n");
+			pr_err("failed to destroy queue\n");
 			return;
 		}
 	}
@@ -117,8 +104,6 @@ static int create_cp_queue(struct process_queue_manager *pqm,
 {
 	int retval;
 
-	retval = 0;
-
 	/* Doorbell initialized in user space*/
 	q_properties->doorbell_ptr = NULL;
 
@@ -131,16 +116,13 @@ static int create_cp_queue(struct process_queue_manager *pqm,
 
 	retval = init_queue(q, q_properties);
 	if (retval != 0)
-		goto err_init_queue;
+		return retval;
 
 	(*q)->device = dev;
 	(*q)->process = pqm->process;
 
-	pr_debug("kfd: PQM After init queue");
-
-	return retval;
+	pr_debug("PQM After init queue");
 
-err_init_queue:
 	return retval;
 }
 
@@ -161,8 +143,6 @@ int pqm_create_queue(struct process_queue_manager *pqm,
 	int num_queues = 0;
 	struct queue *cur;
 
-	BUG_ON(!pqm || !dev || !properties || !qid);
-
 	memset(&q_properties, 0, sizeof(struct queue_properties));
 	memcpy(&q_properties, properties, sizeof(struct queue_properties));
 	q = NULL;
@@ -185,7 +165,7 @@ int pqm_create_queue(struct process_queue_manager *pqm,
 		list_for_each_entry(cur, &pdd->qpd.queues_list, list)
 			num_queues++;
 		if (num_queues >= dev->device_info->max_no_of_hqd/2)
-			return (-ENOSPC);
+			return -ENOSPC;
 	}
 
 	retval = find_available_queue_slot(pqm, qid);
@@ -197,7 +177,7 @@ int pqm_create_queue(struct process_queue_manager *pqm,
 		dev->dqm->ops.register_process(dev->dqm, &pdd->qpd);
 	}
 
-	pqn = kzalloc(sizeof(struct process_queue_node), GFP_KERNEL);
+	pqn = kzalloc(sizeof(*pqn), GFP_KERNEL);
 	if (!pqn) {
 		retval = -ENOMEM;
 		goto err_allocate_pqn;
@@ -210,7 +190,7 @@ int pqm_create_queue(struct process_queue_manager *pqm,
 		if ((sched_policy == KFD_SCHED_POLICY_HWS_NO_OVERSUBSCRIPTION) &&
 		((dev->dqm->processes_count >= VMID_PER_DEVICE) ||
 		(dev->dqm->queue_count >= get_queues_num(dev->dqm)))) {
-			pr_err("kfd: over-subscription is not allowed in radeon_kfd.sched_policy == 1\n");
+			pr_err("Over-subscription is not allowed in radeon_kfd.sched_policy == 1\n");
 			retval = -EPERM;
 			goto err_create_queue;
 		}
@@ -227,7 +207,7 @@ int pqm_create_queue(struct process_queue_manager *pqm,
 		break;
 	case KFD_QUEUE_TYPE_DIQ:
 		kq = kernel_queue_init(dev, KFD_QUEUE_TYPE_DIQ);
-		if (kq == NULL) {
+		if (!kq) {
 			retval = -ENOMEM;
 			goto err_create_queue;
 		}
@@ -238,22 +218,22 @@ int pqm_create_queue(struct process_queue_manager *pqm,
 							kq, &pdd->qpd);
 		break;
 	default:
-		BUG();
-		break;
+		WARN(1, "Invalid queue type %d", type);
+		retval = -EINVAL;
 	}
 
 	if (retval != 0) {
-		pr_debug("Error dqm create queue\n");
+		pr_err("DQM create queue failed\n");
 		goto err_create_queue;
 	}
 
-	pr_debug("kfd: PQM After DQM create queue\n");
+	pr_debug("PQM After DQM create queue\n");
 
 	list_add(&pqn->process_queue_list, &pqm->queues);
 
 	if (q) {
 		*properties = q->properties;
-		pr_debug("kfd: PQM done creating queue\n");
+		pr_debug("PQM done creating queue\n");
 		print_queue_properties(properties);
 	}
 
@@ -279,14 +259,11 @@ int pqm_destroy_queue(struct process_queue_manager *pqm, unsigned int qid)
 
 	dqm = NULL;
 
-	BUG_ON(!pqm);
 	retval = 0;
 
-	pr_debug("kfd: In Func %s\n", __func__);
-
 	pqn = get_queue_by_qid(pqm, qid);
-	if (pqn == NULL) {
-		pr_err("kfd: queue id does not match any known queue\n");
+	if (!pqn) {
+		pr_err("Queue id does not match any known queue\n");
 		return -EINVAL;
 	}
 
@@ -295,7 +272,8 @@ int pqm_destroy_queue(struct process_queue_manager *pqm, unsigned int qid)
 		dev = pqn->kq->dev;
 	if (pqn->q)
 		dev = pqn->q->device;
-	BUG_ON(!dev);
+	if (WARN_ON(!dev))
+		return -ENODEV;
 
 	pdd = kfd_get_process_device_data(dev, pqm->process);
 	if (!pdd) {
@@ -335,12 +313,9 @@ int pqm_update_queue(struct process_queue_manager *pqm, unsigned int qid,
 	int retval;
 	struct process_queue_node *pqn;
 
-	BUG_ON(!pqm);
-
 	pqn = get_queue_by_qid(pqm, qid);
 	if (!pqn) {
-		pr_debug("amdkfd: No queue %d exists for update operation\n",
-				qid);
+		pr_debug("No queue %d exists for update operation\n", qid);
 		return -EFAULT;
 	}
 
@@ -363,8 +338,6 @@ struct kernel_queue *pqm_get_kernel_queue(
 {
 	struct process_queue_node *pqn;
 
-	BUG_ON(!pqm);
-
 	pqn = get_queue_by_qid(pqm, qid);
 	if (pqn && pqn->kq)
 		return pqn->kq;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_queue.c b/drivers/gpu/drm/amd/amdkfd/kfd_queue.c
index 0ab197077f2d..a5315d4f1c95 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_queue.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_queue.c
@@ -65,17 +65,15 @@ void print_queue(struct queue *q)
 
 int init_queue(struct queue **q, const struct queue_properties *properties)
 {
-	struct queue *tmp;
+	struct queue *tmp_q;
 
-	BUG_ON(!q);
-
-	tmp = kzalloc(sizeof(struct queue), GFP_KERNEL);
-	if (!tmp)
+	tmp_q = kzalloc(sizeof(*tmp_q), GFP_KERNEL);
+	if (!tmp_q)
 		return -ENOMEM;
 
-	memcpy(&tmp->properties, properties, sizeof(struct queue_properties));
+	memcpy(&tmp_q->properties, properties, sizeof(*properties));
 
-	*q = tmp;
+	*q = tmp_q;
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
index 1e5064749959..19ce59028d6b 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
@@ -108,9 +108,6 @@ static int kfd_topology_get_crat_acpi(void *crat_image, size_t *size)
 static void kfd_populated_cu_info_cpu(struct kfd_topology_device *dev,
 		struct crat_subtype_computeunit *cu)
 {
-	BUG_ON(!dev);
-	BUG_ON(!cu);
-
 	dev->node_props.cpu_cores_count = cu->num_cpu_cores;
 	dev->node_props.cpu_core_id_base = cu->processor_id_low;
 	if (cu->hsa_capability & CRAT_CU_FLAGS_IOMMU_PRESENT)
@@ -123,9 +120,6 @@ static void kfd_populated_cu_info_cpu(struct kfd_topology_device *dev,
 static void kfd_populated_cu_info_gpu(struct kfd_topology_device *dev,
 		struct crat_subtype_computeunit *cu)
 {
-	BUG_ON(!dev);
-	BUG_ON(!cu);
-
 	dev->node_props.simd_id_base = cu->processor_id_low;
 	dev->node_props.simd_count = cu->num_simd_cores;
 	dev->node_props.lds_size_in_kb = cu->lds_size_in_kb;
@@ -148,8 +142,6 @@ static int kfd_parse_subtype_cu(struct crat_subtype_computeunit *cu)
 	struct kfd_topology_device *dev;
 	int i = 0;
 
-	BUG_ON(!cu);
-
 	pr_info("Found CU entry in CRAT table with proximity_domain=%d caps=%x\n",
 			cu->proximity_domain, cu->hsa_capability);
 	list_for_each_entry(dev, &topology_device_list, list) {
@@ -177,8 +169,6 @@ static int kfd_parse_subtype_mem(struct crat_subtype_memory *mem)
 	struct kfd_topology_device *dev;
 	int i = 0;
 
-	BUG_ON(!mem);
-
 	pr_info("Found memory entry in CRAT table with proximity_domain=%d\n",
 			mem->promixity_domain);
 	list_for_each_entry(dev, &topology_device_list, list) {
@@ -223,8 +213,6 @@ static int kfd_parse_subtype_cache(struct crat_subtype_cache *cache)
 	struct kfd_topology_device *dev;
 	uint32_t id;
 
-	BUG_ON(!cache);
-
 	id = cache->processor_id_low;
 
 	pr_info("Found cache entry in CRAT table with processor_id=%d\n", id);
@@ -274,8 +262,6 @@ static int kfd_parse_subtype_iolink(struct crat_subtype_iolink *iolink)
 	uint32_t id_from;
 	uint32_t id_to;
 
-	BUG_ON(!iolink);
-
 	id_from = iolink->proximity_domain_from;
 	id_to = iolink->proximity_domain_to;
 
@@ -323,8 +309,6 @@ static int kfd_parse_subtype(struct crat_subtype_generic *sub_type_hdr)
 	struct crat_subtype_iolink *iolink;
 	int ret = 0;
 
-	BUG_ON(!sub_type_hdr);
-
 	switch (sub_type_hdr->type) {
 	case CRAT_SUBTYPE_COMPUTEUNIT_AFFINITY:
 		cu = (struct crat_subtype_computeunit *)sub_type_hdr;
@@ -368,8 +352,6 @@ static void kfd_release_topology_device(struct kfd_topology_device *dev)
 	struct kfd_cache_properties *cache;
 	struct kfd_iolink_properties *iolink;
 
-	BUG_ON(!dev);
-
 	list_del(&dev->list);
 
 	while (dev->mem_props.next != &dev->mem_props) {
@@ -416,7 +398,7 @@ static struct kfd_topology_device *kfd_create_topology_device(void)
 	struct kfd_topology_device *dev;
 
 	dev = kfd_alloc_struct(dev);
-	if (dev == NULL) {
+	if (!dev) {
 		pr_err("No memory to allocate a topology device");
 		return NULL;
 	}
@@ -666,7 +648,7 @@ static ssize_t node_show(struct kobject *kobj, struct attribute *attr,
 			dev->node_props.simd_count);
 
 	if (dev->mem_bank_count < dev->node_props.mem_banks_count) {
-		pr_info_once("kfd: mem_banks_count truncated from %d to %d\n",
+		pr_info_once("mem_banks_count truncated from %d to %d\n",
 				dev->node_props.mem_banks_count,
 				dev->mem_bank_count);
 		sysfs_show_32bit_prop(buffer, "mem_banks_count",
@@ -763,8 +745,6 @@ static void kfd_remove_sysfs_node_entry(struct kfd_topology_device *dev)
 	struct kfd_cache_properties *cache;
 	struct kfd_mem_properties *mem;
 
-	BUG_ON(!dev);
-
 	if (dev->kobj_iolink) {
 		list_for_each_entry(iolink, &dev->io_link_props, list)
 			if (iolink->kobj) {
@@ -819,12 +799,12 @@ static int kfd_build_sysfs_node_entry(struct kfd_topology_device *dev,
 	int ret;
 	uint32_t i;
 
-	BUG_ON(!dev);
+	if (WARN_ON(dev->kobj_node))
+		return -EEXIST;
 
 	/*
 	 * Creating the sysfs folders
 	 */
-	BUG_ON(dev->kobj_node);
 	dev->kobj_node = kfd_alloc_struct(dev->kobj_node);
 	if (!dev->kobj_node)
 		return -ENOMEM;
@@ -957,7 +937,7 @@ static int kfd_topology_update_sysfs(void)
 	int ret;
 
 	pr_info("Creating topology SYSFS entries\n");
-	if (sys_props.kobj_topology == NULL) {
+	if (!sys_props.kobj_topology) {
 		sys_props.kobj_topology =
 				kfd_alloc_struct(sys_props.kobj_topology);
 		if (!sys_props.kobj_topology)
@@ -1117,10 +1097,8 @@ static struct kfd_topology_device *kfd_assign_gpu(struct kfd_dev *gpu)
 	struct kfd_topology_device *dev;
 	struct kfd_topology_device *out_dev = NULL;
 
-	BUG_ON(!gpu);
-
 	list_for_each_entry(dev, &topology_device_list, list)
-		if (dev->gpu == NULL && dev->node_props.simd_count > 0) {
+		if (!dev->gpu && (dev->node_props.simd_count > 0)) {
 			dev->gpu = gpu;
 			out_dev = dev;
 			break;
@@ -1143,11 +1121,9 @@ int kfd_topology_add_device(struct kfd_dev *gpu)
 	struct kfd_topology_device *dev;
 	int res;
 
-	BUG_ON(!gpu);
-
 	gpu_id = kfd_generate_gpu_id(gpu);
 
-	pr_debug("kfd: Adding new GPU (ID: 0x%x) to topology\n", gpu_id);
+	pr_debug("Adding new GPU (ID: 0x%x) to topology\n", gpu_id);
 
 	down_write(&topology_lock);
 	/*
@@ -1170,8 +1146,8 @@ int kfd_topology_add_device(struct kfd_dev *gpu)
 		 * GPU vBIOS
 		 */
 
-		/*
-		 * Update the SYSFS tree, since we added another topology device
+		/* Update the SYSFS tree, since we added another topology
+		 * device
 		 */
 		if (kfd_topology_update_sysfs() < 0)
 			kfd_topology_release_sysfs();
@@ -1190,7 +1166,7 @@ int kfd_topology_add_device(struct kfd_dev *gpu)
 
 	if (dev->gpu->device_info->asic_family == CHIP_CARRIZO) {
 		dev->node_props.capability |= HSA_CAP_DOORBELL_PACKET_TYPE;
-		pr_info("amdkfd: adding doorbell packet type capability\n");
+		pr_info("Adding doorbell packet type capability\n");
 	}
 
 	res = 0;
@@ -1210,8 +1186,6 @@ int kfd_topology_remove_device(struct kfd_dev *gpu)
 	uint32_t gpu_id;
 	int res = -ENODEV;
 
-	BUG_ON(!gpu);
-
 	down_write(&topology_lock);
 
 	list_for_each_entry(dev, &topology_device_list, list)
diff --git a/drivers/gpu/drm/amd/include/kgd_kfd_interface.h b/drivers/gpu/drm/amd/include/kgd_kfd_interface.h
index 36f376677a53..94277cb734d2 100644
--- a/drivers/gpu/drm/amd/include/kgd_kfd_interface.h
+++ b/drivers/gpu/drm/amd/include/kgd_kfd_interface.h
@@ -41,6 +41,11 @@ struct kgd_dev;
 
 struct kgd_mem;
 
+enum kfd_preempt_type {
+	KFD_PREEMPT_TYPE_WAVEFRONT_DRAIN = 0,
+	KFD_PREEMPT_TYPE_WAVEFRONT_RESET,
+};
+
 enum kgd_memory_pool {
 	KGD_POOL_SYSTEM_CACHEABLE = 1,
 	KGD_POOL_SYSTEM_WRITECOMBINE = 2,
@@ -82,6 +87,17 @@ struct kgd2kfd_shared_resources {
 	size_t doorbell_start_offset;
 };
 
+struct tile_config {
+	uint32_t *tile_config_ptr;
+	uint32_t *macro_tile_config_ptr;
+	uint32_t num_tile_configs;
+	uint32_t num_macro_tile_configs;
+
+	uint32_t gb_addr_config;
+	uint32_t num_banks;
+	uint32_t num_ranks;
+};
+
 /**
  * struct kfd2kgd_calls
  *
@@ -123,6 +139,11 @@ struct kgd2kfd_shared_resources {
  *
  * @get_fw_version: Returns FW versions from the header
  *
+ * @set_scratch_backing_va: Sets VA for scratch backing memory of a VMID.
+ * Only used for no cp scheduling mode
+ *
+ * @get_tile_config: Returns GPU-specific tiling mode information
+ *
  * This structure contains function pointers to services that the kgd driver
  * provides to amdkfd driver.
  *
@@ -153,14 +174,16 @@ struct kfd2kgd_calls {
 	int (*init_interrupts)(struct kgd_dev *kgd, uint32_t pipe_id);
 
 	int (*hqd_load)(struct kgd_dev *kgd, void *mqd, uint32_t pipe_id,
-			uint32_t queue_id, uint32_t __user *wptr);
+			uint32_t queue_id, uint32_t __user *wptr,
+			uint32_t wptr_shift, uint32_t wptr_mask,
+			struct mm_struct *mm);
 
 	int (*hqd_sdma_load)(struct kgd_dev *kgd, void *mqd);
 
 	bool (*hqd_is_occupied)(struct kgd_dev *kgd, uint64_t queue_address,
 				uint32_t pipe_id, uint32_t queue_id);
 
-	int (*hqd_destroy)(struct kgd_dev *kgd, uint32_t reset_type,
+	int (*hqd_destroy)(struct kgd_dev *kgd, void *mqd, uint32_t reset_type,
 				unsigned int timeout, uint32_t pipe_id,
 				uint32_t queue_id);
 
@@ -192,6 +215,9 @@ struct kfd2kgd_calls {
 
 	uint16_t (*get_fw_version)(struct kgd_dev *kgd,
 				enum kgd_engine_type type);
+	void (*set_scratch_backing_va)(struct kgd_dev *kgd,
+				uint64_t va, uint32_t vmid);
+	int (*get_tile_config)(struct kgd_dev *kgd, struct tile_config *config);
 };
 
 /**
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/cz_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/cz_hwmgr.c
index 0b74da3dca8b..bc839ff0bdd0 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/cz_hwmgr.c
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/cz_hwmgr.c
@@ -1240,13 +1240,18 @@ static int cz_phm_force_dpm_highest(struct pp_hwmgr *hwmgr)
 {
 	struct cz_hwmgr *cz_hwmgr = (struct cz_hwmgr *)(hwmgr->backend);
 
-	if (cz_hwmgr->sclk_dpm.soft_min_clk !=
-				cz_hwmgr->sclk_dpm.soft_max_clk)
-		smum_send_msg_to_smc_with_parameter(hwmgr->smumgr,
-						PPSMC_MSG_SetSclkSoftMin,
-						cz_get_sclk_level(hwmgr,
-						cz_hwmgr->sclk_dpm.soft_max_clk,
-						PPSMC_MSG_SetSclkSoftMin));
+	smum_send_msg_to_smc_with_parameter(hwmgr->smumgr,
+					PPSMC_MSG_SetSclkSoftMin,
+					cz_get_sclk_level(hwmgr,
+					cz_hwmgr->sclk_dpm.soft_max_clk,
+					PPSMC_MSG_SetSclkSoftMin));
+
+	smum_send_msg_to_smc_with_parameter(hwmgr->smumgr,
+				PPSMC_MSG_SetSclkSoftMax,
+				cz_get_sclk_level(hwmgr,
+				cz_hwmgr->sclk_dpm.soft_max_clk,
+				PPSMC_MSG_SetSclkSoftMax));
+
 	return 0;
 }
 
@@ -1292,17 +1297,55 @@ static int cz_phm_force_dpm_lowest(struct pp_hwmgr *hwmgr)
 {
 	struct cz_hwmgr *cz_hwmgr = (struct cz_hwmgr *)(hwmgr->backend);
 
-	if (cz_hwmgr->sclk_dpm.soft_min_clk !=
-				cz_hwmgr->sclk_dpm.soft_max_clk) {
-		cz_hwmgr->sclk_dpm.soft_max_clk =
-			cz_hwmgr->sclk_dpm.soft_min_clk;
+	smum_send_msg_to_smc_with_parameter(hwmgr->smumgr,
+			PPSMC_MSG_SetSclkSoftMax,
+			cz_get_sclk_level(hwmgr,
+			cz_hwmgr->sclk_dpm.soft_min_clk,
+			PPSMC_MSG_SetSclkSoftMax));
 
-		smum_send_msg_to_smc_with_parameter(hwmgr->smumgr,
+	smum_send_msg_to_smc_with_parameter(hwmgr->smumgr,
+				PPSMC_MSG_SetSclkSoftMin,
+				cz_get_sclk_level(hwmgr,
+				cz_hwmgr->sclk_dpm.soft_min_clk,
+				PPSMC_MSG_SetSclkSoftMin));
+
+	return 0;
+}
+
+static int cz_phm_force_dpm_sclk(struct pp_hwmgr *hwmgr, uint32_t sclk)
+{
+	smum_send_msg_to_smc_with_parameter(hwmgr->smumgr,
+				PPSMC_MSG_SetSclkSoftMin,
+				cz_get_sclk_level(hwmgr,
+				sclk,
+				PPSMC_MSG_SetSclkSoftMin));
+
+	smum_send_msg_to_smc_with_parameter(hwmgr->smumgr,
 				PPSMC_MSG_SetSclkSoftMax,
 				cz_get_sclk_level(hwmgr,
-				cz_hwmgr->sclk_dpm.soft_max_clk,
+				sclk,
 				PPSMC_MSG_SetSclkSoftMax));
+	return 0;
+}
+
+static int cz_get_profiling_clk(struct pp_hwmgr *hwmgr, uint32_t *sclk)
+{
+	struct phm_clock_voltage_dependency_table *table =
+		hwmgr->dyn_state.vddc_dependency_on_sclk;
+	int32_t tmp_sclk;
+	int32_t count;
+
+	tmp_sclk = table->entries[table->count-1].clk * 70 / 100;
+
+	for (count = table->count-1; count >= 0; count--) {
+		if (tmp_sclk >= table->entries[count].clk) {
+			tmp_sclk = table->entries[count].clk;
+			*sclk = tmp_sclk;
+			break;
+		}
 	}
+	if (count < 0)
+		*sclk = table->entries[0].clk;
 
 	return 0;
 }
@@ -1310,30 +1353,70 @@ static int cz_phm_force_dpm_lowest(struct pp_hwmgr *hwmgr)
 static int cz_dpm_force_dpm_level(struct pp_hwmgr *hwmgr,
 				enum amd_dpm_forced_level level)
 {
+	uint32_t sclk = 0;
 	int ret = 0;
+	uint32_t profile_mode_mask = AMD_DPM_FORCED_LEVEL_PROFILE_STANDARD |
+					AMD_DPM_FORCED_LEVEL_PROFILE_MIN_SCLK |
+					AMD_DPM_FORCED_LEVEL_PROFILE_PEAK;
+
+	if (level == hwmgr->dpm_level)
+		return ret;
+
+	if (!(hwmgr->dpm_level & profile_mode_mask)) {
+		/* enter profile mode, save current level, disable gfx cg*/
+		if (level & profile_mode_mask) {
+			hwmgr->saved_dpm_level = hwmgr->dpm_level;
+			cgs_set_clockgating_state(hwmgr->device,
+						AMD_IP_BLOCK_TYPE_GFX,
+						AMD_CG_STATE_UNGATE);
+		}
+	} else {
+		/* exit profile mode, restore level, enable gfx cg*/
+		if (!(level & profile_mode_mask)) {
+			if (level == AMD_DPM_FORCED_LEVEL_PROFILE_EXIT)
+				level = hwmgr->saved_dpm_level;
+			cgs_set_clockgating_state(hwmgr->device,
+					AMD_IP_BLOCK_TYPE_GFX,
+					AMD_CG_STATE_GATE);
+		}
+	}
 
 	switch (level) {
 	case AMD_DPM_FORCED_LEVEL_HIGH:
+	case AMD_DPM_FORCED_LEVEL_PROFILE_PEAK:
 		ret = cz_phm_force_dpm_highest(hwmgr);
 		if (ret)
 			return ret;
+		hwmgr->dpm_level = level;
 		break;
 	case AMD_DPM_FORCED_LEVEL_LOW:
+	case AMD_DPM_FORCED_LEVEL_PROFILE_MIN_SCLK:
 		ret = cz_phm_force_dpm_lowest(hwmgr);
 		if (ret)
 			return ret;
+		hwmgr->dpm_level = level;
 		break;
 	case AMD_DPM_FORCED_LEVEL_AUTO:
 		ret = cz_phm_unforce_dpm_levels(hwmgr);
 		if (ret)
 			return ret;
+		hwmgr->dpm_level = level;
+		break;
+	case AMD_DPM_FORCED_LEVEL_PROFILE_STANDARD:
+		ret = cz_get_profiling_clk(hwmgr, &sclk);
+		if (ret)
+			return ret;
+		hwmgr->dpm_level = level;
+		cz_phm_force_dpm_sclk(hwmgr, sclk);
+		break;
+	case AMD_DPM_FORCED_LEVEL_MANUAL:
+		hwmgr->dpm_level = level;
 		break;
+	case AMD_DPM_FORCED_LEVEL_PROFILE_EXIT:
 	default:
 		break;
 	}
 
-	hwmgr->dpm_level = level;
-
 	return ret;
 }
 
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/hwmgr.c
index d025653c7823..9547f265a8bb 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/hwmgr.c
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/hwmgr.c
@@ -557,9 +557,8 @@ uint16_t phm_find_closest_vddci(struct pp_atomctrl_voltage_table *vddci_table, u
 			return vddci_table->entries[i].value;
 	}
 
-	PP_ASSERT_WITH_CODE(false,
-			"VDDCI is larger than max VDDCI in VDDCI Voltage Table!",
-			return vddci_table->entries[i-1].value);
+	pr_debug("vddci is larger than max value in vddci_table\n");
+	return vddci_table->entries[i-1].value;
 }
 
 int phm_find_boot_level(void *table,
@@ -583,26 +582,26 @@ int phm_get_sclk_for_voltage_evv(struct pp_hwmgr *hwmgr,
 	phm_ppt_v1_voltage_lookup_table *lookup_table,
 	uint16_t virtual_voltage_id, int32_t *sclk)
 {
-	uint8_t entryId;
-	uint8_t voltageId;
+	uint8_t entry_id;
+	uint8_t voltage_id;
 	struct phm_ppt_v1_information *table_info =
 			(struct phm_ppt_v1_information *)(hwmgr->pptable);
 
 	PP_ASSERT_WITH_CODE(lookup_table->count != 0, "Lookup table is empty", return -EINVAL);
 
 	/* search for leakage voltage ID 0xff01 ~ 0xff08 and sckl */
-	for (entryId = 0; entryId < table_info->vdd_dep_on_sclk->count; entryId++) {
-		voltageId = table_info->vdd_dep_on_sclk->entries[entryId].vddInd;
-		if (lookup_table->entries[voltageId].us_vdd == virtual_voltage_id)
+	for (entry_id = 0; entry_id < table_info->vdd_dep_on_sclk->count; entry_id++) {
+		voltage_id = table_info->vdd_dep_on_sclk->entries[entry_id].vddInd;
+		if (lookup_table->entries[voltage_id].us_vdd == virtual_voltage_id)
 			break;
 	}
 
-	PP_ASSERT_WITH_CODE(entryId < table_info->vdd_dep_on_sclk->count,
-			"Can't find requested voltage id in vdd_dep_on_sclk table!",
-			return -EINVAL;
-			);
+	if (entry_id >= table_info->vdd_dep_on_sclk->count) {
+		pr_debug("Can't find requested voltage id in vdd_dep_on_sclk table\n");
+		return -EINVAL;
+	}
 
-	*sclk = table_info->vdd_dep_on_sclk->entries[entryId].clk;
+	*sclk = table_info->vdd_dep_on_sclk->entries[entry_id].clk;
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/ppatomfwctrl.c b/drivers/gpu/drm/amd/powerplay/hwmgr/ppatomfwctrl.c
index cd33eb179db2..c062844b15f3 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/ppatomfwctrl.c
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/ppatomfwctrl.c
@@ -142,7 +142,7 @@ int pp_atomfwctrl_get_voltage_table_v4(struct pp_hwmgr *hwmgr,
 		}
 	} else if (voltage_mode == VOLTAGE_OBJ_SVID2) {
 		voltage_table->psi1_enable =
-			voltage_object->svid2_voltage_obj.loadline_psi1 & 0x1;
+			(voltage_object->svid2_voltage_obj.loadline_psi1 & 0x20) >> 5;
 		voltage_table->psi0_enable =
 			voltage_object->svid2_voltage_obj.psi0_enable & 0x1;
 		voltage_table->max_vid_step =
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/rv_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/rv_hwmgr.c
index 4c7f430b36eb..edc5fb6412d9 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/rv_hwmgr.c
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/rv_hwmgr.c
@@ -265,6 +265,15 @@ static int rv_tf_set_clock_limit(struct pp_hwmgr *hwmgr, void *input,
 		}
 	} */
 
+	if (((hwmgr->uvd_arbiter.vclk_soft_min / 100) != rv_data->vclk_soft_min) ||
+	    ((hwmgr->uvd_arbiter.dclk_soft_min / 100) != rv_data->dclk_soft_min)) {
+		rv_data->vclk_soft_min = hwmgr->uvd_arbiter.vclk_soft_min / 100;
+		rv_data->dclk_soft_min = hwmgr->uvd_arbiter.dclk_soft_min / 100;
+		smum_send_msg_to_smc_with_parameter(hwmgr->smumgr,
+			PPSMC_MSG_SetSoftMinVcn,
+			(rv_data->vclk_soft_min << 16) | rv_data->vclk_soft_min);
+	}
+
 	if((hwmgr->gfx_arbiter.sclk_hard_min != 0) &&
 		((hwmgr->gfx_arbiter.sclk_hard_min / 100) != rv_data->soc_actual_hard_min_freq)) {
 		smum_send_msg_to_smc_with_parameter(hwmgr->smumgr,
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/rv_hwmgr.h b/drivers/gpu/drm/amd/powerplay/hwmgr/rv_hwmgr.h
index afb852295a15..2472b50e54cf 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/rv_hwmgr.h
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/rv_hwmgr.h
@@ -280,6 +280,8 @@ struct rv_hwmgr {
 
 	uint32_t                        f_actual_hard_min_freq;
 	uint32_t                        fabric_actual_soft_min_freq;
+	uint32_t                        vclk_soft_min;
+	uint32_t                        dclk_soft_min;
 	uint32_t                        gfx_actual_soft_min_freq;
 
 	bool                           vcn_power_gated;
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c
index f01cda93f178..c2743233ba10 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c
@@ -1962,9 +1962,6 @@ static int smu7_thermal_parameter_init(struct pp_hwmgr *hwmgr)
 			temp_reg = PHM_SET_FIELD(temp_reg, CNB_PWRMGT_CNTL, DPM_ENABLED, 0x1);
 			break;
 		default:
-			PP_ASSERT_WITH_CODE(0,
-			"Failed to setup PCC HW register! Wrong GPIO assigned for VDDC_PCC_GPIO_PINID!",
-			);
 			break;
 		}
 		cgs_write_ind_register(hwmgr->device, CGS_IND_REG__SMC, ixCNB_PWRMGT_CNTL, temp_reg);
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c
index 01ff5054041b..9d71a259d97d 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c
@@ -2313,7 +2313,7 @@ static int vega10_acg_enable(struct pp_hwmgr *hwmgr)
 		smum_send_msg_to_smc(hwmgr->smumgr, PPSMC_MSG_InitializeAcg);
 
 		smum_send_msg_to_smc(hwmgr->smumgr, PPSMC_MSG_RunAcgBtc);
-		vega10_read_arg_from_smc(hwmgr->smumgr, &agc_btc_response);;
+		vega10_read_arg_from_smc(hwmgr->smumgr, &agc_btc_response);
 
 		if (1 == agc_btc_response) {
 			if (1 == data->acg_loop_state)
@@ -2522,6 +2522,9 @@ static int vega10_init_smc_table(struct pp_hwmgr *hwmgr)
 	pp_table->DisplayDpmVoltageMode =
 			(uint8_t)(table_info->uc_dcef_dpm_voltage_mode);
 
+	data->vddc_voltage_table.psi0_enable = voltage_table.psi0_enable;
+	data->vddc_voltage_table.psi1_enable = voltage_table.psi1_enable;
+
 	if (data->registry_data.ulv_support &&
 			table_info->us_ulv_voltage_offset) {
 		result = vega10_populate_ulv_state(hwmgr);
@@ -3701,10 +3704,22 @@ static void vega10_apply_dal_minimum_voltage_request(
 	return;
 }
 
+static int vega10_get_soc_index_for_max_uclk(struct pp_hwmgr *hwmgr)
+{
+	struct phm_ppt_v1_clock_voltage_dependency_table *vdd_dep_table_on_mclk;
+	struct phm_ppt_v2_information *table_info =
+			(struct phm_ppt_v2_information *)(hwmgr->pptable);
+
+	vdd_dep_table_on_mclk  = table_info->vdd_dep_on_mclk;
+
+	return vdd_dep_table_on_mclk->entries[NUM_UCLK_DPM_LEVELS - 1].vddInd + 1;
+}
+
 static int vega10_upload_dpm_bootup_level(struct pp_hwmgr *hwmgr)
 {
 	struct vega10_hwmgr *data =
 			(struct vega10_hwmgr *)(hwmgr->backend);
+	uint32_t socclk_idx;
 
 	vega10_apply_dal_minimum_voltage_request(hwmgr);
 
@@ -3725,13 +3740,22 @@ static int vega10_upload_dpm_bootup_level(struct pp_hwmgr *hwmgr)
 	if (!data->registry_data.mclk_dpm_key_disabled) {
 		if (data->smc_state_table.mem_boot_level !=
 				data->dpm_table.mem_table.dpm_state.soft_min_level) {
+			if (data->smc_state_table.mem_boot_level == NUM_UCLK_DPM_LEVELS - 1) {
+				socclk_idx = vega10_get_soc_index_for_max_uclk(hwmgr);
 				PP_ASSERT_WITH_CODE(!smum_send_msg_to_smc_with_parameter(
-				hwmgr->smumgr,
-				 PPSMC_MSG_SetSoftMinUclkByIndex,
-				data->smc_state_table.mem_boot_level),
-				"Failed to set soft min mclk index!",
-				return -EINVAL);
-
+							hwmgr->smumgr,
+						PPSMC_MSG_SetSoftMinSocclkByIndex,
+						socclk_idx),
+						"Failed to set soft min uclk index!",
+						return -EINVAL);
+			} else {
+				PP_ASSERT_WITH_CODE(!smum_send_msg_to_smc_with_parameter(
+						hwmgr->smumgr,
+						PPSMC_MSG_SetSoftMinUclkByIndex,
+						data->smc_state_table.mem_boot_level),
+						"Failed to set soft min uclk index!",
+						return -EINVAL);
+			}
 			data->dpm_table.mem_table.dpm_state.soft_min_level =
 					data->smc_state_table.mem_boot_level;
 		}
@@ -4138,7 +4162,7 @@ static int vega10_notify_smc_display_config_after_ps_adjustment(
 			pr_info("Attempt to set Hard Min for DCEFCLK Failed!");
 		}
 	} else {
-		pr_info("Cannot find requested DCEFCLK!");
+		pr_debug("Cannot find requested DCEFCLK!");
 	}
 
 	if (min_clocks.memoryClock != 0) {
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_powertune.c b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_powertune.c
index fbafc849ea71..e7fa67063cdc 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_powertune.c
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_powertune.c
@@ -543,7 +543,7 @@ static const struct vega10_didt_config_reg SEEDCCtrlForceStallConfig_Vega10[] =
  * ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
  */
 	/* SQ */
-	{   ixDIDT_SQ_EDC_CTRL,                DIDT_SQ_EDC_CTRL__EDC_EN_MASK,                       DIDT_SQ_EDC_CTRL__EDC_EN__SHIFT,                        0x0001 },
+	{   ixDIDT_SQ_EDC_CTRL,                DIDT_SQ_EDC_CTRL__EDC_EN_MASK,                       DIDT_SQ_EDC_CTRL__EDC_EN__SHIFT,                        0x0000 },
 	{   ixDIDT_SQ_EDC_CTRL,                DIDT_SQ_EDC_CTRL__EDC_SW_RST_MASK,                   DIDT_SQ_EDC_CTRL__EDC_SW_RST__SHIFT,                    0x0000 },
 	{   ixDIDT_SQ_EDC_CTRL,                DIDT_SQ_EDC_CTRL__EDC_CLK_EN_OVERRIDE_MASK,          DIDT_SQ_EDC_CTRL__EDC_CLK_EN_OVERRIDE__SHIFT,           0x0000 },
 	{   ixDIDT_SQ_EDC_CTRL,                DIDT_SQ_EDC_CTRL__EDC_FORCE_STALL_MASK,              DIDT_SQ_EDC_CTRL__EDC_FORCE_STALL__SHIFT,               0x0001 },
@@ -556,7 +556,7 @@ static const struct vega10_didt_config_reg SEEDCCtrlForceStallConfig_Vega10[] =
 	{   ixDIDT_SQ_EDC_CTRL,                DIDT_SQ_EDC_CTRL__SE_EDC_LEVEL_COMB_EN_MASK,         DIDT_SQ_EDC_CTRL__SE_EDC_LEVEL_COMB_EN__SHIFT,          0x0001 },
 
 	/* TD */
-	{   ixDIDT_TD_EDC_CTRL,                DIDT_TD_EDC_CTRL__EDC_EN_MASK,                       DIDT_TD_EDC_CTRL__EDC_EN__SHIFT,                        0x0001 },
+	{   ixDIDT_TD_EDC_CTRL,                DIDT_TD_EDC_CTRL__EDC_EN_MASK,                       DIDT_TD_EDC_CTRL__EDC_EN__SHIFT,                        0x0000 },
 	{   ixDIDT_TD_EDC_CTRL,                DIDT_TD_EDC_CTRL__EDC_SW_RST_MASK,                   DIDT_TD_EDC_CTRL__EDC_SW_RST__SHIFT,                    0x0000 },
 	{   ixDIDT_TD_EDC_CTRL,                DIDT_TD_EDC_CTRL__EDC_CLK_EN_OVERRIDE_MASK,          DIDT_TD_EDC_CTRL__EDC_CLK_EN_OVERRIDE__SHIFT,           0x0000 },
 	{   ixDIDT_TD_EDC_CTRL,                DIDT_TD_EDC_CTRL__EDC_FORCE_STALL_MASK,              DIDT_TD_EDC_CTRL__EDC_FORCE_STALL__SHIFT,               0x0001 },
@@ -1208,7 +1208,7 @@ static int vega10_enable_se_edc_force_stall_config(struct pp_hwmgr *hwmgr)
 	if (0 != result)
 		return result;
 
-	vega10_didt_set_mask(hwmgr, true);
+	vega10_didt_set_mask(hwmgr, false);
 
 	cgs_enter_safe_mode(hwmgr->device, false);
 
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_thermal.c b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_thermal.c
index e7ab8eb8a0cf..d44243441d28 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_thermal.c
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_thermal.c
@@ -321,10 +321,7 @@ int vega10_fan_ctrl_reset_fan_speed_to_default(struct pp_hwmgr *hwmgr)
 
 	if (phm_cap_enabled(hwmgr->platform_descriptor.platformCaps,
 			PHM_PlatformCaps_MicrocodeFanControl)) {
-		result = vega10_fan_ctrl_set_static_mode(hwmgr,
-				FDO_PWM_MODE_STATIC);
-		if (!result)
-			result = vega10_fan_ctrl_start_smc_fan_control(hwmgr);
+		result = vega10_fan_ctrl_start_smc_fan_control(hwmgr);
 	} else
 		result = vega10_fan_ctrl_set_default_mode(hwmgr);
 
@@ -633,7 +630,6 @@ int tf_vega10_thermal_start_smc_fan_control(struct pp_hwmgr *hwmgr,
 	if (phm_cap_enabled(hwmgr->platform_descriptor.platformCaps,
 			PHM_PlatformCaps_MicrocodeFanControl)) {
 		vega10_fan_ctrl_start_smc_fan_control(hwmgr);
-		vega10_fan_ctrl_set_static_mode(hwmgr, FDO_PWM_MODE_STATIC);
 	}
 
 	return 0;
diff --git a/drivers/gpu/drm/amd/powerplay/inc/hwmgr.h b/drivers/gpu/drm/amd/powerplay/inc/hwmgr.h
index 47e57bd2c36f..91b0105e8240 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/hwmgr.h
+++ b/drivers/gpu/drm/amd/powerplay/inc/hwmgr.h
@@ -128,6 +128,8 @@ struct phm_uvd_arbiter {
 	uint32_t dclk;
 	uint32_t vclk_ceiling;
 	uint32_t dclk_ceiling;
+	uint32_t vclk_soft_min;
+	uint32_t dclk_soft_min;
 };
 
 struct phm_vce_arbiter {
diff --git a/drivers/gpu/drm/amd/powerplay/inc/rv_ppsmc.h b/drivers/gpu/drm/amd/powerplay/inc/rv_ppsmc.h
index e0e106f1b23a..901c960cfe21 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/rv_ppsmc.h
+++ b/drivers/gpu/drm/amd/powerplay/inc/rv_ppsmc.h
@@ -66,7 +66,12 @@
 #define PPSMC_MSG_SetMinVddcrSocVoltage         0x22
 #define PPSMC_MSG_SetMinVideoFclkFreq           0x23
 #define PPSMC_MSG_SetMinDeepSleepDcefclk        0x24
-#define PPSMC_Message_Count                     0x25
+#define PPSMC_MSG_ForcePowerDownGfx             0x25
+#define PPSMC_MSG_SetPhyclkVoltageByFreq        0x26
+#define PPSMC_MSG_SetDppclkVoltageByFreq        0x27
+#define PPSMC_MSG_SetSoftMinVcn                 0x28
+#define PPSMC_Message_Count                     0x29
+
 
 typedef uint16_t PPSMC_Result;
 typedef int      PPSMC_Msg;
diff --git a/drivers/gpu/drm/arc/arcpgu_drv.c b/drivers/gpu/drm/arc/arcpgu_drv.c
index e3c13aa202b8..289eda54e5aa 100644
--- a/drivers/gpu/drm/arc/arcpgu_drv.c
+++ b/drivers/gpu/drm/arc/arcpgu_drv.c
@@ -31,7 +31,7 @@ static void arcpgu_fb_output_poll_changed(struct drm_device *dev)
 	drm_fbdev_cma_hotplug_event(arcpgu->fbdev);
 }
 
-static struct drm_mode_config_funcs arcpgu_drm_modecfg_funcs = {
+static const struct drm_mode_config_funcs arcpgu_drm_modecfg_funcs = {
 	.fb_create  = drm_fb_cma_create,
 	.output_poll_changed = arcpgu_fb_output_poll_changed,
 	.atomic_check = drm_atomic_helper_check,
diff --git a/drivers/gpu/drm/ast/ast_drv.c b/drivers/gpu/drm/ast/ast_drv.c
index 3022b39c00f3..69dab82a3771 100644
--- a/drivers/gpu/drm/ast/ast_drv.c
+++ b/drivers/gpu/drm/ast/ast_drv.c
@@ -209,7 +209,6 @@ static struct drm_driver driver = {
 	.gem_free_object_unlocked = ast_gem_free_object,
 	.dumb_create = ast_dumb_create,
 	.dumb_map_offset = ast_dumb_mmap_offset,
-	.dumb_destroy = drm_gem_dumb_destroy,
 
 };
 
diff --git a/drivers/gpu/drm/ast/ast_fb.c b/drivers/gpu/drm/ast/ast_fb.c
index 9052ebeae8d0..0cd827e11fa2 100644
--- a/drivers/gpu/drm/ast/ast_fb.c
+++ b/drivers/gpu/drm/ast/ast_fb.c
@@ -266,7 +266,7 @@ static void ast_fbdev_destroy(struct drm_device *dev,
 	drm_fb_helper_unregister_fbi(&afbdev->helper);
 
 	if (afb->obj) {
-		drm_gem_object_unreference_unlocked(afb->obj);
+		drm_gem_object_put_unlocked(afb->obj);
 		afb->obj = NULL;
 	}
 	drm_fb_helper_fini(&afbdev->helper);
diff --git a/drivers/gpu/drm/ast/ast_main.c b/drivers/gpu/drm/ast/ast_main.c
index 9a44cdec3bca..dac355812adc 100644
--- a/drivers/gpu/drm/ast/ast_main.c
+++ b/drivers/gpu/drm/ast/ast_main.c
@@ -387,7 +387,7 @@ static void ast_user_framebuffer_destroy(struct drm_framebuffer *fb)
 {
 	struct ast_framebuffer *ast_fb = to_ast_framebuffer(fb);
 
-	drm_gem_object_unreference_unlocked(ast_fb->obj);
+	drm_gem_object_put_unlocked(ast_fb->obj);
 	drm_framebuffer_cleanup(fb);
 	kfree(ast_fb);
 }
@@ -429,13 +429,13 @@ ast_user_framebuffer_create(struct drm_device *dev,
 
 	ast_fb = kzalloc(sizeof(*ast_fb), GFP_KERNEL);
 	if (!ast_fb) {
-		drm_gem_object_unreference_unlocked(obj);
+		drm_gem_object_put_unlocked(obj);
 		return ERR_PTR(-ENOMEM);
 	}
 
 	ret = ast_framebuffer_init(dev, ast_fb, mode_cmd, obj);
 	if (ret) {
-		drm_gem_object_unreference_unlocked(obj);
+		drm_gem_object_put_unlocked(obj);
 		kfree(ast_fb);
 		return ERR_PTR(ret);
 	}
@@ -628,7 +628,7 @@ int ast_dumb_create(struct drm_file *file,
 		return ret;
 
 	ret = drm_gem_handle_create(file, gobj, &handle);
-	drm_gem_object_unreference_unlocked(gobj);
+	drm_gem_object_put_unlocked(gobj);
 	if (ret)
 		return ret;
 
@@ -676,7 +676,7 @@ ast_dumb_mmap_offset(struct drm_file *file,
 	bo = gem_to_ast_bo(obj);
 	*offset = ast_bo_mmap_offset(bo);
 
-	drm_gem_object_unreference_unlocked(obj);
+	drm_gem_object_put_unlocked(obj);
 
 	return 0;
 
diff --git a/drivers/gpu/drm/ast/ast_mode.c b/drivers/gpu/drm/ast/ast_mode.c
index 43245229f437..6f3849ec0c1d 100644
--- a/drivers/gpu/drm/ast/ast_mode.c
+++ b/drivers/gpu/drm/ast/ast_mode.c
@@ -950,7 +950,7 @@ static void ast_cursor_fini(struct drm_device *dev)
 {
 	struct ast_private *ast = dev->dev_private;
 	ttm_bo_kunmap(&ast->cache_kmap);
-	drm_gem_object_unreference_unlocked(ast->cursor_cache);
+	drm_gem_object_put_unlocked(ast->cursor_cache);
 }
 
 int ast_mode_init(struct drm_device *dev)
@@ -1215,10 +1215,10 @@ static int ast_cursor_set(struct drm_crtc *crtc,
 
 	ast_show_cursor(crtc);
 
-	drm_gem_object_unreference_unlocked(obj);
+	drm_gem_object_put_unlocked(obj);
 	return 0;
 fail:
-	drm_gem_object_unreference_unlocked(obj);
+	drm_gem_object_put_unlocked(obj);
 	return ret;
 }
 
diff --git a/drivers/gpu/drm/bochs/bochs_drv.c b/drivers/gpu/drm/bochs/bochs_drv.c
index a1d28845da5f..7b20318483e4 100644
--- a/drivers/gpu/drm/bochs/bochs_drv.c
+++ b/drivers/gpu/drm/bochs/bochs_drv.c
@@ -93,7 +93,6 @@ static struct drm_driver bochs_driver = {
 	.gem_free_object_unlocked = bochs_gem_free_object,
 	.dumb_create            = bochs_dumb_create,
 	.dumb_map_offset        = bochs_dumb_mmap_offset,
-	.dumb_destroy           = drm_gem_dumb_destroy,
 };
 
 /* ---------------------------------------------------------------------- */
diff --git a/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c b/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c
index 682c090fa3ed..b2431aee7887 100644
--- a/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c
+++ b/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c
@@ -785,7 +785,7 @@ adv7511_connector_detect(struct drm_connector *connector, bool force)
 	return adv7511_detect(adv, connector);
 }
 
-static struct drm_connector_funcs adv7511_connector_funcs = {
+static const struct drm_connector_funcs adv7511_connector_funcs = {
 	.fill_modes = drm_helper_probe_single_connector_modes,
 	.detect = adv7511_connector_detect,
 	.destroy = drm_connector_cleanup,
@@ -856,7 +856,7 @@ static int adv7511_bridge_attach(struct drm_bridge *bridge)
 	return ret;
 }
 
-static struct drm_bridge_funcs adv7511_bridge_funcs = {
+static const struct drm_bridge_funcs adv7511_bridge_funcs = {
 	.enable = adv7511_bridge_enable,
 	.disable = adv7511_bridge_disable,
 	.mode_set = adv7511_bridge_mode_set,
diff --git a/drivers/gpu/drm/bridge/synopsys/dw-hdmi-ahb-audio.c b/drivers/gpu/drm/bridge/synopsys/dw-hdmi-ahb-audio.c
index 8f2d1379c880..cf3f0caf9c63 100644
--- a/drivers/gpu/drm/bridge/synopsys/dw-hdmi-ahb-audio.c
+++ b/drivers/gpu/drm/bridge/synopsys/dw-hdmi-ahb-audio.c
@@ -517,7 +517,7 @@ static snd_pcm_uframes_t dw_hdmi_pointer(struct snd_pcm_substream *substream)
 	return bytes_to_frames(runtime, dw->buf_offset);
 }
 
-static struct snd_pcm_ops snd_dw_hdmi_ops = {
+static const struct snd_pcm_ops snd_dw_hdmi_ops = {
 	.open = dw_hdmi_open,
 	.close = dw_hdmi_close,
 	.ioctl = snd_pcm_lib_ioctl,
diff --git a/drivers/gpu/drm/bridge/synopsys/dw-mipi-dsi.c b/drivers/gpu/drm/bridge/synopsys/dw-mipi-dsi.c
index 36f5ccbd1794..63c7a01b7053 100644
--- a/drivers/gpu/drm/bridge/synopsys/dw-mipi-dsi.c
+++ b/drivers/gpu/drm/bridge/synopsys/dw-mipi-dsi.c
@@ -811,7 +811,7 @@ static int dw_mipi_dsi_bridge_attach(struct drm_bridge *bridge)
 	return drm_bridge_attach(bridge->encoder, dsi->panel_bridge, bridge);
 }
 
-static struct drm_bridge_funcs dw_mipi_dsi_bridge_funcs = {
+static const struct drm_bridge_funcs dw_mipi_dsi_bridge_funcs = {
 	.mode_set     = dw_mipi_dsi_bridge_mode_set,
 	.enable	      = dw_mipi_dsi_bridge_enable,
 	.post_disable = dw_mipi_dsi_bridge_post_disable,
diff --git a/drivers/gpu/drm/bridge/tc358767.c b/drivers/gpu/drm/bridge/tc358767.c
index 503252d6a74d..8571cfd877c5 100644
--- a/drivers/gpu/drm/bridge/tc358767.c
+++ b/drivers/gpu/drm/bridge/tc358767.c
@@ -1254,7 +1254,7 @@ static int tc_probe(struct i2c_client *client, const struct i2c_device_id *id)
 
 	/* port@2 is the output port */
 	ret = drm_of_find_panel_or_bridge(dev->of_node, 2, 0, &tc->panel, NULL);
-	if (ret)
+	if (ret && ret != -ENODEV)
 		return ret;
 
 	/* Shut down GPIO is optional */
diff --git a/drivers/gpu/drm/cirrus/cirrus_drv.c b/drivers/gpu/drm/cirrus/cirrus_drv.c
index 910c300f5c37..69c4e352dd78 100644
--- a/drivers/gpu/drm/cirrus/cirrus_drv.c
+++ b/drivers/gpu/drm/cirrus/cirrus_drv.c
@@ -142,7 +142,6 @@ static struct drm_driver driver = {
 	.gem_free_object_unlocked = cirrus_gem_free_object,
 	.dumb_create = cirrus_dumb_create,
 	.dumb_map_offset = cirrus_dumb_mmap_offset,
-	.dumb_destroy = drm_gem_dumb_destroy,
 };
 
 static const struct dev_pm_ops cirrus_pm_ops = {
diff --git a/drivers/gpu/drm/cirrus/cirrus_fbdev.c b/drivers/gpu/drm/cirrus/cirrus_fbdev.c
index 0f6815f35ad2..32fbfba2c623 100644
--- a/drivers/gpu/drm/cirrus/cirrus_fbdev.c
+++ b/drivers/gpu/drm/cirrus/cirrus_fbdev.c
@@ -251,7 +251,7 @@ static int cirrus_fbdev_destroy(struct drm_device *dev,
 	drm_fb_helper_unregister_fbi(&gfbdev->helper);
 
 	if (gfb->obj) {
-		drm_gem_object_unreference_unlocked(gfb->obj);
+		drm_gem_object_put_unlocked(gfb->obj);
 		gfb->obj = NULL;
 	}
 
diff --git a/drivers/gpu/drm/cirrus/cirrus_main.c b/drivers/gpu/drm/cirrus/cirrus_main.c
index e7fc95f63dca..b5f528543956 100644
--- a/drivers/gpu/drm/cirrus/cirrus_main.c
+++ b/drivers/gpu/drm/cirrus/cirrus_main.c
@@ -18,7 +18,7 @@ static void cirrus_user_framebuffer_destroy(struct drm_framebuffer *fb)
 {
 	struct cirrus_framebuffer *cirrus_fb = to_cirrus_framebuffer(fb);
 
-	drm_gem_object_unreference_unlocked(cirrus_fb->obj);
+	drm_gem_object_put_unlocked(cirrus_fb->obj);
 	drm_framebuffer_cleanup(fb);
 	kfree(fb);
 }
@@ -67,13 +67,13 @@ cirrus_user_framebuffer_create(struct drm_device *dev,
 
 	cirrus_fb = kzalloc(sizeof(*cirrus_fb), GFP_KERNEL);
 	if (!cirrus_fb) {
-		drm_gem_object_unreference_unlocked(obj);
+		drm_gem_object_put_unlocked(obj);
 		return ERR_PTR(-ENOMEM);
 	}
 
 	ret = cirrus_framebuffer_init(dev, cirrus_fb, mode_cmd, obj);
 	if (ret) {
-		drm_gem_object_unreference_unlocked(obj);
+		drm_gem_object_put_unlocked(obj);
 		kfree(cirrus_fb);
 		return ERR_PTR(ret);
 	}
@@ -261,7 +261,7 @@ int cirrus_dumb_create(struct drm_file *file,
 		return ret;
 
 	ret = drm_gem_handle_create(file, gobj, &handle);
-	drm_gem_object_unreference_unlocked(gobj);
+	drm_gem_object_put_unlocked(gobj);
 	if (ret)
 		return ret;
 
@@ -310,7 +310,7 @@ cirrus_dumb_mmap_offset(struct drm_file *file,
 	bo = gem_to_cirrus_bo(obj);
 	*offset = cirrus_bo_mmap_offset(bo);
 
-	drm_gem_object_unreference_unlocked(obj);
+	drm_gem_object_put_unlocked(obj);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/drm_dp_helper.c b/drivers/gpu/drm/drm_dp_helper.c
index 213fb837e1c4..08af8d6b844b 100644
--- a/drivers/gpu/drm/drm_dp_helper.c
+++ b/drivers/gpu/drm/drm_dp_helper.c
@@ -544,7 +544,7 @@ void drm_dp_downstream_debug(struct seq_file *m,
 				 DP_DETAILED_CAP_INFO_AVAILABLE;
 	int clk;
 	int bpc;
-	char id[6];
+	char id[7];
 	int len;
 	uint8_t rev[2];
 	int type = port_cap[0] & DP_DS_PORT_TYPE_MASK;
@@ -583,6 +583,7 @@ void drm_dp_downstream_debug(struct seq_file *m,
 		seq_puts(m, "\t\tType: N/A\n");
 	}
 
+	memset(id, 0, sizeof(id));
 	drm_dp_downstream_id(aux, id);
 	seq_printf(m, "\t\tID: %s\n", id);
 
@@ -591,7 +592,7 @@ void drm_dp_downstream_debug(struct seq_file *m,
 		seq_printf(m, "\t\tHW: %d.%d\n",
 			   (rev[0] & 0xf0) >> 4, rev[0] & 0xf);
 
-	len = drm_dp_dpcd_read(aux, DP_BRANCH_SW_REV, &rev, 2);
+	len = drm_dp_dpcd_read(aux, DP_BRANCH_SW_REV, rev, 2);
 	if (len > 0)
 		seq_printf(m, "\t\tSW: %d.%d\n", rev[0], rev[1]);
 
diff --git a/drivers/gpu/drm/drm_drv.c b/drivers/gpu/drm/drm_drv.c
index 2ed2d919beae..be38ac7050d4 100644
--- a/drivers/gpu/drm/drm_drv.c
+++ b/drivers/gpu/drm/drm_drv.c
@@ -291,7 +291,7 @@ struct drm_minor *drm_minor_acquire(unsigned int minor_id)
 
 	if (!minor) {
 		return ERR_PTR(-ENODEV);
-	} else if (drm_device_is_unplugged(minor->dev)) {
+	} else if (drm_dev_is_unplugged(minor->dev)) {
 		drm_dev_unref(minor->dev);
 		return ERR_PTR(-ENODEV);
 	}
@@ -364,26 +364,32 @@ void drm_put_dev(struct drm_device *dev)
 }
 EXPORT_SYMBOL(drm_put_dev);
 
-void drm_unplug_dev(struct drm_device *dev)
+static void drm_device_set_unplugged(struct drm_device *dev)
 {
-	/* for a USB device */
-	if (drm_core_check_feature(dev, DRIVER_MODESET))
-		drm_modeset_unregister_all(dev);
+	smp_wmb();
+	atomic_set(&dev->unplugged, 1);
+}
 
-	drm_minor_unregister(dev, DRM_MINOR_PRIMARY);
-	drm_minor_unregister(dev, DRM_MINOR_RENDER);
-	drm_minor_unregister(dev, DRM_MINOR_CONTROL);
+/**
+ * drm_dev_unplug - unplug a DRM device
+ * @dev: DRM device
+ *
+ * This unplugs a hotpluggable DRM device, which makes it inaccessible to
+ * userspace operations. Entry-points can use drm_dev_is_unplugged(). This
+ * essentially unregisters the device like drm_dev_unregister(), but can be
+ * called while there are still open users of @dev.
+ */
+void drm_dev_unplug(struct drm_device *dev)
+{
+	drm_dev_unregister(dev);
 
 	mutex_lock(&drm_global_mutex);
-
 	drm_device_set_unplugged(dev);
-
-	if (dev->open_count == 0) {
-		drm_put_dev(dev);
-	}
+	if (dev->open_count == 0)
+		drm_dev_unref(dev);
 	mutex_unlock(&drm_global_mutex);
 }
-EXPORT_SYMBOL(drm_unplug_dev);
+EXPORT_SYMBOL(drm_dev_unplug);
 
 /*
  * DRM internal mount
@@ -835,6 +841,9 @@ EXPORT_SYMBOL(drm_dev_register);
  * drm_dev_register() but does not deallocate the device. The caller must call
  * drm_dev_unref() to drop their final reference.
  *
+ * A special form of unregistering for hotpluggable devices is drm_dev_unplug(),
+ * which can be called while there are still open users of @dev.
+ *
  * This should be called first in the device teardown code to make sure
  * userspace can't access the device instance any more.
  */
@@ -842,7 +851,8 @@ void drm_dev_unregister(struct drm_device *dev)
 {
 	struct drm_map_list *r_list, *list_temp;
 
-	drm_lastclose(dev);
+	if (drm_core_check_feature(dev, DRIVER_LEGACY))
+		drm_lastclose(dev);
 
 	dev->registered = false;
 
diff --git a/drivers/gpu/drm/drm_fb_cma_helper.c b/drivers/gpu/drm/drm_fb_cma_helper.c
index ade319d10e70..f2ee88363015 100644
--- a/drivers/gpu/drm/drm_fb_cma_helper.c
+++ b/drivers/gpu/drm/drm_fb_cma_helper.c
@@ -18,27 +18,17 @@
  */
 
 #include <drm/drmP.h>
-#include <drm/drm_atomic.h>
-#include <drm/drm_crtc.h>
 #include <drm/drm_fb_helper.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_framebuffer.h>
 #include <drm/drm_gem_cma_helper.h>
+#include <drm/drm_gem_framebuffer_helper.h>
 #include <drm/drm_fb_cma_helper.h>
-#include <linux/dma-buf.h>
-#include <linux/dma-mapping.h>
 #include <linux/module.h>
-#include <linux/reservation.h>
 
 #define DEFAULT_FBDEFIO_DELAY_MS 50
 
-struct drm_fb_cma {
-	struct drm_framebuffer		fb;
-	struct drm_gem_cma_object	*obj[4];
-};
-
 struct drm_fbdev_cma {
 	struct drm_fb_helper	fb_helper;
-	struct drm_fb_cma	*fb;
 	const struct drm_framebuffer_funcs *fb_funcs;
 };
 
@@ -90,69 +80,19 @@ static inline struct drm_fbdev_cma *to_fbdev_cma(struct drm_fb_helper *helper)
 	return container_of(helper, struct drm_fbdev_cma, fb_helper);
 }
 
-static inline struct drm_fb_cma *to_fb_cma(struct drm_framebuffer *fb)
-{
-	return container_of(fb, struct drm_fb_cma, fb);
-}
-
 void drm_fb_cma_destroy(struct drm_framebuffer *fb)
 {
-	struct drm_fb_cma *fb_cma = to_fb_cma(fb);
-	int i;
-
-	for (i = 0; i < 4; i++) {
-		if (fb_cma->obj[i])
-			drm_gem_object_put_unlocked(&fb_cma->obj[i]->base);
-	}
-
-	drm_framebuffer_cleanup(fb);
-	kfree(fb_cma);
+	drm_gem_fb_destroy(fb);
 }
 EXPORT_SYMBOL(drm_fb_cma_destroy);
 
 int drm_fb_cma_create_handle(struct drm_framebuffer *fb,
 	struct drm_file *file_priv, unsigned int *handle)
 {
-	struct drm_fb_cma *fb_cma = to_fb_cma(fb);
-
-	return drm_gem_handle_create(file_priv,
-			&fb_cma->obj[0]->base, handle);
+	return drm_gem_fb_create_handle(fb, file_priv, handle);
 }
 EXPORT_SYMBOL(drm_fb_cma_create_handle);
 
-static struct drm_framebuffer_funcs drm_fb_cma_funcs = {
-	.destroy	= drm_fb_cma_destroy,
-	.create_handle	= drm_fb_cma_create_handle,
-};
-
-static struct drm_fb_cma *drm_fb_cma_alloc(struct drm_device *dev,
-	const struct drm_mode_fb_cmd2 *mode_cmd,
-	struct drm_gem_cma_object **obj,
-	unsigned int num_planes, const struct drm_framebuffer_funcs *funcs)
-{
-	struct drm_fb_cma *fb_cma;
-	int ret;
-	int i;
-
-	fb_cma = kzalloc(sizeof(*fb_cma), GFP_KERNEL);
-	if (!fb_cma)
-		return ERR_PTR(-ENOMEM);
-
-	drm_helper_mode_fill_fb_struct(dev, &fb_cma->fb, mode_cmd);
-
-	for (i = 0; i < num_planes; i++)
-		fb_cma->obj[i] = obj[i];
-
-	ret = drm_framebuffer_init(dev, &fb_cma->fb, funcs);
-	if (ret) {
-		dev_err(dev->dev, "Failed to initialize framebuffer: %d\n", ret);
-		kfree(fb_cma);
-		return ERR_PTR(ret);
-	}
-
-	return fb_cma;
-}
-
 /**
  * drm_fb_cma_create_with_funcs() - helper function for the
  *                                  &drm_mode_config_funcs.fb_create
@@ -170,53 +110,7 @@ struct drm_framebuffer *drm_fb_cma_create_with_funcs(struct drm_device *dev,
 	struct drm_file *file_priv, const struct drm_mode_fb_cmd2 *mode_cmd,
 	const struct drm_framebuffer_funcs *funcs)
 {
-	const struct drm_format_info *info;
-	struct drm_fb_cma *fb_cma;
-	struct drm_gem_cma_object *objs[4];
-	struct drm_gem_object *obj;
-	int ret;
-	int i;
-
-	info = drm_get_format_info(dev, mode_cmd);
-	if (!info)
-		return ERR_PTR(-EINVAL);
-
-	for (i = 0; i < info->num_planes; i++) {
-		unsigned int width = mode_cmd->width / (i ? info->hsub : 1);
-		unsigned int height = mode_cmd->height / (i ? info->vsub : 1);
-		unsigned int min_size;
-
-		obj = drm_gem_object_lookup(file_priv, mode_cmd->handles[i]);
-		if (!obj) {
-			dev_err(dev->dev, "Failed to lookup GEM object\n");
-			ret = -ENOENT;
-			goto err_gem_object_put;
-		}
-
-		min_size = (height - 1) * mode_cmd->pitches[i]
-			 + width * info->cpp[i]
-			 + mode_cmd->offsets[i];
-
-		if (obj->size < min_size) {
-			drm_gem_object_put_unlocked(obj);
-			ret = -EINVAL;
-			goto err_gem_object_put;
-		}
-		objs[i] = to_drm_gem_cma_obj(obj);
-	}
-
-	fb_cma = drm_fb_cma_alloc(dev, mode_cmd, objs, i, funcs);
-	if (IS_ERR(fb_cma)) {
-		ret = PTR_ERR(fb_cma);
-		goto err_gem_object_put;
-	}
-
-	return &fb_cma->fb;
-
-err_gem_object_put:
-	for (i--; i >= 0; i--)
-		drm_gem_object_put_unlocked(&objs[i]->base);
-	return ERR_PTR(ret);
+	return drm_gem_fb_create_with_funcs(dev, file_priv, mode_cmd, funcs);
 }
 EXPORT_SYMBOL_GPL(drm_fb_cma_create_with_funcs);
 
@@ -233,8 +127,7 @@ EXPORT_SYMBOL_GPL(drm_fb_cma_create_with_funcs);
 struct drm_framebuffer *drm_fb_cma_create(struct drm_device *dev,
 	struct drm_file *file_priv, const struct drm_mode_fb_cmd2 *mode_cmd)
 {
-	return drm_fb_cma_create_with_funcs(dev, file_priv, mode_cmd,
-					    &drm_fb_cma_funcs);
+	return drm_gem_fb_create(dev, file_priv, mode_cmd);
 }
 EXPORT_SYMBOL_GPL(drm_fb_cma_create);
 
@@ -250,12 +143,13 @@ EXPORT_SYMBOL_GPL(drm_fb_cma_create);
 struct drm_gem_cma_object *drm_fb_cma_get_gem_obj(struct drm_framebuffer *fb,
 						  unsigned int plane)
 {
-	struct drm_fb_cma *fb_cma = to_fb_cma(fb);
+	struct drm_gem_object *gem;
 
-	if (plane >= 4)
+	gem = drm_gem_fb_get_obj(fb, plane);
+	if (!gem)
 		return NULL;
 
-	return fb_cma->obj[plane];
+	return to_drm_gem_cma_obj(gem);
 }
 EXPORT_SYMBOL_GPL(drm_fb_cma_get_gem_obj);
 
@@ -272,13 +166,14 @@ dma_addr_t drm_fb_cma_get_gem_addr(struct drm_framebuffer *fb,
 				   struct drm_plane_state *state,
 				   unsigned int plane)
 {
-	struct drm_fb_cma *fb_cma = to_fb_cma(fb);
+	struct drm_gem_cma_object *obj;
 	dma_addr_t paddr;
 
-	if (plane >= 4)
+	obj = drm_fb_cma_get_gem_obj(fb, plane);
+	if (!obj)
 		return 0;
 
-	paddr = fb_cma->obj[plane]->paddr + fb->offsets[plane];
+	paddr = obj->paddr + fb->offsets[plane];
 	paddr += fb->format->cpp[plane] * (state->src_x >> 16);
 	paddr += fb->pitches[plane] * (state->src_y >> 16);
 
@@ -302,26 +197,13 @@ EXPORT_SYMBOL_GPL(drm_fb_cma_get_gem_addr);
 int drm_fb_cma_prepare_fb(struct drm_plane *plane,
 			  struct drm_plane_state *state)
 {
-	struct dma_buf *dma_buf;
-	struct dma_fence *fence;
-
-	if ((plane->state->fb == state->fb) || !state->fb)
-		return 0;
-
-	dma_buf = drm_fb_cma_get_gem_obj(state->fb, 0)->base.dma_buf;
-	if (dma_buf) {
-		fence = reservation_object_get_excl_rcu(dma_buf->resv);
-		drm_atomic_set_fence_for_plane(state, fence);
-	}
-
-	return 0;
+	return drm_gem_fb_prepare_fb(plane, state);
 }
 EXPORT_SYMBOL_GPL(drm_fb_cma_prepare_fb);
 
 #ifdef CONFIG_DEBUG_FS
 static void drm_fb_cma_describe(struct drm_framebuffer *fb, struct seq_file *m)
 {
-	struct drm_fb_cma *fb_cma = to_fb_cma(fb);
 	int i;
 
 	seq_printf(m, "fb: %dx%d@%4.4s\n", fb->width, fb->height,
@@ -330,7 +212,7 @@ static void drm_fb_cma_describe(struct drm_framebuffer *fb, struct seq_file *m)
 	for (i = 0; i < fb->format->num_planes; i++) {
 		seq_printf(m, "   %d: offset=%d pitch=%d, obj: ",
 				i, fb->offsets[i], fb->pitches[i]);
-		drm_gem_cma_describe(fb_cma->obj[i], m);
+		drm_gem_cma_describe(drm_fb_cma_get_gem_obj(fb, i), m);
 	}
 }
 
@@ -431,7 +313,6 @@ drm_fbdev_cma_create(struct drm_fb_helper *helper,
 	struct drm_fb_helper_surface_size *sizes)
 {
 	struct drm_fbdev_cma *fbdev_cma = to_fbdev_cma(helper);
-	struct drm_mode_fb_cmd2 mode_cmd = { 0 };
 	struct drm_device *dev = helper->dev;
 	struct drm_gem_cma_object *obj;
 	struct drm_framebuffer *fb;
@@ -446,14 +327,7 @@ drm_fbdev_cma_create(struct drm_fb_helper *helper,
 			sizes->surface_bpp);
 
 	bytes_per_pixel = DIV_ROUND_UP(sizes->surface_bpp, 8);
-
-	mode_cmd.width = sizes->surface_width;
-	mode_cmd.height = sizes->surface_height;
-	mode_cmd.pitches[0] = sizes->surface_width * bytes_per_pixel;
-	mode_cmd.pixel_format = drm_mode_legacy_fb_format(sizes->surface_bpp,
-		sizes->surface_depth);
-
-	size = mode_cmd.pitches[0] * mode_cmd.height;
+	size = sizes->surface_width * sizes->surface_height * bytes_per_pixel;
 	obj = drm_gem_cma_create(dev, size);
 	if (IS_ERR(obj))
 		return -ENOMEM;
@@ -464,15 +338,14 @@ drm_fbdev_cma_create(struct drm_fb_helper *helper,
 		goto err_gem_free_object;
 	}
 
-	fbdev_cma->fb = drm_fb_cma_alloc(dev, &mode_cmd, &obj, 1,
-					 fbdev_cma->fb_funcs);
-	if (IS_ERR(fbdev_cma->fb)) {
+	fb = drm_gem_fbdev_fb_create(dev, sizes, 0, &obj->base,
+				     fbdev_cma->fb_funcs);
+	if (IS_ERR(fb)) {
 		dev_err(dev->dev, "Failed to allocate DRM framebuffer.\n");
-		ret = PTR_ERR(fbdev_cma->fb);
+		ret = PTR_ERR(fb);
 		goto err_fb_info_destroy;
 	}
 
-	fb = &fbdev_cma->fb->fb;
 	helper->fb = fb;
 
 	fbi->par = helper;
@@ -500,7 +373,7 @@ drm_fbdev_cma_create(struct drm_fb_helper *helper,
 	return 0;
 
 err_cma_destroy:
-	drm_framebuffer_remove(&fbdev_cma->fb->fb);
+	drm_framebuffer_remove(fb);
 err_fb_info_destroy:
 	drm_fb_helper_fini(helper);
 err_gem_free_object:
@@ -570,6 +443,11 @@ err_free:
 }
 EXPORT_SYMBOL_GPL(drm_fbdev_cma_init_with_funcs);
 
+static const struct drm_framebuffer_funcs drm_fb_cma_funcs = {
+	.destroy	= drm_gem_fb_destroy,
+	.create_handle	= drm_gem_fb_create_handle,
+};
+
 /**
  * drm_fbdev_cma_init() - Allocate and initializes a drm_fbdev_cma struct
  * @dev: DRM device
@@ -597,8 +475,8 @@ void drm_fbdev_cma_fini(struct drm_fbdev_cma *fbdev_cma)
 	if (fbdev_cma->fb_helper.fbdev)
 		drm_fbdev_cma_defio_fini(fbdev_cma->fb_helper.fbdev);
 
-	if (fbdev_cma->fb)
-		drm_framebuffer_remove(&fbdev_cma->fb->fb);
+	if (fbdev_cma->fb_helper.fb)
+		drm_framebuffer_remove(fbdev_cma->fb_helper.fb);
 
 	drm_fb_helper_fini(&fbdev_cma->fb_helper);
 	kfree(fbdev_cma);
diff --git a/drivers/gpu/drm/drm_file.c b/drivers/gpu/drm/drm_file.c
index 59b75a974357..b3c6e997ccdb 100644
--- a/drivers/gpu/drm/drm_file.c
+++ b/drivers/gpu/drm/drm_file.c
@@ -436,7 +436,7 @@ int drm_release(struct inode *inode, struct file *filp)
 
 	if (!--dev->open_count) {
 		drm_lastclose(dev);
-		if (drm_device_is_unplugged(dev))
+		if (drm_dev_is_unplugged(dev))
 			drm_put_dev(dev);
 	}
 	mutex_unlock(&drm_global_mutex);
diff --git a/drivers/gpu/drm/drm_gem.c b/drivers/gpu/drm/drm_gem.c
index a8d396bed6a4..ad4e9cfe48a2 100644
--- a/drivers/gpu/drm/drm_gem.c
+++ b/drivers/gpu/drm/drm_gem.c
@@ -1001,7 +1001,7 @@ int drm_gem_mmap(struct file *filp, struct vm_area_struct *vma)
 	struct drm_vma_offset_node *node;
 	int ret;
 
-	if (drm_device_is_unplugged(dev))
+	if (drm_dev_is_unplugged(dev))
 		return -ENODEV;
 
 	drm_vma_offset_lock_lookup(dev->vma_offset_manager);
diff --git a/drivers/gpu/drm/drm_gem_cma_helper.c b/drivers/gpu/drm/drm_gem_cma_helper.c
index 275ab872b34f..373e33f22be4 100644
--- a/drivers/gpu/drm/drm_gem_cma_helper.c
+++ b/drivers/gpu/drm/drm_gem_cma_helper.c
@@ -264,41 +264,6 @@ int drm_gem_cma_dumb_create(struct drm_file *file_priv,
 }
 EXPORT_SYMBOL_GPL(drm_gem_cma_dumb_create);
 
-/**
- * drm_gem_cma_dumb_map_offset - return the fake mmap offset for a CMA GEM
- *     object
- * @file_priv: DRM file-private structure containing the GEM object
- * @drm: DRM device
- * @handle: GEM object handle
- * @offset: return location for the fake mmap offset
- *
- * This function look up an object by its handle and returns the fake mmap
- * offset associated with it. Drivers using the CMA helpers should set this
- * as their &drm_driver.dumb_map_offset callback.
- *
- * Returns:
- * 0 on success or a negative error code on failure.
- */
-int drm_gem_cma_dumb_map_offset(struct drm_file *file_priv,
-				struct drm_device *drm, u32 handle,
-				u64 *offset)
-{
-	struct drm_gem_object *gem_obj;
-
-	gem_obj = drm_gem_object_lookup(file_priv, handle);
-	if (!gem_obj) {
-		dev_err(drm->dev, "failed to lookup GEM object\n");
-		return -EINVAL;
-	}
-
-	*offset = drm_vma_node_offset_addr(&gem_obj->vma_node);
-
-	drm_gem_object_put_unlocked(gem_obj);
-
-	return 0;
-}
-EXPORT_SYMBOL_GPL(drm_gem_cma_dumb_map_offset);
-
 const struct vm_operations_struct drm_gem_cma_vm_ops = {
 	.open = drm_gem_vm_open,
 	.close = drm_gem_vm_close,
@@ -390,7 +355,7 @@ unsigned long drm_gem_cma_get_unmapped_area(struct file *filp,
 	struct drm_device *dev = priv->minor->dev;
 	struct drm_vma_offset_node *node;
 
-	if (drm_device_is_unplugged(dev))
+	if (drm_dev_is_unplugged(dev))
 		return -ENODEV;
 
 	drm_vma_offset_lock_lookup(dev->vma_offset_manager);
diff --git a/drivers/gpu/drm/drm_gem_framebuffer_helper.c b/drivers/gpu/drm/drm_gem_framebuffer_helper.c
new file mode 100644
index 000000000000..d54a083dc5dd
--- /dev/null
+++ b/drivers/gpu/drm/drm_gem_framebuffer_helper.c
@@ -0,0 +1,283 @@
+/*
+ * drm gem framebuffer helper functions
+ *
+ * Copyright (C) 2017 Noralf Trønnes
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#include <linux/dma-buf.h>
+#include <linux/dma-fence.h>
+#include <linux/reservation.h>
+#include <linux/slab.h>
+
+#include <drm/drmP.h>
+#include <drm/drm_atomic.h>
+#include <drm/drm_fb_helper.h>
+#include <drm/drm_fourcc.h>
+#include <drm/drm_framebuffer.h>
+#include <drm/drm_gem.h>
+#include <drm/drm_gem_framebuffer_helper.h>
+#include <drm/drm_modeset_helper.h>
+
+/**
+ * DOC: overview
+ *
+ * This library provides helpers for drivers that don't subclass
+ * &drm_framebuffer and and use &drm_gem_object for their backing storage.
+ *
+ * Drivers without additional needs to validate framebuffers can simply use
+ * drm_gem_fb_create() and everything is wired up automatically. But all
+ * parts can be used individually.
+ */
+
+/**
+ * drm_gem_fb_get_obj() - Get GEM object for framebuffer
+ * @fb: The framebuffer
+ * @plane: Which plane
+ *
+ * Returns the GEM object for given framebuffer.
+ */
+struct drm_gem_object *drm_gem_fb_get_obj(struct drm_framebuffer *fb,
+					  unsigned int plane)
+{
+	if (plane >= 4)
+		return NULL;
+
+	return fb->obj[plane];
+}
+EXPORT_SYMBOL_GPL(drm_gem_fb_get_obj);
+
+static struct drm_framebuffer *
+drm_gem_fb_alloc(struct drm_device *dev,
+		 const struct drm_mode_fb_cmd2 *mode_cmd,
+		 struct drm_gem_object **obj, unsigned int num_planes,
+		 const struct drm_framebuffer_funcs *funcs)
+{
+	struct drm_framebuffer *fb;
+	int ret, i;
+
+	fb = kzalloc(sizeof(*fb), GFP_KERNEL);
+	if (!fb)
+		return ERR_PTR(-ENOMEM);
+
+	drm_helper_mode_fill_fb_struct(dev, fb, mode_cmd);
+
+	for (i = 0; i < num_planes; i++)
+		fb->obj[i] = obj[i];
+
+	ret = drm_framebuffer_init(dev, fb, funcs);
+	if (ret) {
+		DRM_DEV_ERROR(dev->dev, "Failed to init framebuffer: %d\n",
+			      ret);
+		kfree(fb);
+		return ERR_PTR(ret);
+	}
+
+	return fb;
+}
+
+/**
+ * drm_gem_fb_destroy - Free GEM backed framebuffer
+ * @fb: DRM framebuffer
+ *
+ * Frees a GEM backed framebuffer with its backing buffer(s) and the structure
+ * itself. Drivers can use this as their &drm_framebuffer_funcs->destroy
+ * callback.
+ */
+void drm_gem_fb_destroy(struct drm_framebuffer *fb)
+{
+	int i;
+
+	for (i = 0; i < 4; i++)
+		drm_gem_object_put_unlocked(fb->obj[i]);
+
+	drm_framebuffer_cleanup(fb);
+	kfree(fb);
+}
+EXPORT_SYMBOL(drm_gem_fb_destroy);
+
+/**
+ * drm_gem_fb_create_handle - Create handle for GEM backed framebuffer
+ * @fb: DRM framebuffer
+ * @file: drm file
+ * @handle: handle created
+ *
+ * Drivers can use this as their &drm_framebuffer_funcs->create_handle
+ * callback.
+ *
+ * Returns:
+ * 0 on success or a negative error code on failure.
+ */
+int drm_gem_fb_create_handle(struct drm_framebuffer *fb, struct drm_file *file,
+			     unsigned int *handle)
+{
+	return drm_gem_handle_create(file, fb->obj[0], handle);
+}
+EXPORT_SYMBOL(drm_gem_fb_create_handle);
+
+/**
+ * drm_gem_fb_create_with_funcs() - helper function for the
+ *                                  &drm_mode_config_funcs.fb_create
+ *                                  callback
+ * @dev: DRM device
+ * @file: drm file for the ioctl call
+ * @mode_cmd: metadata from the userspace fb creation request
+ * @funcs: vtable to be used for the new framebuffer object
+ *
+ * This can be used to set &drm_framebuffer_funcs for drivers that need the
+ * &drm_framebuffer_funcs.dirty callback. Use drm_gem_fb_create() if you don't
+ * need to change &drm_framebuffer_funcs.
+ * The function does buffer size validation.
+ */
+struct drm_framebuffer *
+drm_gem_fb_create_with_funcs(struct drm_device *dev, struct drm_file *file,
+			     const struct drm_mode_fb_cmd2 *mode_cmd,
+			     const struct drm_framebuffer_funcs *funcs)
+{
+	const struct drm_format_info *info;
+	struct drm_gem_object *objs[4];
+	struct drm_framebuffer *fb;
+	int ret, i;
+
+	info = drm_get_format_info(dev, mode_cmd);
+	if (!info)
+		return ERR_PTR(-EINVAL);
+
+	for (i = 0; i < info->num_planes; i++) {
+		unsigned int width = mode_cmd->width / (i ? info->hsub : 1);
+		unsigned int height = mode_cmd->height / (i ? info->vsub : 1);
+		unsigned int min_size;
+
+		objs[i] = drm_gem_object_lookup(file, mode_cmd->handles[i]);
+		if (!objs[i]) {
+			DRM_DEV_ERROR(dev->dev, "Failed to lookup GEM\n");
+			ret = -ENOENT;
+			goto err_gem_object_put;
+		}
+
+		min_size = (height - 1) * mode_cmd->pitches[i]
+			 + width * info->cpp[i]
+			 + mode_cmd->offsets[i];
+
+		if (objs[i]->size < min_size) {
+			drm_gem_object_put_unlocked(objs[i]);
+			ret = -EINVAL;
+			goto err_gem_object_put;
+		}
+	}
+
+	fb = drm_gem_fb_alloc(dev, mode_cmd, objs, i, funcs);
+	if (IS_ERR(fb)) {
+		ret = PTR_ERR(fb);
+		goto err_gem_object_put;
+	}
+
+	return fb;
+
+err_gem_object_put:
+	for (i--; i >= 0; i--)
+		drm_gem_object_put_unlocked(objs[i]);
+
+	return ERR_PTR(ret);
+}
+EXPORT_SYMBOL_GPL(drm_gem_fb_create_with_funcs);
+
+static const struct drm_framebuffer_funcs drm_gem_fb_funcs = {
+	.destroy	= drm_gem_fb_destroy,
+	.create_handle	= drm_gem_fb_create_handle,
+};
+
+/**
+ * drm_gem_fb_create() - &drm_mode_config_funcs.fb_create callback function
+ * @dev: DRM device
+ * @file: drm file for the ioctl call
+ * @mode_cmd: metadata from the userspace fb creation request
+ *
+ * If your hardware has special alignment or pitch requirements these should be
+ * checked before calling this function. The function does buffer size
+ * validation. Use drm_gem_fb_create_with_funcs() if you need to set
+ * &drm_framebuffer_funcs.dirty.
+ */
+struct drm_framebuffer *
+drm_gem_fb_create(struct drm_device *dev, struct drm_file *file,
+		  const struct drm_mode_fb_cmd2 *mode_cmd)
+{
+	return drm_gem_fb_create_with_funcs(dev, file, mode_cmd,
+					    &drm_gem_fb_funcs);
+}
+EXPORT_SYMBOL_GPL(drm_gem_fb_create);
+
+/**
+ * drm_gem_fb_prepare_fb() - Prepare gem framebuffer
+ * @plane: Which plane
+ * @state: Plane state attach fence to
+ *
+ * This can be used as the &drm_plane_helper_funcs.prepare_fb hook.
+ *
+ * This function checks if the plane FB has an dma-buf attached, extracts
+ * the exclusive fence and attaches it to plane state for the atomic helper
+ * to wait on.
+ *
+ * There is no need for &drm_plane_helper_funcs.cleanup_fb hook for simple
+ * gem based framebuffer drivers which have their buffers always pinned in
+ * memory.
+ */
+int drm_gem_fb_prepare_fb(struct drm_plane *plane,
+			  struct drm_plane_state *state)
+{
+	struct dma_buf *dma_buf;
+	struct dma_fence *fence;
+
+	if ((plane->state->fb == state->fb) || !state->fb)
+		return 0;
+
+	dma_buf = drm_gem_fb_get_obj(state->fb, 0)->dma_buf;
+	if (dma_buf) {
+		fence = reservation_object_get_excl_rcu(dma_buf->resv);
+		drm_atomic_set_fence_for_plane(state, fence);
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(drm_gem_fb_prepare_fb);
+
+/**
+ * drm_gem_fbdev_fb_create - Create a drm_framebuffer for fbdev emulation
+ * @dev: DRM device
+ * @sizes: fbdev size description
+ * @pitch_align: optional pitch alignment
+ * @obj: GEM object backing the framebuffer
+ * @funcs: vtable to be used for the new framebuffer object
+ *
+ * This function creates a framebuffer for use with fbdev emulation.
+ *
+ * Returns:
+ * Pointer to a drm_framebuffer on success or an error pointer on failure.
+ */
+struct drm_framebuffer *
+drm_gem_fbdev_fb_create(struct drm_device *dev,
+			struct drm_fb_helper_surface_size *sizes,
+			unsigned int pitch_align, struct drm_gem_object *obj,
+			const struct drm_framebuffer_funcs *funcs)
+{
+	struct drm_mode_fb_cmd2 mode_cmd = { 0 };
+
+	mode_cmd.width = sizes->surface_width;
+	mode_cmd.height = sizes->surface_height;
+	mode_cmd.pitches[0] = sizes->surface_width *
+			      DIV_ROUND_UP(sizes->surface_bpp, 8);
+	if (pitch_align)
+		mode_cmd.pitches[0] = roundup(mode_cmd.pitches[0],
+					      pitch_align);
+	mode_cmd.pixel_format = drm_mode_legacy_fb_format(sizes->surface_bpp,
+							sizes->surface_depth);
+	if (obj->size < mode_cmd.pitches[0] * mode_cmd.height)
+		return ERR_PTR(-EINVAL);
+
+	return drm_gem_fb_alloc(dev, &mode_cmd, &obj, 1, funcs);
+}
+EXPORT_SYMBOL(drm_gem_fbdev_fb_create);
diff --git a/drivers/gpu/drm/drm_internal.h b/drivers/gpu/drm/drm_internal.h
index 4e906b82a170..fbc3f308fa19 100644
--- a/drivers/gpu/drm/drm_internal.h
+++ b/drivers/gpu/drm/drm_internal.h
@@ -167,3 +167,9 @@ int drm_syncobj_handle_to_fd_ioctl(struct drm_device *dev, void *data,
 				   struct drm_file *file_private);
 int drm_syncobj_fd_to_handle_ioctl(struct drm_device *dev, void *data,
 				   struct drm_file *file_private);
+int drm_syncobj_wait_ioctl(struct drm_device *dev, void *data,
+			   struct drm_file *file_private);
+int drm_syncobj_reset_ioctl(struct drm_device *dev, void *data,
+			    struct drm_file *file_private);
+int drm_syncobj_signal_ioctl(struct drm_device *dev, void *data,
+			     struct drm_file *file_private);
diff --git a/drivers/gpu/drm/drm_ioctl.c b/drivers/gpu/drm/drm_ioctl.c
index 8bfeb32f8a10..a9ae6dd2d593 100644
--- a/drivers/gpu/drm/drm_ioctl.c
+++ b/drivers/gpu/drm/drm_ioctl.c
@@ -657,6 +657,12 @@ static const struct drm_ioctl_desc drm_ioctls[] = {
 		      DRM_UNLOCKED|DRM_RENDER_ALLOW),
 	DRM_IOCTL_DEF(DRM_IOCTL_SYNCOBJ_FD_TO_HANDLE, drm_syncobj_fd_to_handle_ioctl,
 		      DRM_UNLOCKED|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF(DRM_IOCTL_SYNCOBJ_WAIT, drm_syncobj_wait_ioctl,
+		      DRM_UNLOCKED|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF(DRM_IOCTL_SYNCOBJ_RESET, drm_syncobj_reset_ioctl,
+		      DRM_UNLOCKED|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF(DRM_IOCTL_SYNCOBJ_SIGNAL, drm_syncobj_signal_ioctl,
+		      DRM_UNLOCKED|DRM_RENDER_ALLOW),
 };
 
 #define DRM_CORE_IOCTL_COUNT	ARRAY_SIZE( drm_ioctls )
@@ -716,7 +722,7 @@ long drm_ioctl_kernel(struct file *file, drm_ioctl_t *func, void *kdata,
 	struct drm_device *dev = file_priv->minor->dev;
 	int retcode;
 
-	if (drm_device_is_unplugged(dev))
+	if (drm_dev_is_unplugged(dev))
 		return -ENODEV;
 
 	retcode = drm_ioctl_permit(flags, file_priv);
@@ -765,7 +771,7 @@ long drm_ioctl(struct file *filp,
 
 	dev = file_priv->minor->dev;
 
-	if (drm_device_is_unplugged(dev))
+	if (drm_dev_is_unplugged(dev))
 		return -ENODEV;
 
 	is_driver_ioctl = nr >= DRM_COMMAND_BASE && nr < DRM_COMMAND_END;
diff --git a/drivers/gpu/drm/drm_plane.c b/drivers/gpu/drm/drm_plane.c
index 5c14beee52ff..85ab1eec73e5 100644
--- a/drivers/gpu/drm/drm_plane.c
+++ b/drivers/gpu/drm/drm_plane.c
@@ -126,7 +126,7 @@ static int create_in_format_blob(struct drm_device *dev, struct drm_plane *plane
 							       plane->format_types[j],
 							       plane->modifiers[i])) {
 
-				mod->formats |= 1 << j;
+				mod->formats |= 1ULL << j;
 			}
 		}
 
diff --git a/drivers/gpu/drm/drm_syncobj.c b/drivers/gpu/drm/drm_syncobj.c
index a5b38a80a99a..0422b8c2c2e7 100644
--- a/drivers/gpu/drm/drm_syncobj.c
+++ b/drivers/gpu/drm/drm_syncobj.c
@@ -1,5 +1,7 @@
 /*
  * Copyright 2017 Red Hat
+ * Parts ported from amdgpu (fence wait code).
+ * Copyright 2016 Advanced Micro Devices, Inc.
  *
  * Permission is hereby granted, free of charge, to any person obtaining a
  * copy of this software and associated documentation files (the "Software"),
@@ -31,6 +33,9 @@
  * that contain an optional fence. The fence can be updated with a new
  * fence, or be NULL.
  *
+ * syncobj's can be waited upon, where it will wait for the underlying
+ * fence.
+ *
  * syncobj's can be export to fd's and back, these fd's are opaque and
  * have no other use case, except passing the syncobj between processes.
  *
@@ -46,6 +51,7 @@
 #include <linux/fs.h>
 #include <linux/anon_inodes.h>
 #include <linux/sync_file.h>
+#include <linux/sched/signal.h>
 
 #include "drm_internal.h"
 #include <drm/drm_syncobj.h>
@@ -75,6 +81,75 @@ struct drm_syncobj *drm_syncobj_find(struct drm_file *file_private,
 }
 EXPORT_SYMBOL(drm_syncobj_find);
 
+static void drm_syncobj_add_callback_locked(struct drm_syncobj *syncobj,
+					    struct drm_syncobj_cb *cb,
+					    drm_syncobj_func_t func)
+{
+	cb->func = func;
+	list_add_tail(&cb->node, &syncobj->cb_list);
+}
+
+static int drm_syncobj_fence_get_or_add_callback(struct drm_syncobj *syncobj,
+						 struct dma_fence **fence,
+						 struct drm_syncobj_cb *cb,
+						 drm_syncobj_func_t func)
+{
+	int ret;
+
+	*fence = drm_syncobj_fence_get(syncobj);
+	if (*fence)
+		return 1;
+
+	spin_lock(&syncobj->lock);
+	/* We've already tried once to get a fence and failed.  Now that we
+	 * have the lock, try one more time just to be sure we don't add a
+	 * callback when a fence has already been set.
+	 */
+	if (syncobj->fence) {
+		*fence = dma_fence_get(syncobj->fence);
+		ret = 1;
+	} else {
+		*fence = NULL;
+		drm_syncobj_add_callback_locked(syncobj, cb, func);
+		ret = 0;
+	}
+	spin_unlock(&syncobj->lock);
+
+	return ret;
+}
+
+/**
+ * drm_syncobj_add_callback - adds a callback to syncobj::cb_list
+ * @syncobj: Sync object to which to add the callback
+ * @cb: Callback to add
+ * @func: Func to use when initializing the drm_syncobj_cb struct
+ *
+ * This adds a callback to be called next time the fence is replaced
+ */
+void drm_syncobj_add_callback(struct drm_syncobj *syncobj,
+			      struct drm_syncobj_cb *cb,
+			      drm_syncobj_func_t func)
+{
+	spin_lock(&syncobj->lock);
+	drm_syncobj_add_callback_locked(syncobj, cb, func);
+	spin_unlock(&syncobj->lock);
+}
+EXPORT_SYMBOL(drm_syncobj_add_callback);
+
+/**
+ * drm_syncobj_add_callback - removes a callback to syncobj::cb_list
+ * @syncobj: Sync object from which to remove the callback
+ * @cb: Callback to remove
+ */
+void drm_syncobj_remove_callback(struct drm_syncobj *syncobj,
+				 struct drm_syncobj_cb *cb)
+{
+	spin_lock(&syncobj->lock);
+	list_del_init(&cb->node);
+	spin_unlock(&syncobj->lock);
+}
+EXPORT_SYMBOL(drm_syncobj_remove_callback);
+
 /**
  * drm_syncobj_replace_fence - replace fence in a sync object.
  * @syncobj: Sync object to replace fence in
@@ -86,18 +161,75 @@ void drm_syncobj_replace_fence(struct drm_syncobj *syncobj,
 			       struct dma_fence *fence)
 {
 	struct dma_fence *old_fence;
+	struct drm_syncobj_cb *cur, *tmp;
 
 	if (fence)
 		dma_fence_get(fence);
-	old_fence = xchg(&syncobj->fence, fence);
+
+	spin_lock(&syncobj->lock);
+
+	old_fence = syncobj->fence;
+	syncobj->fence = fence;
+
+	if (fence != old_fence) {
+		list_for_each_entry_safe(cur, tmp, &syncobj->cb_list, node) {
+			list_del_init(&cur->node);
+			cur->func(syncobj, cur);
+		}
+	}
+
+	spin_unlock(&syncobj->lock);
 
 	dma_fence_put(old_fence);
 }
 EXPORT_SYMBOL(drm_syncobj_replace_fence);
 
-int drm_syncobj_fence_get(struct drm_file *file_private,
-			  u32 handle,
-			  struct dma_fence **fence)
+struct drm_syncobj_null_fence {
+	struct dma_fence base;
+	spinlock_t lock;
+};
+
+static const char *drm_syncobj_null_fence_get_name(struct dma_fence *fence)
+{
+        return "syncobjnull";
+}
+
+static bool drm_syncobj_null_fence_enable_signaling(struct dma_fence *fence)
+{
+    dma_fence_enable_sw_signaling(fence);
+    return !dma_fence_is_signaled(fence);
+}
+
+static const struct dma_fence_ops drm_syncobj_null_fence_ops = {
+	.get_driver_name = drm_syncobj_null_fence_get_name,
+	.get_timeline_name = drm_syncobj_null_fence_get_name,
+	.enable_signaling = drm_syncobj_null_fence_enable_signaling,
+	.wait = dma_fence_default_wait,
+	.release = NULL,
+};
+
+static int drm_syncobj_assign_null_handle(struct drm_syncobj *syncobj)
+{
+	struct drm_syncobj_null_fence *fence;
+	fence = kzalloc(sizeof(*fence), GFP_KERNEL);
+	if (fence == NULL)
+		return -ENOMEM;
+
+	spin_lock_init(&fence->lock);
+	dma_fence_init(&fence->base, &drm_syncobj_null_fence_ops,
+		       &fence->lock, 0, 0);
+	dma_fence_signal(&fence->base);
+
+	drm_syncobj_replace_fence(syncobj, &fence->base);
+
+	dma_fence_put(&fence->base);
+
+	return 0;
+}
+
+int drm_syncobj_find_fence(struct drm_file *file_private,
+			   u32 handle,
+			   struct dma_fence **fence)
 {
 	struct drm_syncobj *syncobj = drm_syncobj_find(file_private, handle);
 	int ret = 0;
@@ -105,14 +237,14 @@ int drm_syncobj_fence_get(struct drm_file *file_private,
 	if (!syncobj)
 		return -ENOENT;
 
-	*fence = dma_fence_get(syncobj->fence);
+	*fence = drm_syncobj_fence_get(syncobj);
 	if (!*fence) {
 		ret = -EINVAL;
 	}
 	drm_syncobj_put(syncobj);
 	return ret;
 }
-EXPORT_SYMBOL(drm_syncobj_fence_get);
+EXPORT_SYMBOL(drm_syncobj_find_fence);
 
 /**
  * drm_syncobj_free - free a sync object.
@@ -125,13 +257,13 @@ void drm_syncobj_free(struct kref *kref)
 	struct drm_syncobj *syncobj = container_of(kref,
 						   struct drm_syncobj,
 						   refcount);
-	dma_fence_put(syncobj->fence);
+	drm_syncobj_replace_fence(syncobj, NULL);
 	kfree(syncobj);
 }
 EXPORT_SYMBOL(drm_syncobj_free);
 
 static int drm_syncobj_create(struct drm_file *file_private,
-			      u32 *handle)
+			      u32 *handle, uint32_t flags)
 {
 	int ret;
 	struct drm_syncobj *syncobj;
@@ -141,6 +273,16 @@ static int drm_syncobj_create(struct drm_file *file_private,
 		return -ENOMEM;
 
 	kref_init(&syncobj->refcount);
+	INIT_LIST_HEAD(&syncobj->cb_list);
+	spin_lock_init(&syncobj->lock);
+
+	if (flags & DRM_SYNCOBJ_CREATE_SIGNALED) {
+		ret = drm_syncobj_assign_null_handle(syncobj);
+		if (ret < 0) {
+			drm_syncobj_put(syncobj);
+			return ret;
+		}
+	}
 
 	idr_preload(GFP_KERNEL);
 	spin_lock(&file_private->syncobj_table_lock);
@@ -307,7 +449,7 @@ int drm_syncobj_export_sync_file(struct drm_file *file_private,
 	if (fd < 0)
 		return fd;
 
-	ret = drm_syncobj_fence_get(file_private, handle, &fence);
+	ret = drm_syncobj_find_fence(file_private, handle, &fence);
 	if (ret)
 		goto err_put_fd;
 
@@ -377,11 +519,11 @@ drm_syncobj_create_ioctl(struct drm_device *dev, void *data,
 		return -ENODEV;
 
 	/* no valid flags yet */
-	if (args->flags)
+	if (args->flags & ~DRM_SYNCOBJ_CREATE_SIGNALED)
 		return -EINVAL;
 
 	return drm_syncobj_create(file_private,
-				  &args->handle);
+				  &args->handle, args->flags);
 }
 
 int
@@ -447,3 +589,368 @@ drm_syncobj_fd_to_handle_ioctl(struct drm_device *dev, void *data,
 	return drm_syncobj_fd_to_handle(file_private, args->fd,
 					&args->handle);
 }
+
+struct syncobj_wait_entry {
+	struct task_struct *task;
+	struct dma_fence *fence;
+	struct dma_fence_cb fence_cb;
+	struct drm_syncobj_cb syncobj_cb;
+};
+
+static void syncobj_wait_fence_func(struct dma_fence *fence,
+				    struct dma_fence_cb *cb)
+{
+	struct syncobj_wait_entry *wait =
+		container_of(cb, struct syncobj_wait_entry, fence_cb);
+
+	wake_up_process(wait->task);
+}
+
+static void syncobj_wait_syncobj_func(struct drm_syncobj *syncobj,
+				      struct drm_syncobj_cb *cb)
+{
+	struct syncobj_wait_entry *wait =
+		container_of(cb, struct syncobj_wait_entry, syncobj_cb);
+
+	/* This happens inside the syncobj lock */
+	wait->fence = dma_fence_get(syncobj->fence);
+	wake_up_process(wait->task);
+}
+
+static signed long drm_syncobj_array_wait_timeout(struct drm_syncobj **syncobjs,
+						  uint32_t count,
+						  uint32_t flags,
+						  signed long timeout,
+						  uint32_t *idx)
+{
+	struct syncobj_wait_entry *entries;
+	struct dma_fence *fence;
+	signed long ret;
+	uint32_t signaled_count, i;
+
+	entries = kcalloc(count, sizeof(*entries), GFP_KERNEL);
+	if (!entries)
+		return -ENOMEM;
+
+	/* Walk the list of sync objects and initialize entries.  We do
+	 * this up-front so that we can properly return -EINVAL if there is
+	 * a syncobj with a missing fence and then never have the chance of
+	 * returning -EINVAL again.
+	 */
+	signaled_count = 0;
+	for (i = 0; i < count; ++i) {
+		entries[i].task = current;
+		entries[i].fence = drm_syncobj_fence_get(syncobjs[i]);
+		if (!entries[i].fence) {
+			if (flags & DRM_SYNCOBJ_WAIT_FLAGS_WAIT_FOR_SUBMIT) {
+				continue;
+			} else {
+				ret = -EINVAL;
+				goto cleanup_entries;
+			}
+		}
+
+		if (dma_fence_is_signaled(entries[i].fence)) {
+			if (signaled_count == 0 && idx)
+				*idx = i;
+			signaled_count++;
+		}
+	}
+
+	/* Initialize ret to the max of timeout and 1.  That way, the
+	 * default return value indicates a successful wait and not a
+	 * timeout.
+	 */
+	ret = max_t(signed long, timeout, 1);
+
+	if (signaled_count == count ||
+	    (signaled_count > 0 &&
+	     !(flags & DRM_SYNCOBJ_WAIT_FLAGS_WAIT_ALL)))
+		goto cleanup_entries;
+
+	/* There's a very annoying laxness in the dma_fence API here, in
+	 * that backends are not required to automatically report when a
+	 * fence is signaled prior to fence->ops->enable_signaling() being
+	 * called.  So here if we fail to match signaled_count, we need to
+	 * fallthough and try a 0 timeout wait!
+	 */
+
+	if (flags & DRM_SYNCOBJ_WAIT_FLAGS_WAIT_FOR_SUBMIT) {
+		for (i = 0; i < count; ++i) {
+			drm_syncobj_fence_get_or_add_callback(syncobjs[i],
+							      &entries[i].fence,
+							      &entries[i].syncobj_cb,
+							      syncobj_wait_syncobj_func);
+		}
+	}
+
+	do {
+		set_current_state(TASK_INTERRUPTIBLE);
+
+		signaled_count = 0;
+		for (i = 0; i < count; ++i) {
+			fence = entries[i].fence;
+			if (!fence)
+				continue;
+
+			if (dma_fence_is_signaled(fence) ||
+			    (!entries[i].fence_cb.func &&
+			     dma_fence_add_callback(fence,
+						    &entries[i].fence_cb,
+						    syncobj_wait_fence_func))) {
+				/* The fence has been signaled */
+				if (flags & DRM_SYNCOBJ_WAIT_FLAGS_WAIT_ALL) {
+					signaled_count++;
+				} else {
+					if (idx)
+						*idx = i;
+					goto done_waiting;
+				}
+			}
+		}
+
+		if (signaled_count == count)
+			goto done_waiting;
+
+		if (timeout == 0) {
+			/* If we are doing a 0 timeout wait and we got
+			 * here, then we just timed out.
+			 */
+			ret = 0;
+			goto done_waiting;
+		}
+
+		ret = schedule_timeout(ret);
+
+		if (ret > 0 && signal_pending(current))
+			ret = -ERESTARTSYS;
+	} while (ret > 0);
+
+done_waiting:
+	__set_current_state(TASK_RUNNING);
+
+cleanup_entries:
+	for (i = 0; i < count; ++i) {
+		if (entries[i].syncobj_cb.func)
+			drm_syncobj_remove_callback(syncobjs[i],
+						    &entries[i].syncobj_cb);
+		if (entries[i].fence_cb.func)
+			dma_fence_remove_callback(entries[i].fence,
+						  &entries[i].fence_cb);
+		dma_fence_put(entries[i].fence);
+	}
+	kfree(entries);
+
+	return ret;
+}
+
+/**
+ * drm_timeout_abs_to_jiffies - calculate jiffies timeout from absolute value
+ *
+ * @timeout_nsec: timeout nsec component in ns, 0 for poll
+ *
+ * Calculate the timeout in jiffies from an absolute time in sec/nsec.
+ */
+static signed long drm_timeout_abs_to_jiffies(int64_t timeout_nsec)
+{
+	ktime_t abs_timeout, now;
+	u64 timeout_ns, timeout_jiffies64;
+
+	/* make 0 timeout means poll - absolute 0 doesn't seem valid */
+	if (timeout_nsec == 0)
+		return 0;
+
+	abs_timeout = ns_to_ktime(timeout_nsec);
+	now = ktime_get();
+
+	if (!ktime_after(abs_timeout, now))
+		return 0;
+
+	timeout_ns = ktime_to_ns(ktime_sub(abs_timeout, now));
+
+	timeout_jiffies64 = nsecs_to_jiffies64(timeout_ns);
+	/*  clamp timeout to avoid infinite timeout */
+	if (timeout_jiffies64 >= MAX_SCHEDULE_TIMEOUT - 1)
+		return MAX_SCHEDULE_TIMEOUT - 1;
+
+	return timeout_jiffies64 + 1;
+}
+
+static int drm_syncobj_array_wait(struct drm_device *dev,
+				  struct drm_file *file_private,
+				  struct drm_syncobj_wait *wait,
+				  struct drm_syncobj **syncobjs)
+{
+	signed long timeout = drm_timeout_abs_to_jiffies(wait->timeout_nsec);
+	signed long ret = 0;
+	uint32_t first = ~0;
+
+	ret = drm_syncobj_array_wait_timeout(syncobjs,
+					     wait->count_handles,
+					     wait->flags,
+					     timeout, &first);
+	if (ret < 0)
+		return ret;
+
+	wait->first_signaled = first;
+	if (ret == 0)
+		return -ETIME;
+	return 0;
+}
+
+static int drm_syncobj_array_find(struct drm_file *file_private,
+				  void *user_handles, uint32_t count_handles,
+				  struct drm_syncobj ***syncobjs_out)
+{
+	uint32_t i, *handles;
+	struct drm_syncobj **syncobjs;
+	int ret;
+
+	handles = kmalloc_array(count_handles, sizeof(*handles), GFP_KERNEL);
+	if (handles == NULL)
+		return -ENOMEM;
+
+	if (copy_from_user(handles, user_handles,
+			   sizeof(uint32_t) * count_handles)) {
+		ret = -EFAULT;
+		goto err_free_handles;
+	}
+
+	syncobjs = kmalloc_array(count_handles, sizeof(*syncobjs), GFP_KERNEL);
+	if (syncobjs == NULL) {
+		ret = -ENOMEM;
+		goto err_free_handles;
+	}
+
+	for (i = 0; i < count_handles; i++) {
+		syncobjs[i] = drm_syncobj_find(file_private, handles[i]);
+		if (!syncobjs[i]) {
+			ret = -ENOENT;
+			goto err_put_syncobjs;
+		}
+	}
+
+	kfree(handles);
+	*syncobjs_out = syncobjs;
+	return 0;
+
+err_put_syncobjs:
+	while (i-- > 0)
+		drm_syncobj_put(syncobjs[i]);
+	kfree(syncobjs);
+err_free_handles:
+	kfree(handles);
+
+	return ret;
+}
+
+static void drm_syncobj_array_free(struct drm_syncobj **syncobjs,
+				   uint32_t count)
+{
+	uint32_t i;
+	for (i = 0; i < count; i++)
+		drm_syncobj_put(syncobjs[i]);
+	kfree(syncobjs);
+}
+
+int
+drm_syncobj_wait_ioctl(struct drm_device *dev, void *data,
+		       struct drm_file *file_private)
+{
+	struct drm_syncobj_wait *args = data;
+	struct drm_syncobj **syncobjs;
+	int ret = 0;
+
+	if (!drm_core_check_feature(dev, DRIVER_SYNCOBJ))
+		return -ENODEV;
+
+	if (args->flags & ~(DRM_SYNCOBJ_WAIT_FLAGS_WAIT_ALL |
+			    DRM_SYNCOBJ_WAIT_FLAGS_WAIT_FOR_SUBMIT))
+		return -EINVAL;
+
+	if (args->count_handles == 0)
+		return -EINVAL;
+
+	ret = drm_syncobj_array_find(file_private,
+				     u64_to_user_ptr(args->handles),
+				     args->count_handles,
+				     &syncobjs);
+	if (ret < 0)
+		return ret;
+
+	ret = drm_syncobj_array_wait(dev, file_private,
+				     args, syncobjs);
+
+	drm_syncobj_array_free(syncobjs, args->count_handles);
+
+	return ret;
+}
+
+int
+drm_syncobj_reset_ioctl(struct drm_device *dev, void *data,
+			struct drm_file *file_private)
+{
+	struct drm_syncobj_array *args = data;
+	struct drm_syncobj **syncobjs;
+	uint32_t i;
+	int ret;
+
+	if (!drm_core_check_feature(dev, DRIVER_SYNCOBJ))
+		return -ENODEV;
+
+	if (args->pad != 0)
+		return -EINVAL;
+
+	if (args->count_handles == 0)
+		return -EINVAL;
+
+	ret = drm_syncobj_array_find(file_private,
+				     u64_to_user_ptr(args->handles),
+				     args->count_handles,
+				     &syncobjs);
+	if (ret < 0)
+		return ret;
+
+	for (i = 0; i < args->count_handles; i++)
+		drm_syncobj_replace_fence(syncobjs[i], NULL);
+
+	drm_syncobj_array_free(syncobjs, args->count_handles);
+
+	return 0;
+}
+
+int
+drm_syncobj_signal_ioctl(struct drm_device *dev, void *data,
+			 struct drm_file *file_private)
+{
+	struct drm_syncobj_array *args = data;
+	struct drm_syncobj **syncobjs;
+	uint32_t i;
+	int ret;
+
+	if (!drm_core_check_feature(dev, DRIVER_SYNCOBJ))
+		return -ENODEV;
+
+	if (args->pad != 0)
+		return -EINVAL;
+
+	if (args->count_handles == 0)
+		return -EINVAL;
+
+	ret = drm_syncobj_array_find(file_private,
+				     u64_to_user_ptr(args->handles),
+				     args->count_handles,
+				     &syncobjs);
+	if (ret < 0)
+		return ret;
+
+	for (i = 0; i < args->count_handles; i++) {
+		ret = drm_syncobj_assign_null_handle(syncobjs[i]);
+		if (ret < 0)
+			break;
+	}
+
+	drm_syncobj_array_free(syncobjs, args->count_handles);
+
+	return ret;
+}
diff --git a/drivers/gpu/drm/drm_vm.c b/drivers/gpu/drm/drm_vm.c
index 1170b3209a12..13a59ed2afbc 100644
--- a/drivers/gpu/drm/drm_vm.c
+++ b/drivers/gpu/drm/drm_vm.c
@@ -631,7 +631,7 @@ int drm_legacy_mmap(struct file *filp, struct vm_area_struct *vma)
 	struct drm_device *dev = priv->minor->dev;
 	int ret;
 
-	if (drm_device_is_unplugged(dev))
+	if (drm_dev_is_unplugged(dev))
 		return -ENODEV;
 
 	mutex_lock(&dev->struct_mutex);
diff --git a/drivers/gpu/drm/etnaviv/Kconfig b/drivers/gpu/drm/etnaviv/Kconfig
index 71cee4e9fefb..38b477b5fbf9 100644
--- a/drivers/gpu/drm/etnaviv/Kconfig
+++ b/drivers/gpu/drm/etnaviv/Kconfig
@@ -10,6 +10,8 @@ config DRM_ETNAVIV
 	select IOMMU_API
 	select IOMMU_SUPPORT
 	select WANT_DEV_COREDUMP
+	select CMA if HAVE_DMA_CONTIGUOUS
+	select DMA_CMA if HAVE_DMA_CONTIGUOUS
 	help
 	  DRM driver for Vivante GPUs.
 
diff --git a/drivers/gpu/drm/etnaviv/etnaviv_drv.c b/drivers/gpu/drm/etnaviv/etnaviv_drv.c
index 91e17aeee1da..2cb4773823c2 100644
--- a/drivers/gpu/drm/etnaviv/etnaviv_drv.c
+++ b/drivers/gpu/drm/etnaviv/etnaviv_drv.c
@@ -316,7 +316,7 @@ static int etnaviv_ioctl_gem_cpu_prep(struct drm_device *dev, void *data,
 
 	ret = etnaviv_gem_cpu_prep(obj, args->op, &TS(args->timeout));
 
-	drm_gem_object_unreference_unlocked(obj);
+	drm_gem_object_put_unlocked(obj);
 
 	return ret;
 }
@@ -337,7 +337,7 @@ static int etnaviv_ioctl_gem_cpu_fini(struct drm_device *dev, void *data,
 
 	ret = etnaviv_gem_cpu_fini(obj);
 
-	drm_gem_object_unreference_unlocked(obj);
+	drm_gem_object_put_unlocked(obj);
 
 	return ret;
 }
@@ -357,7 +357,7 @@ static int etnaviv_ioctl_gem_info(struct drm_device *dev, void *data,
 		return -ENOENT;
 
 	ret = etnaviv_gem_mmap_offset(obj, &args->offset);
-	drm_gem_object_unreference_unlocked(obj);
+	drm_gem_object_put_unlocked(obj);
 
 	return ret;
 }
@@ -446,7 +446,7 @@ static int etnaviv_ioctl_gem_wait(struct drm_device *dev, void *data,
 
 	ret = etnaviv_gem_wait_bo(gpu, obj, timeout);
 
-	drm_gem_object_unreference_unlocked(obj);
+	drm_gem_object_put_unlocked(obj);
 
 	return ret;
 }
diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gem.c b/drivers/gpu/drm/etnaviv/etnaviv_gem.c
index 9a3bea738330..5a634594a6ce 100644
--- a/drivers/gpu/drm/etnaviv/etnaviv_gem.c
+++ b/drivers/gpu/drm/etnaviv/etnaviv_gem.c
@@ -68,7 +68,7 @@ static int etnaviv_gem_shmem_get_pages(struct etnaviv_gem_object *etnaviv_obj)
 	struct page **p = drm_gem_get_pages(&etnaviv_obj->base);
 
 	if (IS_ERR(p)) {
-		dev_err(dev->dev, "could not get pages: %ld\n", PTR_ERR(p));
+		dev_dbg(dev->dev, "could not get pages: %ld\n", PTR_ERR(p));
 		return PTR_ERR(p);
 	}
 
@@ -265,7 +265,7 @@ void etnaviv_gem_mapping_reference(struct etnaviv_vram_mapping *mapping)
 {
 	struct etnaviv_gem_object *etnaviv_obj = mapping->object;
 
-	drm_gem_object_reference(&etnaviv_obj->base);
+	drm_gem_object_get(&etnaviv_obj->base);
 
 	mutex_lock(&etnaviv_obj->lock);
 	WARN_ON(mapping->use == 0);
@@ -282,7 +282,7 @@ void etnaviv_gem_mapping_unreference(struct etnaviv_vram_mapping *mapping)
 	mapping->use -= 1;
 	mutex_unlock(&etnaviv_obj->lock);
 
-	drm_gem_object_unreference_unlocked(&etnaviv_obj->base);
+	drm_gem_object_put_unlocked(&etnaviv_obj->base);
 }
 
 struct etnaviv_vram_mapping *etnaviv_gem_mapping_get(
@@ -358,7 +358,7 @@ out:
 		return ERR_PTR(ret);
 
 	/* Take a reference on the object */
-	drm_gem_object_reference(obj);
+	drm_gem_object_get(obj);
 	return mapping;
 }
 
@@ -413,6 +413,16 @@ int etnaviv_gem_cpu_prep(struct drm_gem_object *obj, u32 op,
 	bool write = !!(op & ETNA_PREP_WRITE);
 	int ret;
 
+	if (!etnaviv_obj->sgt) {
+		void *ret;
+
+		mutex_lock(&etnaviv_obj->lock);
+		ret = etnaviv_gem_get_pages(etnaviv_obj);
+		mutex_unlock(&etnaviv_obj->lock);
+		if (IS_ERR(ret))
+			return PTR_ERR(ret);
+	}
+
 	if (op & ETNA_PREP_NOSYNC) {
 		if (!reservation_object_test_signaled_rcu(etnaviv_obj->resv,
 							  write))
@@ -427,16 +437,6 @@ int etnaviv_gem_cpu_prep(struct drm_gem_object *obj, u32 op,
 	}
 
 	if (etnaviv_obj->flags & ETNA_BO_CACHED) {
-		if (!etnaviv_obj->sgt) {
-			void *ret;
-
-			mutex_lock(&etnaviv_obj->lock);
-			ret = etnaviv_gem_get_pages(etnaviv_obj);
-			mutex_unlock(&etnaviv_obj->lock);
-			if (IS_ERR(ret))
-				return PTR_ERR(ret);
-		}
-
 		dma_sync_sg_for_cpu(dev->dev, etnaviv_obj->sgt->sgl,
 				    etnaviv_obj->sgt->nents,
 				    etnaviv_op_to_dma_dir(op));
@@ -662,7 +662,8 @@ static struct drm_gem_object *__etnaviv_gem_new(struct drm_device *dev,
 		 * going to pin these pages.
 		 */
 		mapping = obj->filp->f_mapping;
-		mapping_set_gfp_mask(mapping, GFP_HIGHUSER);
+		mapping_set_gfp_mask(mapping, GFP_HIGHUSER |
+				     __GFP_RETRY_MAYFAIL | __GFP_NOWARN);
 	}
 
 	if (ret)
@@ -671,7 +672,7 @@ static struct drm_gem_object *__etnaviv_gem_new(struct drm_device *dev,
 	return obj;
 
 fail:
-	drm_gem_object_unreference_unlocked(obj);
+	drm_gem_object_put_unlocked(obj);
 	return ERR_PTR(ret);
 }
 
@@ -688,14 +689,14 @@ int etnaviv_gem_new_handle(struct drm_device *dev, struct drm_file *file,
 
 	ret = etnaviv_gem_obj_add(dev, obj);
 	if (ret < 0) {
-		drm_gem_object_unreference_unlocked(obj);
+		drm_gem_object_put_unlocked(obj);
 		return ret;
 	}
 
 	ret = drm_gem_handle_create(file, obj, handle);
 
 	/* drop reference from allocate - handle holds it now */
-	drm_gem_object_unreference_unlocked(obj);
+	drm_gem_object_put_unlocked(obj);
 
 	return ret;
 }
@@ -712,7 +713,7 @@ struct drm_gem_object *etnaviv_gem_new(struct drm_device *dev,
 
 	ret = etnaviv_gem_obj_add(dev, obj);
 	if (ret < 0) {
-		drm_gem_object_unreference_unlocked(obj);
+		drm_gem_object_put_unlocked(obj);
 		return ERR_PTR(ret);
 	}
 
@@ -800,7 +801,7 @@ static void __etnaviv_gem_userptr_get_pages(struct work_struct *_work)
 	}
 
 	mutex_unlock(&etnaviv_obj->lock);
-	drm_gem_object_unreference_unlocked(&etnaviv_obj->base);
+	drm_gem_object_put_unlocked(&etnaviv_obj->base);
 
 	mmput(work->mm);
 	put_task_struct(work->task);
@@ -858,7 +859,7 @@ static int etnaviv_gem_userptr_get_pages(struct etnaviv_gem_object *etnaviv_obj)
 	}
 
 	get_task_struct(current);
-	drm_gem_object_reference(&etnaviv_obj->base);
+	drm_gem_object_get(&etnaviv_obj->base);
 
 	work->mm = mm;
 	work->task = current;
@@ -924,6 +925,6 @@ int etnaviv_gem_new_userptr(struct drm_device *dev, struct drm_file *file,
 	ret = drm_gem_handle_create(file, &etnaviv_obj->base, handle);
 unreference:
 	/* drop reference from allocate - handle holds it now */
-	drm_gem_object_unreference_unlocked(&etnaviv_obj->base);
+	drm_gem_object_put_unlocked(&etnaviv_obj->base);
 	return ret;
 }
diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gem_prime.c b/drivers/gpu/drm/etnaviv/etnaviv_gem_prime.c
index e5da4f2300ba..ae884723e9b1 100644
--- a/drivers/gpu/drm/etnaviv/etnaviv_gem_prime.c
+++ b/drivers/gpu/drm/etnaviv/etnaviv_gem_prime.c
@@ -146,7 +146,7 @@ struct drm_gem_object *etnaviv_gem_prime_import_sg_table(struct drm_device *dev,
 	return &etnaviv_obj->base;
 
 fail:
-	drm_gem_object_unreference_unlocked(&etnaviv_obj->base);
+	drm_gem_object_put_unlocked(&etnaviv_obj->base);
 
 	return ERR_PTR(ret);
 }
diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gem_submit.c b/drivers/gpu/drm/etnaviv/etnaviv_gem_submit.c
index 5bd93169dac2..a7ff2e4c00d2 100644
--- a/drivers/gpu/drm/etnaviv/etnaviv_gem_submit.c
+++ b/drivers/gpu/drm/etnaviv/etnaviv_gem_submit.c
@@ -88,7 +88,7 @@ static int submit_lookup_objects(struct etnaviv_gem_submit *submit,
 		 * Take a refcount on the object. The file table lock
 		 * prevents the object_idr's refcount on this being dropped.
 		 */
-		drm_gem_object_reference(obj);
+		drm_gem_object_get(obj);
 
 		submit->bos[i].obj = to_etnaviv_bo(obj);
 	}
@@ -270,8 +270,8 @@ static int submit_reloc(struct etnaviv_gem_submit *submit, void *stream,
 		if (ret)
 			return ret;
 
-		if (r->reloc_offset >= bo->obj->base.size - sizeof(*ptr)) {
-			DRM_ERROR("relocation %u outside object", i);
+		if (r->reloc_offset > bo->obj->base.size - sizeof(*ptr)) {
+			DRM_ERROR("relocation %u outside object\n", i);
 			return -EINVAL;
 		}
 
@@ -291,7 +291,7 @@ static void submit_cleanup(struct etnaviv_gem_submit *submit)
 		struct etnaviv_gem_object *etnaviv_obj = submit->bos[i].obj;
 
 		submit_unlock_object(submit, i);
-		drm_gem_object_unreference_unlocked(&etnaviv_obj->base);
+		drm_gem_object_put_unlocked(&etnaviv_obj->base);
 	}
 
 	ww_acquire_fini(&submit->ticket);
diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gpu.c b/drivers/gpu/drm/etnaviv/etnaviv_gpu.c
index ada45fdd0eae..fc9a6a83dfc7 100644
--- a/drivers/gpu/drm/etnaviv/etnaviv_gpu.c
+++ b/drivers/gpu/drm/etnaviv/etnaviv_gpu.c
@@ -1622,10 +1622,12 @@ static int etnaviv_gpu_bind(struct device *dev, struct device *master,
 	struct etnaviv_gpu *gpu = dev_get_drvdata(dev);
 	int ret;
 
-	gpu->cooling = thermal_of_cooling_device_register(dev->of_node,
+	if (IS_ENABLED(CONFIG_THERMAL)) {
+		gpu->cooling = thermal_of_cooling_device_register(dev->of_node,
 				(char *)dev_name(dev), gpu, &cooling_ops);
-	if (IS_ERR(gpu->cooling))
-		return PTR_ERR(gpu->cooling);
+		if (IS_ERR(gpu->cooling))
+			return PTR_ERR(gpu->cooling);
+	}
 
 #ifdef CONFIG_PM
 	ret = pm_runtime_get_sync(gpu->dev);
diff --git a/drivers/gpu/drm/exynos/Kconfig b/drivers/gpu/drm/exynos/Kconfig
index 1d185347c64c..305dc3d4ff77 100644
--- a/drivers/gpu/drm/exynos/Kconfig
+++ b/drivers/gpu/drm/exynos/Kconfig
@@ -75,6 +75,7 @@ config DRM_EXYNOS_DP
 config DRM_EXYNOS_HDMI
 	bool "HDMI"
 	depends on DRM_EXYNOS_MIXER || DRM_EXYNOS5433_DECON
+	select CEC_CORE if CEC_NOTIFIER
 	help
 	  Choose this option if you want to use Exynos HDMI for DRM.
 
diff --git a/drivers/gpu/drm/exynos/exynos5433_drm_decon.c b/drivers/gpu/drm/exynos/exynos5433_drm_decon.c
index 5792ca88ab7a..730b8d9db187 100644
--- a/drivers/gpu/drm/exynos/exynos5433_drm_decon.c
+++ b/drivers/gpu/drm/exynos/exynos5433_drm_decon.c
@@ -13,6 +13,7 @@
 #include <linux/platform_device.h>
 #include <linux/clk.h>
 #include <linux/component.h>
+#include <linux/iopoll.h>
 #include <linux/mfd/syscon.h>
 #include <linux/of_device.h>
 #include <linux/of_gpio.h>
@@ -33,9 +34,8 @@
 #define WINDOWS_NR	3
 #define MIN_FB_WIDTH_FOR_16WORD_BURST	128
 
-#define IFTYPE_I80	(1 << 0)
-#define I80_HW_TRG	(1 << 1)
-#define IFTYPE_HDMI	(1 << 2)
+#define I80_HW_TRG	(1 << 0)
+#define IFTYPE_HDMI	(1 << 1)
 
 static const char * const decon_clks_name[] = {
 	"pclk",
@@ -57,6 +57,8 @@ struct decon_context {
 	struct regmap			*sysreg;
 	struct clk			*clks[ARRAY_SIZE(decon_clks_name)];
 	unsigned int			irq;
+	unsigned int			irq_vsync;
+	unsigned int			irq_lcd_sys;
 	unsigned int			te_irq;
 	unsigned long			out_type;
 	int				first_win;
@@ -90,7 +92,7 @@ static int decon_enable_vblank(struct exynos_drm_crtc *crtc)
 	u32 val;
 
 	val = VIDINTCON0_INTEN;
-	if (ctx->out_type & IFTYPE_I80)
+	if (crtc->i80_mode)
 		val |= VIDINTCON0_FRAMEDONE;
 	else
 		val |= VIDINTCON0_INTFRMEN | VIDINTCON0_FRAMESEL_FP;
@@ -139,7 +141,7 @@ static u32 decon_get_frame_count(struct decon_context *ctx, bool end)
 
 	switch (status & (VIDCON1_VSTATUS_MASK | VIDCON1_I80_ACTIVE)) {
 	case VIDCON1_VSTATUS_VS:
-		if (!(ctx->out_type & IFTYPE_I80))
+		if (!(ctx->crtc->i80_mode))
 			--frm;
 		break;
 	case VIDCON1_VSTATUS_BP:
@@ -166,7 +168,7 @@ static u32 decon_get_vblank_counter(struct exynos_drm_crtc *crtc)
 
 static void decon_setup_trigger(struct decon_context *ctx)
 {
-	if (!(ctx->out_type & (IFTYPE_I80 | I80_HW_TRG)))
+	if (!ctx->crtc->i80_mode && !(ctx->out_type & I80_HW_TRG))
 		return;
 
 	if (!(ctx->out_type & I80_HW_TRG)) {
@@ -206,7 +208,7 @@ static void decon_commit(struct exynos_drm_crtc *crtc)
 	val = VIDOUT_LCD_ON;
 	if (interlaced)
 		val |= VIDOUT_INTERLACE_EN_F;
-	if (ctx->out_type & IFTYPE_I80) {
+	if (crtc->i80_mode) {
 		val |= VIDOUT_COMMAND_IF;
 	} else {
 		val |= VIDOUT_RGB_IF;
@@ -222,7 +224,7 @@ static void decon_commit(struct exynos_drm_crtc *crtc)
 			VIDTCON2_HOZVAL(m->hdisplay - 1);
 	writel(val, ctx->addr + DECON_VIDTCON2);
 
-	if (!(ctx->out_type & IFTYPE_I80)) {
+	if (!crtc->i80_mode) {
 		int vbp = m->crtc_vtotal - m->crtc_vsync_end;
 		int vfp = m->crtc_vsync_start - m->crtc_vdisplay;
 
@@ -277,16 +279,14 @@ static void decon_win_set_pixfmt(struct decon_context *ctx, unsigned int win,
 		val |= WINCONx_BURSTLEN_16WORD;
 		break;
 	case DRM_FORMAT_ARGB8888:
+	default:
 		val |= WINCONx_BPPMODE_32BPP_A8888;
 		val |= WINCONx_WSWP_F | WINCONx_BLD_PIX_F | WINCONx_ALPHA_SEL_F;
 		val |= WINCONx_BURSTLEN_16WORD;
 		break;
-	default:
-		DRM_ERROR("Proper pixel format is not set\n");
-		return;
 	}
 
-	DRM_DEBUG_KMS("bpp = %u\n", fb->format->cpp[0] * 8);
+	DRM_DEBUG_KMS("cpp = %u\n", fb->format->cpp[0]);
 
 	/*
 	 * In case of exynos, setting dma-burst to 16Word causes permanent
@@ -329,7 +329,7 @@ static void decon_update_plane(struct exynos_drm_crtc *crtc,
 	struct decon_context *ctx = crtc->ctx;
 	struct drm_framebuffer *fb = state->base.fb;
 	unsigned int win = plane->index;
-	unsigned int bpp = fb->format->cpp[0];
+	unsigned int cpp = fb->format->cpp[0];
 	unsigned int pitch = fb->pitches[0];
 	dma_addr_t dma_addr = exynos_drm_fb_dma_addr(fb, 0);
 	u32 val;
@@ -365,11 +365,11 @@ static void decon_update_plane(struct exynos_drm_crtc *crtc,
 	writel(val, ctx->addr + DECON_VIDW0xADD1B0(win));
 
 	if (!(ctx->out_type & IFTYPE_HDMI))
-		val = BIT_VAL(pitch - state->crtc.w * bpp, 27, 14)
-			| BIT_VAL(state->crtc.w * bpp, 13, 0);
+		val = BIT_VAL(pitch - state->crtc.w * cpp, 27, 14)
+			| BIT_VAL(state->crtc.w * cpp, 13, 0);
 	else
-		val = BIT_VAL(pitch - state->crtc.w * bpp, 29, 15)
-			| BIT_VAL(state->crtc.w * bpp, 14, 0);
+		val = BIT_VAL(pitch - state->crtc.w * cpp, 29, 15)
+			| BIT_VAL(state->crtc.w * cpp, 14, 0);
 	writel(val, ctx->addr + DECON_VIDW0xADD2(win));
 
 	decon_win_set_pixfmt(ctx, win, fb);
@@ -407,24 +407,19 @@ static void decon_atomic_flush(struct exynos_drm_crtc *crtc)
 
 static void decon_swreset(struct decon_context *ctx)
 {
-	unsigned int tries;
 	unsigned long flags;
+	u32 val;
+	int ret;
 
 	writel(0, ctx->addr + DECON_VIDCON0);
-	for (tries = 2000; tries; --tries) {
-		if (~readl(ctx->addr + DECON_VIDCON0) & VIDCON0_STOP_STATUS)
-			break;
-		udelay(10);
-	}
+	readl_poll_timeout(ctx->addr + DECON_VIDCON0, val,
+			   ~val & VIDCON0_STOP_STATUS, 12, 20000);
 
 	writel(VIDCON0_SWRESET, ctx->addr + DECON_VIDCON0);
-	for (tries = 2000; tries; --tries) {
-		if (~readl(ctx->addr + DECON_VIDCON0) & VIDCON0_SWRESET)
-			break;
-		udelay(10);
-	}
+	ret = readl_poll_timeout(ctx->addr + DECON_VIDCON0, val,
+				 ~val & VIDCON0_SWRESET, 12, 20000);
 
-	WARN(tries == 0, "failed to software reset DECON\n");
+	WARN(ret < 0, "failed to software reset DECON\n");
 
 	spin_lock_irqsave(&ctx->vblank_lock, flags);
 	ctx->frame_id = 0;
@@ -515,6 +510,22 @@ err:
 		clk_disable_unprepare(ctx->clks[i]);
 }
 
+static enum drm_mode_status decon_mode_valid(struct exynos_drm_crtc *crtc,
+		const struct drm_display_mode *mode)
+{
+	struct decon_context *ctx = crtc->ctx;
+
+	ctx->irq = crtc->i80_mode ? ctx->irq_lcd_sys : ctx->irq_vsync;
+
+	if (ctx->irq)
+		return MODE_OK;
+
+	dev_info(ctx->dev, "Sink requires %s mode, but appropriate interrupt is not provided.\n",
+			crtc->i80_mode ? "command" : "video");
+
+	return MODE_BAD;
+}
+
 static const struct exynos_drm_crtc_ops decon_crtc_ops = {
 	.enable			= decon_enable,
 	.disable		= decon_disable,
@@ -524,6 +535,7 @@ static const struct exynos_drm_crtc_ops decon_crtc_ops = {
 	.atomic_begin		= decon_atomic_begin,
 	.update_plane		= decon_update_plane,
 	.disable_plane		= decon_disable_plane,
+	.mode_valid		= decon_mode_valid,
 	.atomic_flush		= decon_atomic_flush,
 };
 
@@ -674,19 +686,22 @@ static const struct of_device_id exynos5433_decon_driver_dt_match[] = {
 MODULE_DEVICE_TABLE(of, exynos5433_decon_driver_dt_match);
 
 static int decon_conf_irq(struct decon_context *ctx, const char *name,
-		irq_handler_t handler, unsigned long int flags, bool required)
+		irq_handler_t handler, unsigned long int flags)
 {
 	struct platform_device *pdev = to_platform_device(ctx->dev);
 	int ret, irq = platform_get_irq_byname(pdev, name);
 
 	if (irq < 0) {
-		if (irq == -EPROBE_DEFER)
+		switch (irq) {
+		case -EPROBE_DEFER:
 			return irq;
-		if (required)
-			dev_err(ctx->dev, "cannot get %s IRQ\n", name);
-		else
-			irq = 0;
-		return irq;
+		case -ENODATA:
+		case -ENXIO:
+			return 0;
+		default:
+			dev_err(ctx->dev, "IRQ %s get failed, %d\n", name, irq);
+			return irq;
+		}
 	}
 	irq_set_status_flags(irq, IRQ_NOAUTOEN);
 	ret = devm_request_irq(ctx->dev, irq, handler, flags, "drm_decon", ctx);
@@ -714,11 +729,8 @@ static int exynos5433_decon_probe(struct platform_device *pdev)
 	ctx->out_type = (unsigned long)of_device_get_match_data(dev);
 	spin_lock_init(&ctx->vblank_lock);
 
-	if (ctx->out_type & IFTYPE_HDMI) {
+	if (ctx->out_type & IFTYPE_HDMI)
 		ctx->first_win = 1;
-	} else if (of_get_child_by_name(dev->of_node, "i80-if-timings")) {
-		ctx->out_type |= IFTYPE_I80;
-	}
 
 	for (i = 0; i < ARRAY_SIZE(decon_clks_name); i++) {
 		struct clk *clk;
@@ -742,25 +754,23 @@ static int exynos5433_decon_probe(struct platform_device *pdev)
 		return PTR_ERR(ctx->addr);
 	}
 
-	if (ctx->out_type & IFTYPE_I80) {
-		ret = decon_conf_irq(ctx, "lcd_sys", decon_irq_handler, 0, true);
-		if (ret < 0)
-			return ret;
-		ctx->irq = ret;
+	ret = decon_conf_irq(ctx, "vsync", decon_irq_handler, 0);
+	if (ret < 0)
+		return ret;
+	ctx->irq_vsync = ret;
 
-		ret = decon_conf_irq(ctx, "te", decon_te_irq_handler,
-				     IRQF_TRIGGER_RISING, false);
-		if (ret < 0)
-			return ret;
-		if (ret) {
-			ctx->te_irq = ret;
-			ctx->out_type &= ~I80_HW_TRG;
-		}
-	} else {
-		ret = decon_conf_irq(ctx, "vsync", decon_irq_handler, 0, true);
-		if (ret < 0)
+	ret = decon_conf_irq(ctx, "lcd_sys", decon_irq_handler, 0);
+	if (ret < 0)
+		return ret;
+	ctx->irq_lcd_sys = ret;
+
+	ret = decon_conf_irq(ctx, "te", decon_te_irq_handler,
+			IRQF_TRIGGER_RISING);
+	if (ret < 0)
 			return ret;
-		ctx->irq = ret;
+	if (ret) {
+		ctx->te_irq = ret;
+		ctx->out_type &= ~I80_HW_TRG;
 	}
 
 	if (ctx->out_type & I80_HW_TRG) {
diff --git a/drivers/gpu/drm/exynos/exynos7_drm_decon.c b/drivers/gpu/drm/exynos/exynos7_drm_decon.c
index 3e88269fdc2e..615efcf7782a 100644
--- a/drivers/gpu/drm/exynos/exynos7_drm_decon.c
+++ b/drivers/gpu/drm/exynos/exynos7_drm_decon.c
@@ -309,19 +309,14 @@ static void decon_win_set_pixfmt(struct decon_context *ctx, unsigned int win,
 		val |= WINCONx_BURSTLEN_16WORD;
 		break;
 	case DRM_FORMAT_BGRA8888:
+	default:
 		val |= WINCONx_BPPMODE_32BPP_BGRA | WINCONx_BLD_PIX |
 			WINCONx_ALPHA_SEL;
 		val |= WINCONx_BURSTLEN_16WORD;
 		break;
-	default:
-		DRM_DEBUG_KMS("invalid pixel size so using unpacked 24bpp.\n");
-
-		val |= WINCONx_BPPMODE_24BPP_xRGB;
-		val |= WINCONx_BURSTLEN_16WORD;
-		break;
 	}
 
-	DRM_DEBUG_KMS("bpp = %d\n", fb->format->cpp[0] * 8);
+	DRM_DEBUG_KMS("cpp = %d\n", fb->format->cpp[0]);
 
 	/*
 	 * In case of exynos, setting dma-burst to 16Word causes permanent
@@ -398,7 +393,7 @@ static void decon_update_plane(struct exynos_drm_crtc *crtc,
 	unsigned int last_x;
 	unsigned int last_y;
 	unsigned int win = plane->index;
-	unsigned int bpp = fb->format->cpp[0];
+	unsigned int cpp = fb->format->cpp[0];
 	unsigned int pitch = fb->pitches[0];
 
 	if (ctx->suspended)
@@ -418,7 +413,7 @@ static void decon_update_plane(struct exynos_drm_crtc *crtc,
 	val = (unsigned long)exynos_drm_fb_dma_addr(fb, 0);
 	writel(val, ctx->regs + VIDW_BUF_START(win));
 
-	padding = (pitch / bpp) - fb->width;
+	padding = (pitch / cpp) - fb->width;
 
 	/* buffer size */
 	writel(fb->width + padding, ctx->regs + VIDW_WHOLE_X(win));
diff --git a/drivers/gpu/drm/exynos/exynos_dp.c b/drivers/gpu/drm/exynos/exynos_dp.c
index 385537b726a6..39629e7a80b9 100644
--- a/drivers/gpu/drm/exynos/exynos_dp.c
+++ b/drivers/gpu/drm/exynos/exynos_dp.c
@@ -155,7 +155,7 @@ static int exynos_dp_bind(struct device *dev, struct device *master, void *data)
 	struct exynos_dp_device *dp = dev_get_drvdata(dev);
 	struct drm_encoder *encoder = &dp->encoder;
 	struct drm_device *drm_dev = data;
-	int pipe, ret;
+	int ret;
 
 	/*
 	 * Just like the probe function said, we don't need the
@@ -179,20 +179,15 @@ static int exynos_dp_bind(struct device *dev, struct device *master, void *data)
 			return ret;
 	}
 
-	pipe = exynos_drm_crtc_get_pipe_from_type(drm_dev,
-						  EXYNOS_DISPLAY_TYPE_LCD);
-	if (pipe < 0)
-		return pipe;
-
-	encoder->possible_crtcs = 1 << pipe;
-
-	DRM_DEBUG_KMS("possible_crtcs = 0x%x\n", encoder->possible_crtcs);
-
 	drm_encoder_init(drm_dev, encoder, &exynos_dp_encoder_funcs,
 			 DRM_MODE_ENCODER_TMDS, NULL);
 
 	drm_encoder_helper_add(encoder, &exynos_dp_encoder_helper_funcs);
 
+	ret = exynos_drm_set_possible_crtcs(encoder, EXYNOS_DISPLAY_TYPE_LCD);
+	if (ret < 0)
+		return ret;
+
 	dp->plat_data.encoder = encoder;
 
 	return analogix_dp_bind(dev, dp->drm_dev, &dp->plat_data);
diff --git a/drivers/gpu/drm/exynos/exynos_drm_core.c b/drivers/gpu/drm/exynos/exynos_drm_core.c
index edbd98ff293e..b0c0621fcdf7 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_core.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_core.c
@@ -13,6 +13,7 @@
  */
 
 #include <drm/drmP.h>
+
 #include "exynos_drm_drv.h"
 #include "exynos_drm_crtc.h"
 
diff --git a/drivers/gpu/drm/exynos/exynos_drm_crtc.c b/drivers/gpu/drm/exynos/exynos_drm_crtc.c
index c37078fbe0ea..6ce0821590df 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_crtc.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_crtc.c
@@ -16,6 +16,7 @@
 #include <drm/drm_crtc_helper.h>
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
+#include <drm/drm_encoder.h>
 
 #include "exynos_drm_crtc.h"
 #include "exynos_drm_drv.h"
@@ -83,7 +84,19 @@ static void exynos_crtc_atomic_flush(struct drm_crtc *crtc,
 		exynos_crtc->ops->atomic_flush(exynos_crtc);
 }
 
+static enum drm_mode_status exynos_crtc_mode_valid(struct drm_crtc *crtc,
+	const struct drm_display_mode *mode)
+{
+	struct exynos_drm_crtc *exynos_crtc = to_exynos_crtc(crtc);
+
+	if (exynos_crtc->ops->mode_valid)
+		return exynos_crtc->ops->mode_valid(exynos_crtc, mode);
+
+	return MODE_OK;
+}
+
 static const struct drm_crtc_helper_funcs exynos_crtc_helper_funcs = {
+	.mode_valid	= exynos_crtc_mode_valid,
 	.atomic_check	= exynos_crtc_atomic_check,
 	.atomic_begin	= exynos_crtc_atomic_begin,
 	.atomic_flush	= exynos_crtc_atomic_flush,
@@ -191,16 +204,30 @@ err_crtc:
 	return ERR_PTR(ret);
 }
 
-int exynos_drm_crtc_get_pipe_from_type(struct drm_device *drm_dev,
+struct exynos_drm_crtc *exynos_drm_crtc_get_by_type(struct drm_device *drm_dev,
 				       enum exynos_drm_output_type out_type)
 {
 	struct drm_crtc *crtc;
 
 	drm_for_each_crtc(crtc, drm_dev)
 		if (to_exynos_crtc(crtc)->type == out_type)
-			return drm_crtc_index(crtc);
+			return to_exynos_crtc(crtc);
 
-	return -EPERM;
+	return ERR_PTR(-EPERM);
+}
+
+int exynos_drm_set_possible_crtcs(struct drm_encoder *encoder,
+		enum exynos_drm_output_type out_type)
+{
+	struct exynos_drm_crtc *crtc = exynos_drm_crtc_get_by_type(encoder->dev,
+						out_type);
+
+	if (IS_ERR(crtc))
+		return PTR_ERR(crtc);
+
+	encoder->possible_crtcs = drm_crtc_mask(&crtc->base);
+
+	return 0;
 }
 
 void exynos_drm_crtc_te_handler(struct drm_crtc *crtc)
diff --git a/drivers/gpu/drm/exynos/exynos_drm_crtc.h b/drivers/gpu/drm/exynos/exynos_drm_crtc.h
index ef58b64e3d2d..dec446109e6c 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_crtc.h
+++ b/drivers/gpu/drm/exynos/exynos_drm_crtc.h
@@ -15,21 +15,25 @@
 #ifndef _EXYNOS_DRM_CRTC_H_
 #define _EXYNOS_DRM_CRTC_H_
 
+
 #include "exynos_drm_drv.h"
 
 struct exynos_drm_crtc *exynos_drm_crtc_create(struct drm_device *drm_dev,
 					struct drm_plane *plane,
-					enum exynos_drm_output_type type,
+					enum exynos_drm_output_type out_type,
 					const struct exynos_drm_crtc_ops *ops,
 					void *context);
 void exynos_drm_crtc_wait_pending_update(struct exynos_drm_crtc *exynos_crtc);
 void exynos_drm_crtc_finish_update(struct exynos_drm_crtc *exynos_crtc,
 				   struct exynos_drm_plane *exynos_plane);
 
-/* This function gets pipe value to crtc device matched with out_type. */
-int exynos_drm_crtc_get_pipe_from_type(struct drm_device *drm_dev,
+/* This function gets crtc device matched with out_type. */
+struct exynos_drm_crtc *exynos_drm_crtc_get_by_type(struct drm_device *drm_dev,
 				       enum exynos_drm_output_type out_type);
 
+int exynos_drm_set_possible_crtcs(struct drm_encoder *encoder,
+		enum exynos_drm_output_type out_type);
+
 /*
  * This function calls the crtc device(manager)'s te_handler() callback
  * to trigger to transfer video image at the tearing effect synchronization
diff --git a/drivers/gpu/drm/exynos/exynos_drm_dpi.c b/drivers/gpu/drm/exynos/exynos_drm_dpi.c
index 76d80e5de521..66945e0dc57f 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_dpi.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_dpi.c
@@ -202,19 +202,15 @@ int exynos_dpi_bind(struct drm_device *dev, struct drm_encoder *encoder)
 {
 	int ret;
 
-	ret = exynos_drm_crtc_get_pipe_from_type(dev, EXYNOS_DISPLAY_TYPE_LCD);
-	if (ret < 0)
-		return ret;
-
-	encoder->possible_crtcs = 1 << ret;
-
-	DRM_DEBUG_KMS("possible_crtcs = 0x%x\n", encoder->possible_crtcs);
-
 	drm_encoder_init(dev, encoder, &exynos_dpi_encoder_funcs,
 			 DRM_MODE_ENCODER_TMDS, NULL);
 
 	drm_encoder_helper_add(encoder, &exynos_dpi_encoder_helper_funcs);
 
+	ret = exynos_drm_set_possible_crtcs(encoder, EXYNOS_DISPLAY_TYPE_LCD);
+	if (ret < 0)
+		return ret;
+
 	ret = exynos_dpi_create_connector(encoder);
 	if (ret) {
 		DRM_ERROR("failed to create connector ret = %d\n", ret);
diff --git a/drivers/gpu/drm/exynos/exynos_drm_drv.c b/drivers/gpu/drm/exynos/exynos_drm_drv.c
index cab9e12d7846..b1f7299600f0 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_drv.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_drv.c
@@ -145,8 +145,6 @@ static struct drm_driver exynos_drm_driver = {
 	.gem_free_object_unlocked = exynos_drm_gem_free_object,
 	.gem_vm_ops		= &exynos_drm_gem_vm_ops,
 	.dumb_create		= exynos_drm_gem_dumb_create,
-	.dumb_map_offset	= exynos_drm_gem_dumb_map_offset,
-	.dumb_destroy		= drm_gem_dumb_destroy,
 	.prime_handle_to_fd	= drm_gem_prime_handle_to_fd,
 	.prime_fd_to_handle	= drm_gem_prime_fd_to_handle,
 	.gem_prime_export	= drm_gem_prime_export,
@@ -455,7 +453,6 @@ static int exynos_drm_platform_probe(struct platform_device *pdev)
 	struct component_match *match;
 
 	pdev->dev.coherent_dma_mask = DMA_BIT_MASK(32);
-	exynos_drm_driver.num_ioctls = ARRAY_SIZE(exynos_ioctls);
 
 	match = exynos_drm_match_add(&pdev->dev);
 	if (IS_ERR(match))
diff --git a/drivers/gpu/drm/exynos/exynos_drm_drv.h b/drivers/gpu/drm/exynos/exynos_drm_drv.h
index a93de321706b..cf131c2aa23e 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_drv.h
+++ b/drivers/gpu/drm/exynos/exynos_drm_drv.h
@@ -91,6 +91,7 @@ struct exynos_drm_plane {
 #define EXYNOS_DRM_PLANE_CAP_DOUBLE	(1 << 0)
 #define EXYNOS_DRM_PLANE_CAP_SCALE	(1 << 1)
 #define EXYNOS_DRM_PLANE_CAP_ZPOS	(1 << 2)
+#define EXYNOS_DRM_PLANE_CAP_TILE	(1 << 3)
 
 /*
  * Exynos DRM plane configuration structure.
@@ -117,6 +118,7 @@ struct exynos_drm_plane_config {
  * @disable: disable the device
  * @enable_vblank: specific driver callback for enabling vblank interrupt.
  * @disable_vblank: specific driver callback for disabling vblank interrupt.
+ * @mode_valid: specific driver callback for mode validation
  * @atomic_check: validate state
  * @atomic_begin: prepare device to receive an update
  * @atomic_flush: mark the end of device update
@@ -132,6 +134,8 @@ struct exynos_drm_crtc_ops {
 	int (*enable_vblank)(struct exynos_drm_crtc *crtc);
 	void (*disable_vblank)(struct exynos_drm_crtc *crtc);
 	u32 (*get_vblank_counter)(struct exynos_drm_crtc *crtc);
+	enum drm_mode_status (*mode_valid)(struct exynos_drm_crtc *crtc,
+		const struct drm_display_mode *mode);
 	int (*atomic_check)(struct exynos_drm_crtc *crtc,
 			    struct drm_crtc_state *state);
 	void (*atomic_begin)(struct exynos_drm_crtc *crtc);
@@ -162,6 +166,7 @@ struct exynos_drm_crtc {
 	const struct exynos_drm_crtc_ops	*ops;
 	void				*ctx;
 	struct exynos_drm_clk		*pipe_clk;
+	bool				i80_mode : 1;
 };
 
 static inline void exynos_drm_pipe_clk_enable(struct exynos_drm_crtc *crtc,
diff --git a/drivers/gpu/drm/exynos/exynos_drm_dsi.c b/drivers/gpu/drm/exynos/exynos_drm_dsi.c
index 4ea7cc7cb3de..7904ffa9abfb 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_dsi.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_dsi.c
@@ -254,7 +254,6 @@ struct exynos_dsi {
 	struct drm_encoder encoder;
 	struct mipi_dsi_host dsi_host;
 	struct drm_connector connector;
-	struct device_node *panel_node;
 	struct drm_panel *panel;
 	struct device *dev;
 
@@ -1329,12 +1328,13 @@ static int exynos_dsi_init(struct exynos_dsi *dsi)
 	return 0;
 }
 
-static int exynos_dsi_register_te_irq(struct exynos_dsi *dsi)
+static int exynos_dsi_register_te_irq(struct exynos_dsi *dsi,
+				      struct device *panel)
 {
 	int ret;
 	int te_gpio_irq;
 
-	dsi->te_gpio = of_get_named_gpio(dsi->panel_node, "te-gpios", 0);
+	dsi->te_gpio = of_get_named_gpio(panel->of_node, "te-gpios", 0);
 	if (dsi->te_gpio == -ENOENT)
 		return 0;
 
@@ -1374,85 +1374,6 @@ static void exynos_dsi_unregister_te_irq(struct exynos_dsi *dsi)
 	}
 }
 
-static int exynos_dsi_host_attach(struct mipi_dsi_host *host,
-				  struct mipi_dsi_device *device)
-{
-	struct exynos_dsi *dsi = host_to_dsi(host);
-
-	dsi->lanes = device->lanes;
-	dsi->format = device->format;
-	dsi->mode_flags = device->mode_flags;
-	dsi->panel_node = device->dev.of_node;
-
-	/*
-	 * This is a temporary solution and should be made by more generic way.
-	 *
-	 * If attached panel device is for command mode one, dsi should register
-	 * TE interrupt handler.
-	 */
-	if (!(dsi->mode_flags & MIPI_DSI_MODE_VIDEO)) {
-		int ret = exynos_dsi_register_te_irq(dsi);
-
-		if (ret)
-			return ret;
-	}
-
-	if (dsi->connector.dev)
-		drm_helper_hpd_irq_event(dsi->connector.dev);
-
-	return 0;
-}
-
-static int exynos_dsi_host_detach(struct mipi_dsi_host *host,
-				  struct mipi_dsi_device *device)
-{
-	struct exynos_dsi *dsi = host_to_dsi(host);
-
-	exynos_dsi_unregister_te_irq(dsi);
-
-	dsi->panel_node = NULL;
-
-	if (dsi->connector.dev)
-		drm_helper_hpd_irq_event(dsi->connector.dev);
-
-	return 0;
-}
-
-static ssize_t exynos_dsi_host_transfer(struct mipi_dsi_host *host,
-				        const struct mipi_dsi_msg *msg)
-{
-	struct exynos_dsi *dsi = host_to_dsi(host);
-	struct exynos_dsi_transfer xfer;
-	int ret;
-
-	if (!(dsi->state & DSIM_STATE_ENABLED))
-		return -EINVAL;
-
-	if (!(dsi->state & DSIM_STATE_INITIALIZED)) {
-		ret = exynos_dsi_init(dsi);
-		if (ret)
-			return ret;
-		dsi->state |= DSIM_STATE_INITIALIZED;
-	}
-
-	ret = mipi_dsi_create_packet(&xfer.packet, msg);
-	if (ret < 0)
-		return ret;
-
-	xfer.rx_len = msg->rx_len;
-	xfer.rx_payload = msg->rx_buf;
-	xfer.flags = msg->flags;
-
-	ret = exynos_dsi_transfer(dsi, &xfer);
-	return (ret < 0) ? ret : xfer.rx_done;
-}
-
-static const struct mipi_dsi_host_ops exynos_dsi_ops = {
-	.attach = exynos_dsi_host_attach,
-	.detach = exynos_dsi_host_detach,
-	.transfer = exynos_dsi_host_transfer,
-};
-
 static void exynos_dsi_enable(struct drm_encoder *encoder)
 {
 	struct exynos_dsi *dsi = encoder_to_dsi(encoder);
@@ -1508,25 +1429,7 @@ static void exynos_dsi_disable(struct drm_encoder *encoder)
 static enum drm_connector_status
 exynos_dsi_detect(struct drm_connector *connector, bool force)
 {
-	struct exynos_dsi *dsi = connector_to_dsi(connector);
-
-	if (!dsi->panel) {
-		dsi->panel = of_drm_find_panel(dsi->panel_node);
-		if (dsi->panel)
-			drm_panel_attach(dsi->panel, &dsi->connector);
-	} else if (!dsi->panel_node) {
-		struct drm_encoder *encoder;
-
-		encoder = platform_get_drvdata(to_platform_device(dsi->dev));
-		exynos_dsi_disable(encoder);
-		drm_panel_detach(dsi->panel);
-		dsi->panel = NULL;
-	}
-
-	if (dsi->panel)
-		return connector_status_connected;
-
-	return connector_status_disconnected;
+	return connector->status;
 }
 
 static void exynos_dsi_connector_destroy(struct drm_connector *connector)
@@ -1575,6 +1478,7 @@ static int exynos_dsi_create_connector(struct drm_encoder *encoder)
 		return ret;
 	}
 
+	connector->status = connector_status_disconnected;
 	drm_connector_helper_add(connector, &exynos_dsi_connector_helper_funcs);
 	drm_mode_connector_attach_encoder(connector, encoder);
 
@@ -1611,6 +1515,105 @@ static const struct drm_encoder_funcs exynos_dsi_encoder_funcs = {
 
 MODULE_DEVICE_TABLE(of, exynos_dsi_of_match);
 
+static int exynos_dsi_host_attach(struct mipi_dsi_host *host,
+				  struct mipi_dsi_device *device)
+{
+	struct exynos_dsi *dsi = host_to_dsi(host);
+	struct drm_device *drm = dsi->connector.dev;
+
+	/*
+	 * This is a temporary solution and should be made by more generic way.
+	 *
+	 * If attached panel device is for command mode one, dsi should register
+	 * TE interrupt handler.
+	 */
+	if (!(device->mode_flags & MIPI_DSI_MODE_VIDEO)) {
+		int ret = exynos_dsi_register_te_irq(dsi, &device->dev);
+		if (ret)
+			return ret;
+	}
+
+	mutex_lock(&drm->mode_config.mutex);
+
+	dsi->lanes = device->lanes;
+	dsi->format = device->format;
+	dsi->mode_flags = device->mode_flags;
+	dsi->panel = of_drm_find_panel(device->dev.of_node);
+	if (dsi->panel) {
+		drm_panel_attach(dsi->panel, &dsi->connector);
+		dsi->connector.status = connector_status_connected;
+	}
+	exynos_drm_crtc_get_by_type(drm, EXYNOS_DISPLAY_TYPE_LCD)->i80_mode =
+			!(dsi->mode_flags & MIPI_DSI_MODE_VIDEO);
+
+	mutex_unlock(&drm->mode_config.mutex);
+
+	if (drm->mode_config.poll_enabled)
+		drm_kms_helper_hotplug_event(drm);
+
+	return 0;
+}
+
+static int exynos_dsi_host_detach(struct mipi_dsi_host *host,
+				  struct mipi_dsi_device *device)
+{
+	struct exynos_dsi *dsi = host_to_dsi(host);
+	struct drm_device *drm = dsi->connector.dev;
+
+	mutex_lock(&drm->mode_config.mutex);
+
+	if (dsi->panel) {
+		exynos_dsi_disable(&dsi->encoder);
+		drm_panel_detach(dsi->panel);
+		dsi->panel = NULL;
+		dsi->connector.status = connector_status_disconnected;
+	}
+
+	mutex_unlock(&drm->mode_config.mutex);
+
+	if (drm->mode_config.poll_enabled)
+		drm_kms_helper_hotplug_event(drm);
+
+	exynos_dsi_unregister_te_irq(dsi);
+
+	return 0;
+}
+
+static ssize_t exynos_dsi_host_transfer(struct mipi_dsi_host *host,
+					 const struct mipi_dsi_msg *msg)
+{
+	struct exynos_dsi *dsi = host_to_dsi(host);
+	struct exynos_dsi_transfer xfer;
+	int ret;
+
+	if (!(dsi->state & DSIM_STATE_ENABLED))
+		return -EINVAL;
+
+	if (!(dsi->state & DSIM_STATE_INITIALIZED)) {
+		ret = exynos_dsi_init(dsi);
+		if (ret)
+			return ret;
+		dsi->state |= DSIM_STATE_INITIALIZED;
+	}
+
+	ret = mipi_dsi_create_packet(&xfer.packet, msg);
+	if (ret < 0)
+		return ret;
+
+	xfer.rx_len = msg->rx_len;
+	xfer.rx_payload = msg->rx_buf;
+	xfer.flags = msg->flags;
+
+	ret = exynos_dsi_transfer(dsi, &xfer);
+	return (ret < 0) ? ret : xfer.rx_done;
+}
+
+static const struct mipi_dsi_host_ops exynos_dsi_ops = {
+	.attach = exynos_dsi_host_attach,
+	.detach = exynos_dsi_host_detach,
+	.transfer = exynos_dsi_host_transfer,
+};
+
 static int exynos_dsi_of_read_u32(const struct device_node *np,
 				  const char *propname, u32 *out_value)
 {
@@ -1649,8 +1652,6 @@ static int exynos_dsi_parse_dt(struct exynos_dsi *dsi)
 		return ret;
 
 	dsi->bridge_node = of_graph_get_remote_node(node, DSI_PORT_IN, 0);
-	if (!dsi->bridge_node)
-		return -EINVAL;
 
 	return 0;
 }
@@ -1664,20 +1665,15 @@ static int exynos_dsi_bind(struct device *dev, struct device *master,
 	struct drm_bridge *bridge;
 	int ret;
 
-	ret = exynos_drm_crtc_get_pipe_from_type(drm_dev,
-						  EXYNOS_DISPLAY_TYPE_LCD);
-	if (ret < 0)
-		return ret;
-
-	encoder->possible_crtcs = 1 << ret;
-
-	DRM_DEBUG_KMS("possible_crtcs = 0x%x\n", encoder->possible_crtcs);
-
 	drm_encoder_init(drm_dev, encoder, &exynos_dsi_encoder_funcs,
 			 DRM_MODE_ENCODER_TMDS, NULL);
 
 	drm_encoder_helper_add(encoder, &exynos_dsi_encoder_helper_funcs);
 
+	ret = exynos_drm_set_possible_crtcs(encoder, EXYNOS_DISPLAY_TYPE_LCD);
+	if (ret < 0)
+		return ret;
+
 	ret = exynos_dsi_create_connector(encoder);
 	if (ret) {
 		DRM_ERROR("failed to create connector ret = %d\n", ret);
@@ -1685,9 +1681,11 @@ static int exynos_dsi_bind(struct device *dev, struct device *master,
 		return ret;
 	}
 
-	bridge = of_drm_find_bridge(dsi->bridge_node);
-	if (bridge)
-		drm_bridge_attach(encoder, bridge, NULL);
+	if (dsi->bridge_node) {
+		bridge = of_drm_find_bridge(dsi->bridge_node);
+		if (bridge)
+			drm_bridge_attach(encoder, bridge, NULL);
+	}
 
 	return mipi_dsi_host_register(&dsi->dsi_host);
 }
diff --git a/drivers/gpu/drm/exynos/exynos_drm_fb.c b/drivers/gpu/drm/exynos/exynos_drm_fb.c
index ed1a648d518c..8208df56a88f 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_fb.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_fb.c
@@ -145,13 +145,19 @@ static struct drm_framebuffer *
 exynos_user_fb_create(struct drm_device *dev, struct drm_file *file_priv,
 		      const struct drm_mode_fb_cmd2 *mode_cmd)
 {
+	const struct drm_format_info *info = drm_get_format_info(dev, mode_cmd);
 	struct exynos_drm_gem *exynos_gem[MAX_FB_BUFFER];
 	struct drm_gem_object *obj;
 	struct drm_framebuffer *fb;
 	int i;
 	int ret;
 
-	for (i = 0; i < drm_format_num_planes(mode_cmd->pixel_format); i++) {
+	for (i = 0; i < info->num_planes; i++) {
+		unsigned int height = (i == 0) ? mode_cmd->height :
+				     DIV_ROUND_UP(mode_cmd->height, info->vsub);
+		unsigned long size = height * mode_cmd->pitches[i] +
+				     mode_cmd->offsets[i];
+
 		obj = drm_gem_object_lookup(file_priv, mode_cmd->handles[i]);
 		if (!obj) {
 			DRM_ERROR("failed to lookup gem object\n");
@@ -160,6 +166,12 @@ exynos_user_fb_create(struct drm_device *dev, struct drm_file *file_priv,
 		}
 
 		exynos_gem[i] = to_exynos_gem(obj);
+
+		if (size > exynos_gem[i]->size) {
+			i++;
+			ret = -EINVAL;
+			goto err;
+		}
 	}
 
 	fb = exynos_drm_framebuffer_init(dev, mode_cmd, exynos_gem, i);
@@ -213,4 +225,6 @@ void exynos_drm_mode_config_init(struct drm_device *dev)
 
 	dev->mode_config.funcs = &exynos_drm_mode_config_funcs;
 	dev->mode_config.helper_private = &exynos_drm_mode_config_helpers;
+
+	dev->mode_config.allow_fb_modifiers = true;
 }
diff --git a/drivers/gpu/drm/exynos/exynos_drm_fimd.c b/drivers/gpu/drm/exynos/exynos_drm_fimd.c
index 60f93cad6643..d42ae2bc3e56 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_fimd.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_fimd.c
@@ -583,18 +583,12 @@ static void fimd_win_set_pixfmt(struct fimd_context *ctx, unsigned int win,
 		val |= WINCONx_BURSTLEN_16WORD;
 		break;
 	case DRM_FORMAT_ARGB8888:
+	default:
 		val |= WINCON1_BPPMODE_25BPP_A1888
 			| WINCON1_BLD_PIX | WINCON1_ALPHA_SEL;
 		val |= WINCONx_WSWP;
 		val |= WINCONx_BURSTLEN_16WORD;
 		break;
-	default:
-		DRM_DEBUG_KMS("invalid pixel size so using unpacked 24bpp.\n");
-
-		val |= WINCON0_BPPMODE_24BPP_888;
-		val |= WINCONx_WSWP;
-		val |= WINCONx_BURSTLEN_16WORD;
-		break;
 	}
 
 	/*
@@ -718,13 +712,13 @@ static void fimd_update_plane(struct exynos_drm_crtc *crtc,
 	unsigned long val, size, offset;
 	unsigned int last_x, last_y, buf_offsize, line_size;
 	unsigned int win = plane->index;
-	unsigned int bpp = fb->format->cpp[0];
+	unsigned int cpp = fb->format->cpp[0];
 	unsigned int pitch = fb->pitches[0];
 
 	if (ctx->suspended)
 		return;
 
-	offset = state->src.x * bpp;
+	offset = state->src.x * cpp;
 	offset += state->src.y * pitch;
 
 	/* buffer start address */
@@ -743,8 +737,8 @@ static void fimd_update_plane(struct exynos_drm_crtc *crtc,
 			state->crtc.w, state->crtc.h);
 
 	/* buffer size */
-	buf_offsize = pitch - (state->crtc.w * bpp);
-	line_size = state->crtc.w * bpp;
+	buf_offsize = pitch - (state->crtc.w * cpp);
+	line_size = state->crtc.w * cpp;
 	val = VIDW_BUF_SIZE_OFFSET(buf_offsize) |
 		VIDW_BUF_SIZE_PAGEWIDTH(line_size) |
 		VIDW_BUF_SIZE_OFFSET_E(buf_offsize) |
diff --git a/drivers/gpu/drm/exynos/exynos_drm_gem.c b/drivers/gpu/drm/exynos/exynos_drm_gem.c
index c23479be4850..077de014d610 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_gem.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_gem.c
@@ -286,8 +286,8 @@ int exynos_drm_gem_map_ioctl(struct drm_device *dev, void *data,
 {
 	struct drm_exynos_gem_map *args = data;
 
-	return exynos_drm_gem_dumb_map_offset(file_priv, dev, args->handle,
-					      &args->offset);
+	return drm_gem_dumb_map_offset(file_priv, dev, args->handle,
+				       &args->offset);
 }
 
 dma_addr_t *exynos_drm_gem_get_dma_addr(struct drm_device *dev,
@@ -422,32 +422,6 @@ int exynos_drm_gem_dumb_create(struct drm_file *file_priv,
 	return 0;
 }
 
-int exynos_drm_gem_dumb_map_offset(struct drm_file *file_priv,
-				   struct drm_device *dev, uint32_t handle,
-				   uint64_t *offset)
-{
-	struct drm_gem_object *obj;
-	int ret = 0;
-
-	/*
-	 * get offset of memory allocated for drm framebuffer.
-	 * - this callback would be called by user application
-	 *	with DRM_IOCTL_MODE_MAP_DUMB command.
-	 */
-
-	obj = drm_gem_object_lookup(file_priv, handle);
-	if (!obj) {
-		DRM_ERROR("failed to lookup gem object.\n");
-		return -EINVAL;
-	}
-
-	*offset = drm_vma_node_offset_addr(&obj->vma_node);
-	DRM_DEBUG_KMS("offset = 0x%lx\n", (unsigned long)*offset);
-
-	drm_gem_object_unreference_unlocked(obj);
-	return ret;
-}
-
 int exynos_drm_gem_fault(struct vm_fault *vmf)
 {
 	struct vm_area_struct *vma = vmf->vma;
diff --git a/drivers/gpu/drm/exynos/exynos_drm_gem.h b/drivers/gpu/drm/exynos/exynos_drm_gem.h
index 85457255fcd1..e86d1a9518c3 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_gem.h
+++ b/drivers/gpu/drm/exynos/exynos_drm_gem.h
@@ -110,11 +110,6 @@ int exynos_drm_gem_dumb_create(struct drm_file *file_priv,
 			       struct drm_device *dev,
 			       struct drm_mode_create_dumb *args);
 
-/* map memory region for drm framebuffer to user space. */
-int exynos_drm_gem_dumb_map_offset(struct drm_file *file_priv,
-				   struct drm_device *dev, uint32_t handle,
-				   uint64_t *offset);
-
 /* page fault handler and mmap fault address(virtual) to physical memory. */
 int exynos_drm_gem_fault(struct vm_fault *vmf);
 
diff --git a/drivers/gpu/drm/exynos/exynos_drm_mic.c b/drivers/gpu/drm/exynos/exynos_drm_mic.c
index e45720543a45..ba4a32b132ba 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_mic.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_mic.c
@@ -21,9 +21,12 @@
 #include <linux/component.h>
 #include <linux/pm_runtime.h>
 #include <drm/drmP.h>
+#include <drm/drm_encoder.h>
 #include <linux/mfd/syscon.h>
 #include <linux/regmap.h>
 
+#include "exynos_drm_drv.h"
+
 /* Sysreg registers for MIC */
 #define DSD_CFG_MUX	0x1004
 #define MIC0_RGB_MUX	(1 << 0)
@@ -85,12 +88,6 @@
 
 #define MIC_BS_SIZE_2D(x)	((x) & 0x3fff)
 
-enum {
-	ENDPOINT_DECON_NODE,
-	ENDPOINT_DSI_NODE,
-	NUM_ENDPOINTS
-};
-
 static char *clk_names[] = { "pclk_mic0", "sclk_rgb_vclk_to_mic0" };
 #define NUM_CLKS		ARRAY_SIZE(clk_names)
 static DEFINE_MUTEX(mic_mutex);
@@ -229,36 +226,6 @@ static void mic_set_reg_on(struct exynos_mic *mic, bool enable)
 	writel(reg, mic->reg + MIC_OP);
 }
 
-static int parse_dt(struct exynos_mic *mic)
-{
-	int ret = 0, i, j;
-	struct device_node *remote_node;
-	struct device_node *nodes[3];
-
-	/*
-	 * The order of endpoints does matter.
-	 * The first node must be for decon and the second one must be for dsi.
-	 */
-	for (i = 0, j = 0; i < NUM_ENDPOINTS; i++) {
-		remote_node = of_graph_get_remote_node(mic->dev->of_node, i, 0);
-		if (!remote_node) {
-			ret = -EPIPE;
-			goto exit;
-		}
-		nodes[j++] = remote_node;
-
-		if (i == ENDPOINT_DECON_NODE &&
-			of_get_child_by_name(remote_node, "i80-if-timings"))
-			mic->i80_mode = 1;
-	}
-
-exit:
-	while (--j > -1)
-		of_node_put(nodes[j]);
-
-	return ret;
-}
-
 static void mic_disable(struct drm_bridge *bridge) { }
 
 static void mic_post_disable(struct drm_bridge *bridge)
@@ -286,6 +253,7 @@ static void mic_mode_set(struct drm_bridge *bridge,
 
 	mutex_lock(&mic_mutex);
 	drm_display_mode_to_videomode(mode, &mic->vm);
+	mic->i80_mode = to_exynos_crtc(bridge->encoder->crtc)->i80_mode;
 	mutex_unlock(&mic_mutex);
 }
 
@@ -340,16 +308,10 @@ static int exynos_mic_bind(struct device *dev, struct device *master,
 			   void *data)
 {
 	struct exynos_mic *mic = dev_get_drvdata(dev);
-	int ret;
 
-	mic->bridge.funcs = &mic_bridge_funcs;
-	mic->bridge.of_node = dev->of_node;
 	mic->bridge.driver_private = mic;
-	ret = drm_bridge_add(&mic->bridge);
-	if (ret)
-		DRM_ERROR("mic: Failed to add MIC to the global bridge list\n");
 
-	return ret;
+	return 0;
 }
 
 static void exynos_mic_unbind(struct device *dev, struct device *master,
@@ -365,8 +327,6 @@ static void exynos_mic_unbind(struct device *dev, struct device *master,
 
 already_disabled:
 	mutex_unlock(&mic_mutex);
-
-	drm_bridge_remove(&mic->bridge);
 }
 
 static const struct component_ops exynos_mic_component_ops = {
@@ -425,10 +385,6 @@ static int exynos_mic_probe(struct platform_device *pdev)
 
 	mic->dev = dev;
 
-	ret = parse_dt(mic);
-	if (ret)
-		goto err;
-
 	ret = of_address_to_resource(dev->of_node, 0, &res);
 	if (ret) {
 		DRM_ERROR("mic: Failed to get mem region for MIC\n");
@@ -461,6 +417,15 @@ static int exynos_mic_probe(struct platform_device *pdev)
 
 	platform_set_drvdata(pdev, mic);
 
+	mic->bridge.funcs = &mic_bridge_funcs;
+	mic->bridge.of_node = dev->of_node;
+
+	ret = drm_bridge_add(&mic->bridge);
+	if (ret) {
+		DRM_ERROR("mic: Failed to add MIC to the global bridge list\n");
+		return ret;
+	}
+
 	pm_runtime_enable(dev);
 
 	ret = component_add(dev, &exynos_mic_component_ops);
@@ -479,8 +444,13 @@ err:
 
 static int exynos_mic_remove(struct platform_device *pdev)
 {
+	struct exynos_mic *mic = platform_get_drvdata(pdev);
+
 	component_del(&pdev->dev, &exynos_mic_component_ops);
 	pm_runtime_disable(&pdev->dev);
+
+	drm_bridge_remove(&mic->bridge);
+
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/exynos/exynos_drm_plane.c b/drivers/gpu/drm/exynos/exynos_drm_plane.c
index 8de74009dee4..d2a90dae5c71 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_plane.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_plane.c
@@ -179,6 +179,29 @@ static struct drm_plane_funcs exynos_plane_funcs = {
 };
 
 static int
+exynos_drm_plane_check_format(const struct exynos_drm_plane_config *config,
+			      struct exynos_drm_plane_state *state)
+{
+	struct drm_framebuffer *fb = state->base.fb;
+
+	switch (fb->modifier) {
+	case DRM_FORMAT_MOD_SAMSUNG_64_32_TILE:
+		if (!(config->capabilities & EXYNOS_DRM_PLANE_CAP_TILE))
+			return -ENOTSUPP;
+		break;
+
+	case DRM_FORMAT_MOD_LINEAR:
+		break;
+
+	default:
+		DRM_ERROR("unsupported pixel format modifier");
+		return -ENOTSUPP;
+	}
+
+	return 0;
+}
+
+static int
 exynos_drm_plane_check_size(const struct exynos_drm_plane_config *config,
 			    struct exynos_drm_plane_state *state)
 {
@@ -222,6 +245,10 @@ static int exynos_plane_atomic_check(struct drm_plane *plane,
 	/* translate state into exynos_state */
 	exynos_plane_mode_set(exynos_state);
 
+	ret = exynos_drm_plane_check_format(exynos_plane->config, exynos_state);
+	if (ret)
+		return ret;
+
 	ret = exynos_drm_plane_check_size(exynos_plane->config, exynos_state);
 	return ret;
 }
diff --git a/drivers/gpu/drm/exynos/exynos_drm_vidi.c b/drivers/gpu/drm/exynos/exynos_drm_vidi.c
index 9186a654c3b5..53e03f8af3d5 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_vidi.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_vidi.c
@@ -381,7 +381,7 @@ static int vidi_bind(struct device *dev, struct device *master, void *data)
 	struct exynos_drm_plane *exynos_plane;
 	struct exynos_drm_plane_config plane_config = { 0 };
 	unsigned int i;
-	int pipe, ret;
+	int ret;
 
 	ctx->drm_dev = drm_dev;
 
@@ -406,20 +406,15 @@ static int vidi_bind(struct device *dev, struct device *master, void *data)
 		return PTR_ERR(ctx->crtc);
 	}
 
-	pipe = exynos_drm_crtc_get_pipe_from_type(drm_dev,
-						  EXYNOS_DISPLAY_TYPE_VIDI);
-	if (pipe < 0)
-		return pipe;
-
-	encoder->possible_crtcs = 1 << pipe;
-
-	DRM_DEBUG_KMS("possible_crtcs = 0x%x\n", encoder->possible_crtcs);
-
 	drm_encoder_init(drm_dev, encoder, &exynos_vidi_encoder_funcs,
 			 DRM_MODE_ENCODER_TMDS, NULL);
 
 	drm_encoder_helper_add(encoder, &exynos_vidi_encoder_helper_funcs);
 
+	ret = exynos_drm_set_possible_crtcs(encoder, EXYNOS_DISPLAY_TYPE_VIDI);
+	if (ret < 0)
+		return ret;
+
 	ret = vidi_create_connector(encoder);
 	if (ret) {
 		DRM_ERROR("failed to create connector ret = %d\n", ret);
diff --git a/drivers/gpu/drm/exynos/exynos_hdmi.c b/drivers/gpu/drm/exynos/exynos_hdmi.c
index 0e2a472c3021..214fa5e51963 100644
--- a/drivers/gpu/drm/exynos/exynos_hdmi.c
+++ b/drivers/gpu/drm/exynos/exynos_hdmi.c
@@ -1500,8 +1500,6 @@ static void hdmi_disable(struct drm_encoder *encoder)
 	 */
 	cancel_delayed_work(&hdata->hotplug_work);
 	cec_notifier_set_phys_addr(hdata->notifier, CEC_PHYS_ADDR_INVALID);
-
-	hdmiphy_disable(hdata);
 }
 
 static const struct drm_encoder_helper_funcs exynos_hdmi_encoder_helper_funcs = {
@@ -1675,7 +1673,7 @@ static int hdmi_resources_init(struct hdmi_context *hdata)
 	return hdmi_bridge_init(hdata);
 }
 
-static struct of_device_id hdmi_match_types[] = {
+static const struct of_device_id hdmi_match_types[] = {
 	{
 		.compatible = "samsung,exynos4210-hdmi",
 		.data = &exynos4210_hdmi_driver_data,
@@ -1699,32 +1697,25 @@ static int hdmi_bind(struct device *dev, struct device *master, void *data)
 	struct drm_device *drm_dev = data;
 	struct hdmi_context *hdata = dev_get_drvdata(dev);
 	struct drm_encoder *encoder = &hdata->encoder;
-	struct exynos_drm_crtc *exynos_crtc;
-	struct drm_crtc *crtc;
-	int ret, pipe;
+	struct exynos_drm_crtc *crtc;
+	int ret;
 
 	hdata->drm_dev = drm_dev;
 
-	pipe = exynos_drm_crtc_get_pipe_from_type(drm_dev,
-						  EXYNOS_DISPLAY_TYPE_HDMI);
-	if (pipe < 0)
-		return pipe;
-
 	hdata->phy_clk.enable = hdmiphy_clk_enable;
 
-	crtc = drm_crtc_from_index(drm_dev, pipe);
-	exynos_crtc = to_exynos_crtc(crtc);
-	exynos_crtc->pipe_clk = &hdata->phy_clk;
-
-	encoder->possible_crtcs = 1 << pipe;
-
-	DRM_DEBUG_KMS("possible_crtcs = 0x%x\n", encoder->possible_crtcs);
-
 	drm_encoder_init(drm_dev, encoder, &exynos_hdmi_encoder_funcs,
 			 DRM_MODE_ENCODER_TMDS, NULL);
 
 	drm_encoder_helper_add(encoder, &exynos_hdmi_encoder_helper_funcs);
 
+	ret = exynos_drm_set_possible_crtcs(encoder, EXYNOS_DISPLAY_TYPE_HDMI);
+	if (ret < 0)
+		return ret;
+
+	crtc = exynos_drm_crtc_get_by_type(drm_dev, EXYNOS_DISPLAY_TYPE_HDMI);
+	crtc->pipe_clk = &hdata->phy_clk;
+
 	ret = hdmi_create_connector(encoder);
 	if (ret) {
 		DRM_ERROR("failed to create connector ret = %d\n", ret);
@@ -1933,8 +1924,7 @@ static int hdmi_remove(struct platform_device *pdev)
 	return 0;
 }
 
-#ifdef CONFIG_PM
-static int exynos_hdmi_suspend(struct device *dev)
+static int __maybe_unused exynos_hdmi_suspend(struct device *dev)
 {
 	struct hdmi_context *hdata = dev_get_drvdata(dev);
 
@@ -1943,7 +1933,7 @@ static int exynos_hdmi_suspend(struct device *dev)
 	return 0;
 }
 
-static int exynos_hdmi_resume(struct device *dev)
+static int __maybe_unused exynos_hdmi_resume(struct device *dev)
 {
 	struct hdmi_context *hdata = dev_get_drvdata(dev);
 	int ret;
@@ -1954,7 +1944,6 @@ static int exynos_hdmi_resume(struct device *dev)
 
 	return 0;
 }
-#endif
 
 static const struct dev_pm_ops exynos_hdmi_pm_ops = {
 	SET_RUNTIME_PM_OPS(exynos_hdmi_suspend, exynos_hdmi_resume, NULL)
diff --git a/drivers/gpu/drm/exynos/exynos_mixer.c b/drivers/gpu/drm/exynos/exynos_mixer.c
index 6bed4f3ffcd6..002755415e00 100644
--- a/drivers/gpu/drm/exynos/exynos_mixer.c
+++ b/drivers/gpu/drm/exynos/exynos_mixer.c
@@ -148,7 +148,8 @@ static const struct exynos_drm_plane_config plane_configs[MIXER_WIN_NR] = {
 		.pixel_formats = vp_formats,
 		.num_pixel_formats = ARRAY_SIZE(vp_formats),
 		.capabilities = EXYNOS_DRM_PLANE_CAP_SCALE |
-				EXYNOS_DRM_PLANE_CAP_ZPOS,
+				EXYNOS_DRM_PLANE_CAP_ZPOS |
+				EXYNOS_DRM_PLANE_CAP_TILE,
 	},
 };
 
@@ -483,29 +484,18 @@ static void vp_video_buffer(struct mixer_context *ctx,
 	unsigned int priority = state->base.normalized_zpos + 1;
 	unsigned long flags;
 	dma_addr_t luma_addr[2], chroma_addr[2];
-	bool tiled_mode = false;
-	bool crcb_mode = false;
+	bool is_tiled, is_nv21;
 	u32 val;
 
-	switch (fb->format->format) {
-	case DRM_FORMAT_NV12:
-		crcb_mode = false;
-		break;
-	case DRM_FORMAT_NV21:
-		crcb_mode = true;
-		break;
-	default:
-		DRM_ERROR("pixel format for vp is wrong [%d].\n",
-				fb->format->format);
-		return;
-	}
+	is_nv21 = (fb->format->format == DRM_FORMAT_NV21);
+	is_tiled = (fb->modifier == DRM_FORMAT_MOD_SAMSUNG_64_32_TILE);
 
 	luma_addr[0] = exynos_drm_fb_dma_addr(fb, 0);
 	chroma_addr[0] = exynos_drm_fb_dma_addr(fb, 1);
 
 	if (mode->flags & DRM_MODE_FLAG_INTERLACE) {
 		__set_bit(MXR_BIT_INTERLACE, &ctx->flags);
-		if (tiled_mode) {
+		if (is_tiled) {
 			luma_addr[1] = luma_addr[0] + 0x40;
 			chroma_addr[1] = chroma_addr[0] + 0x40;
 		} else {
@@ -525,14 +515,14 @@ static void vp_video_buffer(struct mixer_context *ctx,
 	vp_reg_writemask(res, VP_MODE, val, VP_MODE_LINE_SKIP);
 
 	/* setup format */
-	val = (crcb_mode ? VP_MODE_NV21 : VP_MODE_NV12);
-	val |= (tiled_mode ? VP_MODE_MEM_TILED : VP_MODE_MEM_LINEAR);
+	val = (is_nv21 ? VP_MODE_NV21 : VP_MODE_NV12);
+	val |= (is_tiled ? VP_MODE_MEM_TILED : VP_MODE_MEM_LINEAR);
 	vp_reg_writemask(res, VP_MODE, val, VP_MODE_FMT_MASK);
 
 	/* setting size of input image */
 	vp_reg_write(res, VP_IMG_SIZE_Y, VP_IMG_HSIZE(fb->pitches[0]) |
 		VP_IMG_VSIZE(fb->height));
-	/* chroma height has to reduced by 2 to avoid chroma distorions */
+	/* chroma plane for NV12/NV21 is half the height of the luma plane */
 	vp_reg_write(res, VP_IMG_SIZE_C, VP_IMG_HSIZE(fb->pitches[0]) |
 		VP_IMG_VSIZE(fb->height / 2));
 
@@ -594,7 +584,7 @@ static void mixer_graph_buffer(struct mixer_context *ctx,
 	unsigned long flags;
 	unsigned int win = plane->index;
 	unsigned int x_ratio = 0, y_ratio = 0;
-	unsigned int src_x_offset, src_y_offset, dst_x_offset, dst_y_offset;
+	unsigned int dst_x_offset, dst_y_offset;
 	dma_addr_t dma_addr;
 	unsigned int fmt;
 	u32 val;
@@ -616,12 +606,9 @@ static void mixer_graph_buffer(struct mixer_context *ctx,
 
 	case DRM_FORMAT_XRGB8888:
 	case DRM_FORMAT_ARGB8888:
+	default:
 		fmt = MXR_FORMAT_ARGB8888;
 		break;
-
-	default:
-		DRM_DEBUG_KMS("pixelformat unsupported by mixer\n");
-		return;
 	}
 
 	/* ratio is already checked by common plane code */
@@ -631,12 +618,10 @@ static void mixer_graph_buffer(struct mixer_context *ctx,
 	dst_x_offset = state->crtc.x;
 	dst_y_offset = state->crtc.y;
 
-	/* converting dma address base and source offset */
+	/* translate dma address base s.t. the source image offset is zero */
 	dma_addr = exynos_drm_fb_dma_addr(fb, 0)
 		+ (state->src.x * fb->format->cpp[0])
 		+ (state->src.y * fb->pitches[0]);
-	src_x_offset = 0;
-	src_y_offset = 0;
 
 	if (mode->flags & DRM_MODE_FLAG_INTERLACE)
 		__set_bit(MXR_BIT_INTERLACE, &ctx->flags);
@@ -667,11 +652,6 @@ static void mixer_graph_buffer(struct mixer_context *ctx,
 	val |= MXR_GRP_WH_V_SCALE(y_ratio);
 	mixer_reg_write(res, MXR_GRAPHIC_WH(win), val);
 
-	/* setup offsets in source image */
-	val  = MXR_GRP_SXY_SX(src_x_offset);
-	val |= MXR_GRP_SXY_SY(src_y_offset);
-	mixer_reg_write(res, MXR_GRAPHIC_SXY(win), val);
-
 	/* setup offsets in display image */
 	val  = MXR_GRP_DXY_DX(dst_x_offset);
 	val |= MXR_GRP_DXY_DY(dst_y_offset);
@@ -748,6 +728,10 @@ static void mixer_win_reset(struct mixer_context *ctx)
 	if (test_bit(MXR_BIT_VP_ENABLED, &ctx->flags))
 		mixer_reg_writemask(res, MXR_CFG, 0, MXR_CFG_VP_ENABLE);
 
+	/* set all source image offsets to zero */
+	mixer_reg_write(res, MXR_GRAPHIC_SXY(0), 0);
+	mixer_reg_write(res, MXR_GRAPHIC_SXY(1), 0);
+
 	spin_unlock_irqrestore(&res->reg_slock, flags);
 }
 
@@ -1094,28 +1078,28 @@ static const struct exynos_drm_crtc_ops mixer_crtc_ops = {
 	.atomic_check		= mixer_atomic_check,
 };
 
-static struct mixer_drv_data exynos5420_mxr_drv_data = {
+static const struct mixer_drv_data exynos5420_mxr_drv_data = {
 	.version = MXR_VER_128_0_0_184,
 	.is_vp_enabled = 0,
 };
 
-static struct mixer_drv_data exynos5250_mxr_drv_data = {
+static const struct mixer_drv_data exynos5250_mxr_drv_data = {
 	.version = MXR_VER_16_0_33_0,
 	.is_vp_enabled = 0,
 };
 
-static struct mixer_drv_data exynos4212_mxr_drv_data = {
+static const struct mixer_drv_data exynos4212_mxr_drv_data = {
 	.version = MXR_VER_0_0_0_16,
 	.is_vp_enabled = 1,
 };
 
-static struct mixer_drv_data exynos4210_mxr_drv_data = {
+static const struct mixer_drv_data exynos4210_mxr_drv_data = {
 	.version = MXR_VER_0_0_0_16,
 	.is_vp_enabled = 1,
 	.has_sclk = 1,
 };
 
-static struct of_device_id mixer_match_types[] = {
+static const struct of_device_id mixer_match_types[] = {
 	{
 		.compatible = "samsung,exynos4210-mixer",
 		.data	= &exynos4210_mxr_drv_data,
diff --git a/drivers/gpu/drm/gma500/gem.c b/drivers/gpu/drm/gma500/gem.c
index 7da061aab729..131239759a75 100644
--- a/drivers/gpu/drm/gma500/gem.c
+++ b/drivers/gpu/drm/gma500/gem.c
@@ -48,36 +48,6 @@ int psb_gem_get_aperture(struct drm_device *dev, void *data,
 }
 
 /**
- *	psb_gem_dumb_map_gtt	-	buffer mapping for dumb interface
- *	@file: our drm client file
- *	@dev: drm device
- *	@handle: GEM handle to the object (from dumb_create)
- *
- *	Do the necessary setup to allow the mapping of the frame buffer
- *	into user memory. We don't have to do much here at the moment.
- */
-int psb_gem_dumb_map_gtt(struct drm_file *file, struct drm_device *dev,
-			 uint32_t handle, uint64_t *offset)
-{
-	int ret = 0;
-	struct drm_gem_object *obj;
-
-	/* GEM does all our handle to object mapping */
-	obj = drm_gem_object_lookup(file, handle);
-	if (obj == NULL)
-		return -ENOENT;
-
-	/* Make it mmapable */
-	ret = drm_gem_create_mmap_offset(obj);
-	if (ret)
-		goto out;
-	*offset = drm_vma_node_offset_addr(&obj->vma_node);
-out:
-	drm_gem_object_unreference_unlocked(obj);
-	return ret;
-}
-
-/**
  *	psb_gem_create		-	create a mappable object
  *	@file: the DRM file of the client
  *	@dev: our device
diff --git a/drivers/gpu/drm/gma500/mdfld_dsi_pkg_sender.c b/drivers/gpu/drm/gma500/mdfld_dsi_pkg_sender.c
index 1616af209bfc..c50534c923df 100644
--- a/drivers/gpu/drm/gma500/mdfld_dsi_pkg_sender.c
+++ b/drivers/gpu/drm/gma500/mdfld_dsi_pkg_sender.c
@@ -520,7 +520,7 @@ static int __read_panel_data(struct mdfld_dsi_pkg_sender *sender, u8 data_type,
 			u8 *data, u16 len, u32 *data_out, u16 len_out, bool hs)
 {
 	unsigned long flags;
-	struct drm_device *dev = sender->dev;
+	struct drm_device *dev;
 	int i;
 	u32 gen_data_reg;
 	int retry = MDFLD_DSI_READ_MAX_COUNT;
@@ -530,6 +530,8 @@ static int __read_panel_data(struct mdfld_dsi_pkg_sender *sender, u8 data_type,
 		return -EINVAL;
 	}
 
+	dev = sender->dev;
+
 	/**
 	 * do reading.
 	 * 0) send out generic read request
diff --git a/drivers/gpu/drm/gma500/psb_drv.c b/drivers/gpu/drm/gma500/psb_drv.c
index 747c06b227c5..37a3be71acd9 100644
--- a/drivers/gpu/drm/gma500/psb_drv.c
+++ b/drivers/gpu/drm/gma500/psb_drv.c
@@ -494,8 +494,6 @@ static struct drm_driver driver = {
 	.gem_vm_ops = &psb_gem_vm_ops,
 
 	.dumb_create = psb_gem_dumb_create,
-	.dumb_map_offset = psb_gem_dumb_map_gtt,
-	.dumb_destroy = drm_gem_dumb_destroy,
 	.ioctls = psb_ioctls,
 	.fops = &psb_gem_fops,
 	.name = DRIVER_NAME,
diff --git a/drivers/gpu/drm/gma500/psb_drv.h b/drivers/gpu/drm/gma500/psb_drv.h
index 83667087d6e5..821497dbd3fc 100644
--- a/drivers/gpu/drm/gma500/psb_drv.h
+++ b/drivers/gpu/drm/gma500/psb_drv.h
@@ -750,8 +750,6 @@ extern int psb_gem_get_aperture(struct drm_device *dev, void *data,
 			struct drm_file *file);
 extern int psb_gem_dumb_create(struct drm_file *file, struct drm_device *dev,
 			struct drm_mode_create_dumb *args);
-extern int psb_gem_dumb_map_gtt(struct drm_file *file, struct drm_device *dev,
-			uint32_t handle, uint64_t *offset);
 extern int psb_gem_fault(struct vm_fault *vmf);
 
 /* psb_device.c */
diff --git a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_fbdev.c b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_fbdev.c
index 9740eed9231a..b92595c477ef 100644
--- a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_fbdev.c
+++ b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_fbdev.c
@@ -157,7 +157,7 @@ out_unpin_bo:
 out_unreserve_ttm_bo:
 	ttm_bo_unreserve(&bo->bo);
 out_unref_gem:
-	drm_gem_object_unreference_unlocked(gobj);
+	drm_gem_object_put_unlocked(gobj);
 
 	return ret;
 }
@@ -172,7 +172,7 @@ static void hibmc_fbdev_destroy(struct hibmc_fbdev *fbdev)
 	drm_fb_helper_fini(fbh);
 
 	if (gfb)
-		drm_framebuffer_unreference(&gfb->fb);
+		drm_framebuffer_put(&gfb->fb);
 }
 
 static const struct drm_fb_helper_funcs hibmc_fbdev_helper_funcs = {
diff --git a/drivers/gpu/drm/hisilicon/hibmc/hibmc_ttm.c b/drivers/gpu/drm/hisilicon/hibmc/hibmc_ttm.c
index ac457c779caa..3518167a7dc4 100644
--- a/drivers/gpu/drm/hisilicon/hibmc/hibmc_ttm.c
+++ b/drivers/gpu/drm/hisilicon/hibmc/hibmc_ttm.c
@@ -444,7 +444,7 @@ int hibmc_dumb_create(struct drm_file *file, struct drm_device *dev,
 	}
 
 	ret = drm_gem_handle_create(file, gobj, &handle);
-	drm_gem_object_unreference_unlocked(gobj);
+	drm_gem_object_put_unlocked(gobj);
 	if (ret) {
 		DRM_ERROR("failed to unreference GEM object: %d\n", ret);
 		return ret;
@@ -479,7 +479,7 @@ int hibmc_dumb_mmap_offset(struct drm_file *file, struct drm_device *dev,
 	bo = gem_to_hibmc_bo(obj);
 	*offset = hibmc_bo_mmap_offset(bo);
 
-	drm_gem_object_unreference_unlocked(obj);
+	drm_gem_object_put_unlocked(obj);
 	return 0;
 }
 
@@ -487,7 +487,7 @@ static void hibmc_user_framebuffer_destroy(struct drm_framebuffer *fb)
 {
 	struct hibmc_framebuffer *hibmc_fb = to_hibmc_framebuffer(fb);
 
-	drm_gem_object_unreference_unlocked(hibmc_fb->obj);
+	drm_gem_object_put_unlocked(hibmc_fb->obj);
 	drm_framebuffer_cleanup(fb);
 	kfree(hibmc_fb);
 }
@@ -543,7 +543,7 @@ hibmc_user_framebuffer_create(struct drm_device *dev,
 
 	hibmc_fb = hibmc_framebuffer_init(dev, mode_cmd, obj);
 	if (IS_ERR(hibmc_fb)) {
-		drm_gem_object_unreference_unlocked(obj);
+		drm_gem_object_put_unlocked(obj);
 		return ERR_PTR((long)hibmc_fb);
 	}
 	return &hibmc_fb->fb;
diff --git a/drivers/gpu/drm/hisilicon/kirin/dw_drm_dsi.c b/drivers/gpu/drm/hisilicon/kirin/dw_drm_dsi.c
index f77dcfaade6c..b4c7af3ab6ae 100644
--- a/drivers/gpu/drm/hisilicon/kirin/dw_drm_dsi.c
+++ b/drivers/gpu/drm/hisilicon/kirin/dw_drm_dsi.c
@@ -603,6 +603,72 @@ static void dsi_encoder_enable(struct drm_encoder *encoder)
 	dsi->enable = true;
 }
 
+static enum drm_mode_status dsi_encoder_phy_mode_valid(
+					struct drm_encoder *encoder,
+					const struct drm_display_mode *mode)
+{
+	struct dw_dsi *dsi = encoder_to_dsi(encoder);
+	struct mipi_phy_params phy;
+	u32 bpp = mipi_dsi_pixel_format_to_bpp(dsi->format);
+	u32 req_kHz, act_kHz, lane_byte_clk_kHz;
+
+	/* Calculate the lane byte clk using the adjusted mode clk */
+	memset(&phy, 0, sizeof(phy));
+	req_kHz = mode->clock * bpp / dsi->lanes;
+	act_kHz = dsi_calc_phy_rate(req_kHz, &phy);
+	lane_byte_clk_kHz = act_kHz / 8;
+
+	DRM_DEBUG_DRIVER("Checking mode %ix%i-%i@%i clock: %i...",
+			mode->hdisplay, mode->vdisplay, bpp,
+			drm_mode_vrefresh(mode), mode->clock);
+
+	/*
+	 * Make sure the adjusted mode clock and the lane byte clk
+	 * have a common denominator base frequency
+	 */
+	if (mode->clock/dsi->lanes == lane_byte_clk_kHz/3) {
+		DRM_DEBUG_DRIVER("OK!\n");
+		return MODE_OK;
+	}
+
+	DRM_DEBUG_DRIVER("BAD!\n");
+	return MODE_BAD;
+}
+
+static enum drm_mode_status dsi_encoder_mode_valid(struct drm_encoder *encoder,
+					const struct drm_display_mode *mode)
+
+{
+	const struct drm_crtc_helper_funcs *crtc_funcs = NULL;
+	struct drm_crtc *crtc = NULL;
+	struct drm_display_mode adj_mode;
+	enum drm_mode_status ret;
+
+	/*
+	 * The crtc might adjust the mode, so go through the
+	 * possible crtcs (technically just one) and call
+	 * mode_fixup to figure out the adjusted mode before we
+	 * validate it.
+	 */
+	drm_for_each_crtc(crtc, encoder->dev) {
+		/*
+		 * reset adj_mode to the mode value each time,
+		 * so we don't adjust the mode twice
+		 */
+		drm_mode_copy(&adj_mode, mode);
+
+		crtc_funcs = crtc->helper_private;
+		if (crtc_funcs && crtc_funcs->mode_fixup)
+			if (!crtc_funcs->mode_fixup(crtc, mode, &adj_mode))
+				return MODE_BAD;
+
+		ret = dsi_encoder_phy_mode_valid(encoder, &adj_mode);
+		if (ret != MODE_OK)
+			return ret;
+	}
+	return MODE_OK;
+}
+
 static void dsi_encoder_mode_set(struct drm_encoder *encoder,
 				 struct drm_display_mode *mode,
 				 struct drm_display_mode *adj_mode)
@@ -622,6 +688,7 @@ static int dsi_encoder_atomic_check(struct drm_encoder *encoder,
 
 static const struct drm_encoder_helper_funcs dw_encoder_helper_funcs = {
 	.atomic_check	= dsi_encoder_atomic_check,
+	.mode_valid	= dsi_encoder_mode_valid,
 	.mode_set	= dsi_encoder_mode_set,
 	.enable		= dsi_encoder_enable,
 	.disable	= dsi_encoder_disable
diff --git a/drivers/gpu/drm/hisilicon/kirin/kirin_drm_ade.c b/drivers/gpu/drm/hisilicon/kirin/kirin_drm_ade.c
index 39f7d15673ed..9823477b1855 100644
--- a/drivers/gpu/drm/hisilicon/kirin/kirin_drm_ade.c
+++ b/drivers/gpu/drm/hisilicon/kirin/kirin_drm_ade.c
@@ -178,6 +178,19 @@ static void ade_init(struct ade_hw_ctx *ctx)
 			FRM_END_START_MASK, REG_EFFECTIVE_IN_ADEEN_FRMEND);
 }
 
+static bool ade_crtc_mode_fixup(struct drm_crtc *crtc,
+				const struct drm_display_mode *mode,
+				struct drm_display_mode *adjusted_mode)
+{
+	struct ade_crtc *acrtc = to_ade_crtc(crtc);
+	struct ade_hw_ctx *ctx = acrtc->ctx;
+
+	adjusted_mode->clock =
+		clk_round_rate(ctx->ade_pix_clk, mode->clock * 1000) / 1000;
+	return true;
+}
+
+
 static void ade_set_pix_clk(struct ade_hw_ctx *ctx,
 			    struct drm_display_mode *mode,
 			    struct drm_display_mode *adj_mode)
@@ -555,6 +568,7 @@ static void ade_crtc_atomic_flush(struct drm_crtc *crtc,
 }
 
 static const struct drm_crtc_helper_funcs ade_crtc_helper_funcs = {
+	.mode_fixup	= ade_crtc_mode_fixup,
 	.mode_set_nofb	= ade_crtc_mode_set_nofb,
 	.atomic_begin	= ade_crtc_atomic_begin,
 	.atomic_flush	= ade_crtc_atomic_flush,
diff --git a/drivers/gpu/drm/hisilicon/kirin/kirin_drm_drv.c b/drivers/gpu/drm/hisilicon/kirin/kirin_drm_drv.c
index 79fcce76f2ad..e27352ca26c4 100644
--- a/drivers/gpu/drm/hisilicon/kirin/kirin_drm_drv.c
+++ b/drivers/gpu/drm/hisilicon/kirin/kirin_drm_drv.c
@@ -157,8 +157,6 @@ static struct drm_driver kirin_drm_driver = {
 	.gem_free_object_unlocked = drm_gem_cma_free_object,
 	.gem_vm_ops		= &drm_gem_cma_vm_ops,
 	.dumb_create		= kirin_gem_cma_dumb_create,
-	.dumb_map_offset	= drm_gem_cma_dumb_map_offset,
-	.dumb_destroy		= drm_gem_dumb_destroy,
 
 	.prime_handle_to_fd	= drm_gem_prime_handle_to_fd,
 	.prime_fd_to_handle	= drm_gem_prime_fd_to_handle,
diff --git a/drivers/gpu/drm/i915/gvt/display.c b/drivers/gpu/drm/i915/gvt/display.c
index 24cc4b012e93..3c318439a659 100644
--- a/drivers/gpu/drm/i915/gvt/display.c
+++ b/drivers/gpu/drm/i915/gvt/display.c
@@ -323,27 +323,27 @@ void intel_gvt_check_vblank_emulation(struct intel_gvt *gvt)
 {
 	struct intel_gvt_irq *irq = &gvt->irq;
 	struct intel_vgpu *vgpu;
-	bool have_enabled_pipe = false;
 	int pipe, id;
 
 	if (WARN_ON(!mutex_is_locked(&gvt->lock)))
 		return;
 
-	hrtimer_cancel(&irq->vblank_timer.timer);
-
 	for_each_active_vgpu(gvt, vgpu, id) {
 		for (pipe = 0; pipe < I915_MAX_PIPES; pipe++) {
-			have_enabled_pipe =
-				pipe_is_enabled(vgpu, pipe);
-			if (have_enabled_pipe)
-				break;
+			if (pipe_is_enabled(vgpu, pipe))
+				goto out;
 		}
 	}
 
-	if (have_enabled_pipe)
-		hrtimer_start(&irq->vblank_timer.timer,
-			ktime_add_ns(ktime_get(), irq->vblank_timer.period),
-			HRTIMER_MODE_ABS);
+	/* all the pipes are disabled */
+	hrtimer_cancel(&irq->vblank_timer.timer);
+	return;
+
+out:
+	hrtimer_start(&irq->vblank_timer.timer,
+		ktime_add_ns(ktime_get(), irq->vblank_timer.period),
+		HRTIMER_MODE_ABS);
+
 }
 
 static void emulate_vblank_on_pipe(struct intel_vgpu *vgpu, int pipe)
diff --git a/drivers/gpu/drm/i915/gvt/execlist.c b/drivers/gpu/drm/i915/gvt/execlist.c
index df11f69edc05..91b4300f3b39 100644
--- a/drivers/gpu/drm/i915/gvt/execlist.c
+++ b/drivers/gpu/drm/i915/gvt/execlist.c
@@ -46,6 +46,8 @@
 #define same_context(a, b) (((a)->context_id == (b)->context_id) && \
 		((a)->lrca == (b)->lrca))
 
+static void clean_workloads(struct intel_vgpu *vgpu, unsigned long engine_mask);
+
 static int context_switch_events[] = {
 	[RCS] = RCS_AS_CONTEXT_SWITCH,
 	[BCS] = BCS_AS_CONTEXT_SWITCH,
@@ -499,10 +501,10 @@ static void release_shadow_wa_ctx(struct intel_shadow_wa_ctx *wa_ctx)
 static int complete_execlist_workload(struct intel_vgpu_workload *workload)
 {
 	struct intel_vgpu *vgpu = workload->vgpu;
-	struct intel_vgpu_execlist *execlist =
-		&vgpu->execlist[workload->ring_id];
+	int ring_id = workload->ring_id;
+	struct intel_vgpu_execlist *execlist = &vgpu->execlist[ring_id];
 	struct intel_vgpu_workload *next_workload;
-	struct list_head *next = workload_q_head(vgpu, workload->ring_id)->next;
+	struct list_head *next = workload_q_head(vgpu, ring_id)->next;
 	bool lite_restore = false;
 	int ret;
 
@@ -512,10 +514,25 @@ static int complete_execlist_workload(struct intel_vgpu_workload *workload)
 	release_shadow_batch_buffer(workload);
 	release_shadow_wa_ctx(&workload->wa_ctx);
 
-	if (workload->status || vgpu->resetting)
+	if (workload->status || (vgpu->resetting_eng & ENGINE_MASK(ring_id))) {
+		/* if workload->status is not successful means HW GPU
+		 * has occurred GPU hang or something wrong with i915/GVT,
+		 * and GVT won't inject context switch interrupt to guest.
+		 * So this error is a vGPU hang actually to the guest.
+		 * According to this we should emunlate a vGPU hang. If
+		 * there are pending workloads which are already submitted
+		 * from guest, we should clean them up like HW GPU does.
+		 *
+		 * if it is in middle of engine resetting, the pending
+		 * workloads won't be submitted to HW GPU and will be
+		 * cleaned up during the resetting process later, so doing
+		 * the workload clean up here doesn't have any impact.
+		 **/
+		clean_workloads(vgpu, ENGINE_MASK(ring_id));
 		goto out;
+	}
 
-	if (!list_empty(workload_q_head(vgpu, workload->ring_id))) {
+	if (!list_empty(workload_q_head(vgpu, ring_id))) {
 		struct execlist_ctx_descriptor_format *this_desc, *next_desc;
 
 		next_workload = container_of(next,
diff --git a/drivers/gpu/drm/i915/gvt/firmware.c b/drivers/gpu/drm/i915/gvt/firmware.c
index 5dad9298b2d5..a26c1705430e 100644
--- a/drivers/gpu/drm/i915/gvt/firmware.c
+++ b/drivers/gpu/drm/i915/gvt/firmware.c
@@ -72,11 +72,13 @@ static int expose_firmware_sysfs(struct intel_gvt *gvt)
 	struct intel_gvt_device_info *info = &gvt->device_info;
 	struct pci_dev *pdev = gvt->dev_priv->drm.pdev;
 	struct intel_gvt_mmio_info *e;
+	struct gvt_mmio_block *block = gvt->mmio.mmio_block;
+	int num = gvt->mmio.num_mmio_block;
 	struct gvt_firmware_header *h;
 	void *firmware;
 	void *p;
 	unsigned long size, crc32_start;
-	int i;
+	int i, j;
 	int ret;
 
 	size = sizeof(*h) + info->mmio_size + info->cfg_space_size;
@@ -105,6 +107,13 @@ static int expose_firmware_sysfs(struct intel_gvt *gvt)
 	hash_for_each(gvt->mmio.mmio_info_table, i, e, node)
 		*(u32 *)(p + e->offset) = I915_READ_NOTRACE(_MMIO(e->offset));
 
+	for (i = 0; i < num; i++, block++) {
+		for (j = 0; j < block->size; j += 4)
+			*(u32 *)(p + INTEL_GVT_MMIO_OFFSET(block->offset) + j) =
+				I915_READ_NOTRACE(_MMIO(INTEL_GVT_MMIO_OFFSET(
+							block->offset) + j));
+	}
+
 	memcpy(gvt->firmware.mmio, p, info->mmio_size);
 
 	crc32_start = offsetof(struct gvt_firmware_header, crc32) + 4;
diff --git a/drivers/gpu/drm/i915/gvt/gvt.h b/drivers/gpu/drm/i915/gvt/gvt.h
index ea736717e051..44b719eda8c4 100644
--- a/drivers/gpu/drm/i915/gvt/gvt.h
+++ b/drivers/gpu/drm/i915/gvt/gvt.h
@@ -149,7 +149,7 @@ struct intel_vgpu {
 	bool active;
 	bool pv_notified;
 	bool failsafe;
-	bool resetting;
+	unsigned int resetting_eng;
 	void *sched_data;
 	struct vgpu_sched_ctl sched_ctl;
 
@@ -196,6 +196,15 @@ struct intel_gvt_fence {
 	unsigned long vgpu_allocated_fence_num;
 };
 
+/* Special MMIO blocks. */
+struct gvt_mmio_block {
+	unsigned int device;
+	i915_reg_t   offset;
+	unsigned int size;
+	gvt_mmio_func read;
+	gvt_mmio_func write;
+};
+
 #define INTEL_GVT_MMIO_HASH_BITS 11
 
 struct intel_gvt_mmio {
@@ -215,6 +224,9 @@ struct intel_gvt_mmio {
 /* This reg could be accessed by unaligned address */
 #define F_UNALIGN	(1 << 6)
 
+	struct gvt_mmio_block *mmio_block;
+	unsigned int num_mmio_block;
+
 	DECLARE_HASHTABLE(mmio_info_table, INTEL_GVT_MMIO_HASH_BITS);
 	unsigned int num_tracked_mmio;
 };
diff --git a/drivers/gpu/drm/i915/gvt/handlers.c b/drivers/gpu/drm/i915/gvt/handlers.c
index 9220a756ecfc..3502a59166ff 100644
--- a/drivers/gpu/drm/i915/gvt/handlers.c
+++ b/drivers/gpu/drm/i915/gvt/handlers.c
@@ -2873,31 +2873,15 @@ static int init_skl_mmio_info(struct intel_gvt *gvt)
 	return 0;
 }
 
-/* Special MMIO blocks. */
-static struct gvt_mmio_block {
-	unsigned int device;
-	i915_reg_t   offset;
-	unsigned int size;
-	gvt_mmio_func read;
-	gvt_mmio_func write;
-} gvt_mmio_blocks[] = {
-	{D_SKL_PLUS, _MMIO(CSR_MMIO_START_RANGE), 0x3000, NULL, NULL},
-	{D_ALL, _MMIO(MCHBAR_MIRROR_BASE_SNB), 0x40000, NULL, NULL},
-	{D_ALL, _MMIO(VGT_PVINFO_PAGE), VGT_PVINFO_SIZE,
-		pvinfo_mmio_read, pvinfo_mmio_write},
-	{D_ALL, LGC_PALETTE(PIPE_A, 0), 1024, NULL, NULL},
-	{D_ALL, LGC_PALETTE(PIPE_B, 0), 1024, NULL, NULL},
-	{D_ALL, LGC_PALETTE(PIPE_C, 0), 1024, NULL, NULL},
-};
-
 static struct gvt_mmio_block *find_mmio_block(struct intel_gvt *gvt,
 					      unsigned int offset)
 {
 	unsigned long device = intel_gvt_get_device_type(gvt);
-	struct gvt_mmio_block *block = gvt_mmio_blocks;
+	struct gvt_mmio_block *block = gvt->mmio.mmio_block;
+	int num = gvt->mmio.num_mmio_block;
 	int i;
 
-	for (i = 0; i < ARRAY_SIZE(gvt_mmio_blocks); i++, block++) {
+	for (i = 0; i < num; i++, block++) {
 		if (!(device & block->device))
 			continue;
 		if (offset >= INTEL_GVT_MMIO_OFFSET(block->offset) &&
@@ -2928,6 +2912,17 @@ void intel_gvt_clean_mmio_info(struct intel_gvt *gvt)
 	gvt->mmio.mmio_attribute = NULL;
 }
 
+/* Special MMIO blocks. */
+static struct gvt_mmio_block mmio_blocks[] = {
+	{D_SKL_PLUS, _MMIO(CSR_MMIO_START_RANGE), 0x3000, NULL, NULL},
+	{D_ALL, _MMIO(MCHBAR_MIRROR_BASE_SNB), 0x40000, NULL, NULL},
+	{D_ALL, _MMIO(VGT_PVINFO_PAGE), VGT_PVINFO_SIZE,
+		pvinfo_mmio_read, pvinfo_mmio_write},
+	{D_ALL, LGC_PALETTE(PIPE_A, 0), 1024, NULL, NULL},
+	{D_ALL, LGC_PALETTE(PIPE_B, 0), 1024, NULL, NULL},
+	{D_ALL, LGC_PALETTE(PIPE_C, 0), 1024, NULL, NULL},
+};
+
 /**
  * intel_gvt_setup_mmio_info - setup MMIO information table for GVT device
  * @gvt: GVT device
@@ -2967,6 +2962,9 @@ int intel_gvt_setup_mmio_info(struct intel_gvt *gvt)
 			goto err;
 	}
 
+	gvt->mmio.mmio_block = mmio_blocks;
+	gvt->mmio.num_mmio_block = ARRAY_SIZE(mmio_blocks);
+
 	gvt_dbg_mmio("traced %u virtual mmio registers\n",
 		     gvt->mmio.num_tracked_mmio);
 	return 0;
@@ -3046,7 +3044,7 @@ int intel_vgpu_mmio_reg_rw(struct intel_vgpu *vgpu, unsigned int offset,
 	gvt_mmio_func func;
 	int ret;
 
-	if (WARN_ON(bytes > 4))
+	if (WARN_ON(bytes > 8))
 		return -EINVAL;
 
 	/*
diff --git a/drivers/gpu/drm/i915/gvt/scheduler.c b/drivers/gpu/drm/i915/gvt/scheduler.c
index 025aba8a72e0..391800d2067b 100644
--- a/drivers/gpu/drm/i915/gvt/scheduler.c
+++ b/drivers/gpu/drm/i915/gvt/scheduler.c
@@ -479,7 +479,8 @@ static void complete_current_workload(struct intel_gvt *gvt, int ring_id)
 
 		i915_gem_request_put(fetch_and_zero(&workload->req));
 
-		if (!workload->status && !vgpu->resetting) {
+		if (!workload->status && !(vgpu->resetting_eng &
+					   ENGINE_MASK(ring_id))) {
 			update_guest_context(workload);
 
 			for_each_set_bit(event, workload->pending_events,
diff --git a/drivers/gpu/drm/i915/gvt/vgpu.c b/drivers/gpu/drm/i915/gvt/vgpu.c
index 5896ead8529e..02c61a1ad56a 100644
--- a/drivers/gpu/drm/i915/gvt/vgpu.c
+++ b/drivers/gpu/drm/i915/gvt/vgpu.c
@@ -481,11 +481,13 @@ void intel_gvt_reset_vgpu_locked(struct intel_vgpu *vgpu, bool dmlr,
 {
 	struct intel_gvt *gvt = vgpu->gvt;
 	struct intel_gvt_workload_scheduler *scheduler = &gvt->scheduler;
+	unsigned int resetting_eng = dmlr ? ALL_ENGINES : engine_mask;
 
 	gvt_dbg_core("------------------------------------------\n");
 	gvt_dbg_core("resseting vgpu%d, dmlr %d, engine_mask %08x\n",
 		     vgpu->id, dmlr, engine_mask);
-	vgpu->resetting = true;
+
+	vgpu->resetting_eng = resetting_eng;
 
 	intel_vgpu_stop_schedule(vgpu);
 	/*
@@ -498,7 +500,7 @@ void intel_gvt_reset_vgpu_locked(struct intel_vgpu *vgpu, bool dmlr,
 		mutex_lock(&gvt->lock);
 	}
 
-	intel_vgpu_reset_execlist(vgpu, dmlr ? ALL_ENGINES : engine_mask);
+	intel_vgpu_reset_execlist(vgpu, resetting_eng);
 
 	/* full GPU reset or device model level reset */
 	if (engine_mask == ALL_ENGINES || dmlr) {
@@ -521,7 +523,7 @@ void intel_gvt_reset_vgpu_locked(struct intel_vgpu *vgpu, bool dmlr,
 		}
 	}
 
-	vgpu->resetting = false;
+	vgpu->resetting_eng = 0;
 	gvt_dbg_core("reset vgpu%d done\n", vgpu->id);
 	gvt_dbg_core("------------------------------------------\n");
 }
diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
index 7741c1c39037..8a9d37ac16d4 100644
--- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c
+++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
@@ -2141,9 +2141,7 @@ await_fence_array(struct i915_execbuffer *eb,
 		if (!(flags & I915_EXEC_FENCE_WAIT))
 			continue;
 
-		rcu_read_lock();
-		fence = dma_fence_get_rcu_safe(&syncobj->fence);
-		rcu_read_unlock();
+		fence = drm_syncobj_fence_get(syncobj);
 		if (!fence)
 			return -EINVAL;
 
diff --git a/drivers/gpu/drm/i915/intel_gvt.c b/drivers/gpu/drm/i915/intel_gvt.c
index 52d5b82790d9..c17ed0e62b67 100644
--- a/drivers/gpu/drm/i915/intel_gvt.c
+++ b/drivers/gpu/drm/i915/intel_gvt.c
@@ -45,7 +45,7 @@ static bool is_supported_device(struct drm_i915_private *dev_priv)
 		return true;
 	if (IS_SKYLAKE(dev_priv))
 		return true;
-	if (IS_KABYLAKE(dev_priv) && INTEL_DEVID(dev_priv) == 0x591D)
+	if (IS_KABYLAKE(dev_priv))
 		return true;
 	return false;
 }
diff --git a/drivers/gpu/drm/imx/ipuv3-plane.c b/drivers/gpu/drm/imx/ipuv3-plane.c
index debde2dae7bf..227309b01206 100644
--- a/drivers/gpu/drm/imx/ipuv3-plane.c
+++ b/drivers/gpu/drm/imx/ipuv3-plane.c
@@ -496,6 +496,27 @@ static int ipu_chan_assign_axi_id(int ipu_chan)
 	}
 }
 
+static void ipu_calculate_bursts(u32 width, u32 cpp, u32 stride,
+				 u8 *burstsize, u8 *num_bursts)
+{
+	const unsigned int width_bytes = width * cpp;
+	unsigned int npb, bursts;
+
+	/* Maximum number of pixels per burst without overshooting stride */
+	for (npb = 64 / cpp; npb > 0; --npb) {
+		if (round_up(width_bytes, npb * cpp) <= stride)
+			break;
+	}
+	*burstsize = npb;
+
+	/* Maximum number of consecutive bursts without overshooting stride */
+	for (bursts = 8; bursts > 1; bursts /= 2) {
+		if (round_up(width_bytes, npb * cpp * bursts) <= stride)
+			break;
+	}
+	*num_bursts = bursts;
+}
+
 static void ipu_plane_atomic_update(struct drm_plane *plane,
 				    struct drm_plane_state *old_state)
 {
@@ -509,6 +530,9 @@ static void ipu_plane_atomic_update(struct drm_plane *plane,
 	unsigned long alpha_eba = 0;
 	enum ipu_color_space ics;
 	unsigned int axi_id = 0;
+	const struct drm_format_info *info;
+	u8 burstsize, num_bursts;
+	u32 width, height;
 	int active;
 
 	if (ipu_plane->dp_flow == IPU_DP_FLOW_SYNC_FG)
@@ -525,8 +549,8 @@ static void ipu_plane_atomic_update(struct drm_plane *plane,
 		ipu_prg_channel_configure(ipu_plane->ipu_ch, axi_id,
 					  drm_rect_width(&state->src) >> 16,
 					  drm_rect_height(&state->src) >> 16,
-					  state->fb->pitches[0],
-					  state->fb->format->format, &eba);
+					  fb->pitches[0],
+					  fb->format->format, &eba);
 	}
 
 	if (old_state->fb && !drm_atomic_crtc_needs_modeset(crtc_state)) {
@@ -553,11 +577,11 @@ static void ipu_plane_atomic_update(struct drm_plane *plane,
 		ipu_dp_set_global_alpha(ipu_plane->dp, true, 0, true);
 		break;
 	case IPU_DP_FLOW_SYNC_FG:
-		ics = ipu_drm_fourcc_to_colorspace(state->fb->format->format);
+		ics = ipu_drm_fourcc_to_colorspace(fb->format->format);
 		ipu_dp_setup_channel(ipu_plane->dp, ics,
 					IPUV3_COLORSPACE_UNKNOWN);
 		/* Enable local alpha on partial plane */
-		switch (state->fb->format->format) {
+		switch (fb->format->format) {
 		case DRM_FORMAT_ARGB1555:
 		case DRM_FORMAT_ABGR1555:
 		case DRM_FORMAT_RGBA5551:
@@ -583,15 +607,21 @@ static void ipu_plane_atomic_update(struct drm_plane *plane,
 
 	ipu_dmfc_config_wait4eot(ipu_plane->dmfc, drm_rect_width(dst));
 
+	width = drm_rect_width(&state->src) >> 16;
+	height = drm_rect_height(&state->src) >> 16;
+	info = drm_format_info(fb->format->format);
+	ipu_calculate_bursts(width, info->cpp[0], fb->pitches[0],
+			     &burstsize, &num_bursts);
+
 	ipu_cpmem_zero(ipu_plane->ipu_ch);
-	ipu_cpmem_set_resolution(ipu_plane->ipu_ch,
-				 drm_rect_width(&state->src) >> 16,
-				 drm_rect_height(&state->src) >> 16);
-	ipu_cpmem_set_fmt(ipu_plane->ipu_ch, state->fb->format->format);
+	ipu_cpmem_set_resolution(ipu_plane->ipu_ch, width, height);
+	ipu_cpmem_set_fmt(ipu_plane->ipu_ch, fb->format->format);
+	ipu_cpmem_set_burstsize(ipu_plane->ipu_ch, burstsize);
 	ipu_cpmem_set_high_priority(ipu_plane->ipu_ch);
 	ipu_idmac_set_double_buffer(ipu_plane->ipu_ch, 1);
-	ipu_cpmem_set_stride(ipu_plane->ipu_ch, state->fb->pitches[0]);
+	ipu_cpmem_set_stride(ipu_plane->ipu_ch, fb->pitches[0]);
 	ipu_cpmem_set_axi_id(ipu_plane->ipu_ch, axi_id);
+
 	switch (fb->format->format) {
 	case DRM_FORMAT_YUV420:
 	case DRM_FORMAT_YVU420:
@@ -631,6 +661,7 @@ static void ipu_plane_atomic_update(struct drm_plane *plane,
 	case DRM_FORMAT_RGBX8888_A8:
 	case DRM_FORMAT_BGRX8888_A8:
 		alpha_eba = drm_plane_state_to_eba(state, 1);
+		num_bursts = 0;
 
 		dev_dbg(ipu_plane->base.dev->dev, "phys = %lu %lu, x = %d, y = %d",
 			eba, alpha_eba, state->src.x1 >> 16, state->src.y1 >> 16);
@@ -644,8 +675,7 @@ static void ipu_plane_atomic_update(struct drm_plane *plane,
 		ipu_cpmem_set_format_passthrough(ipu_plane->alpha_ch, 8);
 		ipu_cpmem_set_high_priority(ipu_plane->alpha_ch);
 		ipu_idmac_set_double_buffer(ipu_plane->alpha_ch, 1);
-		ipu_cpmem_set_stride(ipu_plane->alpha_ch,
-				     state->fb->pitches[1]);
+		ipu_cpmem_set_stride(ipu_plane->alpha_ch, fb->pitches[1]);
 		ipu_cpmem_set_burstsize(ipu_plane->alpha_ch, 16);
 		ipu_cpmem_set_buffer(ipu_plane->alpha_ch, 0, alpha_eba);
 		ipu_cpmem_set_buffer(ipu_plane->alpha_ch, 1, alpha_eba);
@@ -657,6 +687,7 @@ static void ipu_plane_atomic_update(struct drm_plane *plane,
 	}
 	ipu_cpmem_set_buffer(ipu_plane->ipu_ch, 0, eba);
 	ipu_cpmem_set_buffer(ipu_plane->ipu_ch, 1, eba);
+	ipu_idmac_lock_enable(ipu_plane->ipu_ch, num_bursts);
 	ipu_plane_enable(ipu_plane);
 }
 
diff --git a/drivers/gpu/drm/mediatek/mtk_drm_fb.c b/drivers/gpu/drm/mediatek/mtk_drm_fb.c
index d4246c9dceae..0d8d506695f9 100644
--- a/drivers/gpu/drm/mediatek/mtk_drm_fb.c
+++ b/drivers/gpu/drm/mediatek/mtk_drm_fb.c
@@ -58,7 +58,7 @@ static void mtk_drm_fb_destroy(struct drm_framebuffer *fb)
 
 	drm_framebuffer_cleanup(fb);
 
-	drm_gem_object_unreference_unlocked(mtk_fb->gem_obj);
+	drm_gem_object_put_unlocked(mtk_fb->gem_obj);
 
 	kfree(mtk_fb);
 }
@@ -160,6 +160,6 @@ struct drm_framebuffer *mtk_drm_mode_fb_create(struct drm_device *dev,
 	return &mtk_fb->base;
 
 unreference:
-	drm_gem_object_unreference_unlocked(gem);
+	drm_gem_object_put_unlocked(gem);
 	return ERR_PTR(ret);
 }
diff --git a/drivers/gpu/drm/mediatek/mtk_drm_gem.c b/drivers/gpu/drm/mediatek/mtk_drm_gem.c
index 8ec963fff8b1..f595ac816b55 100644
--- a/drivers/gpu/drm/mediatek/mtk_drm_gem.c
+++ b/drivers/gpu/drm/mediatek/mtk_drm_gem.c
@@ -122,7 +122,7 @@ int mtk_drm_gem_dumb_create(struct drm_file *file_priv, struct drm_device *dev,
 		goto err_handle_create;
 
 	/* drop reference from allocate - handle holds it now. */
-	drm_gem_object_unreference_unlocked(&mtk_gem->base);
+	drm_gem_object_put_unlocked(&mtk_gem->base);
 
 	return 0;
 
diff --git a/drivers/gpu/drm/meson/meson_drv.c b/drivers/gpu/drm/meson/meson_drv.c
index 5375e6dccdd7..7742c7d81ed8 100644
--- a/drivers/gpu/drm/meson/meson_drv.c
+++ b/drivers/gpu/drm/meson/meson_drv.c
@@ -116,8 +116,6 @@ static struct drm_driver meson_driver = {
 
 	/* GEM Ops */
 	.dumb_create		= drm_gem_cma_dumb_create,
-	.dumb_destroy		= drm_gem_dumb_destroy,
-	.dumb_map_offset	= drm_gem_cma_dumb_map_offset,
 	.gem_free_object_unlocked = drm_gem_cma_free_object,
 	.gem_vm_ops		= &drm_gem_cma_vm_ops,
 
diff --git a/drivers/gpu/drm/mgag200/mgag200_cursor.c b/drivers/gpu/drm/mgag200/mgag200_cursor.c
index 2ac3fcbfea7b..968e20379d54 100644
--- a/drivers/gpu/drm/mgag200/mgag200_cursor.c
+++ b/drivers/gpu/drm/mgag200/mgag200_cursor.c
@@ -248,7 +248,7 @@ int mga_crtc_cursor_set(struct drm_crtc *crtc,
 out_unreserve1:
 	mgag200_bo_unreserve(pixels_2);
 out_unref:
-	drm_gem_object_unreference_unlocked(obj);
+	drm_gem_object_put_unlocked(obj);
 
 	return ret;
 }
diff --git a/drivers/gpu/drm/mgag200/mgag200_drv.c b/drivers/gpu/drm/mgag200/mgag200_drv.c
index 4189160af726..74cdde2ee474 100644
--- a/drivers/gpu/drm/mgag200/mgag200_drv.c
+++ b/drivers/gpu/drm/mgag200/mgag200_drv.c
@@ -102,7 +102,6 @@ static struct drm_driver driver = {
 	.gem_free_object_unlocked = mgag200_gem_free_object,
 	.dumb_create = mgag200_dumb_create,
 	.dumb_map_offset = mgag200_dumb_mmap_offset,
-	.dumb_destroy = drm_gem_dumb_destroy,
 };
 
 static struct pci_driver mgag200_pci_driver = {
diff --git a/drivers/gpu/drm/mgag200/mgag200_fb.c b/drivers/gpu/drm/mgag200/mgag200_fb.c
index 9d914ca69996..30726c9fe28c 100644
--- a/drivers/gpu/drm/mgag200/mgag200_fb.c
+++ b/drivers/gpu/drm/mgag200/mgag200_fb.c
@@ -232,7 +232,7 @@ static int mgag200fb_create(struct drm_fb_helper *helper,
 err_alloc_fbi:
 	vfree(sysram);
 err_sysram:
-	drm_gem_object_unreference_unlocked(gobj);
+	drm_gem_object_put_unlocked(gobj);
 
 	return ret;
 }
@@ -245,7 +245,7 @@ static int mga_fbdev_destroy(struct drm_device *dev,
 	drm_fb_helper_unregister_fbi(&mfbdev->helper);
 
 	if (mfb->obj) {
-		drm_gem_object_unreference_unlocked(mfb->obj);
+		drm_gem_object_put_unlocked(mfb->obj);
 		mfb->obj = NULL;
 	}
 	drm_fb_helper_fini(&mfbdev->helper);
diff --git a/drivers/gpu/drm/mgag200/mgag200_main.c b/drivers/gpu/drm/mgag200/mgag200_main.c
index dce8a3eb5a10..780f983b0294 100644
--- a/drivers/gpu/drm/mgag200/mgag200_main.c
+++ b/drivers/gpu/drm/mgag200/mgag200_main.c
@@ -18,7 +18,7 @@ static void mga_user_framebuffer_destroy(struct drm_framebuffer *fb)
 {
 	struct mga_framebuffer *mga_fb = to_mga_framebuffer(fb);
 
-	drm_gem_object_unreference_unlocked(mga_fb->obj);
+	drm_gem_object_put_unlocked(mga_fb->obj);
 	drm_framebuffer_cleanup(fb);
 	kfree(fb);
 }
@@ -59,13 +59,13 @@ mgag200_user_framebuffer_create(struct drm_device *dev,
 
 	mga_fb = kzalloc(sizeof(*mga_fb), GFP_KERNEL);
 	if (!mga_fb) {
-		drm_gem_object_unreference_unlocked(obj);
+		drm_gem_object_put_unlocked(obj);
 		return ERR_PTR(-ENOMEM);
 	}
 
 	ret = mgag200_framebuffer_init(dev, mga_fb, mode_cmd, obj);
 	if (ret) {
-		drm_gem_object_unreference_unlocked(obj);
+		drm_gem_object_put_unlocked(obj);
 		kfree(mga_fb);
 		return ERR_PTR(ret);
 	}
@@ -317,7 +317,7 @@ int mgag200_dumb_create(struct drm_file *file,
 		return ret;
 
 	ret = drm_gem_handle_create(file, gobj, &handle);
-	drm_gem_object_unreference_unlocked(gobj);
+	drm_gem_object_put_unlocked(gobj);
 	if (ret)
 		return ret;
 
@@ -366,6 +366,6 @@ mgag200_dumb_mmap_offset(struct drm_file *file,
 	bo = gem_to_mga_bo(obj);
 	*offset = mgag200_bo_mmap_offset(bo);
 
-	drm_gem_object_unreference_unlocked(obj);
+	drm_gem_object_put_unlocked(obj);
 	return 0;
 }
diff --git a/drivers/gpu/drm/msm/Kconfig b/drivers/gpu/drm/msm/Kconfig
index b638d192ce5e..99d39b2aefa6 100644
--- a/drivers/gpu/drm/msm/Kconfig
+++ b/drivers/gpu/drm/msm/Kconfig
@@ -5,7 +5,7 @@ config DRM_MSM
 	depends on ARCH_QCOM || (ARM && COMPILE_TEST)
 	depends on OF && COMMON_CLK
 	depends on MMU
-	select QCOM_MDT_LOADER
+	select QCOM_MDT_LOADER if ARCH_QCOM
 	select REGULATOR
 	select DRM_KMS_HELPER
 	select DRM_PANEL
diff --git a/drivers/gpu/drm/msm/adreno/a3xx_gpu.c b/drivers/gpu/drm/msm/adreno/a3xx_gpu.c
index 0e3828ed1e46..7791313405b5 100644
--- a/drivers/gpu/drm/msm/adreno/a3xx_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/a3xx_gpu.c
@@ -486,8 +486,6 @@ struct msm_gpu *a3xx_gpu_init(struct drm_device *dev)
 	adreno_gpu = &a3xx_gpu->base;
 	gpu = &adreno_gpu->base;
 
-	a3xx_gpu->pdev = pdev;
-
 	gpu->perfcntrs = perfcntrs;
 	gpu->num_perfcntrs = ARRAY_SIZE(perfcntrs);
 
diff --git a/drivers/gpu/drm/msm/adreno/a3xx_gpu.h b/drivers/gpu/drm/msm/adreno/a3xx_gpu.h
index 85ff66cbddd6..ab60dc9e344e 100644
--- a/drivers/gpu/drm/msm/adreno/a3xx_gpu.h
+++ b/drivers/gpu/drm/msm/adreno/a3xx_gpu.h
@@ -28,7 +28,6 @@
 
 struct a3xx_gpu {
 	struct adreno_gpu base;
-	struct platform_device *pdev;
 
 	/* if OCMEM is used for GMEM: */
 	uint32_t ocmem_base;
diff --git a/drivers/gpu/drm/msm/adreno/a4xx_gpu.c b/drivers/gpu/drm/msm/adreno/a4xx_gpu.c
index 19abf229b08d..58341ef6f15b 100644
--- a/drivers/gpu/drm/msm/adreno/a4xx_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/a4xx_gpu.c
@@ -568,8 +568,6 @@ struct msm_gpu *a4xx_gpu_init(struct drm_device *dev)
 	adreno_gpu = &a4xx_gpu->base;
 	gpu = &adreno_gpu->base;
 
-	a4xx_gpu->pdev = pdev;
-
 	gpu->perfcntrs = NULL;
 	gpu->num_perfcntrs = 0;
 
diff --git a/drivers/gpu/drm/msm/adreno/a4xx_gpu.h b/drivers/gpu/drm/msm/adreno/a4xx_gpu.h
index 01247204ac92..f757184328a3 100644
--- a/drivers/gpu/drm/msm/adreno/a4xx_gpu.h
+++ b/drivers/gpu/drm/msm/adreno/a4xx_gpu.h
@@ -23,7 +23,6 @@
 
 struct a4xx_gpu {
 	struct adreno_gpu base;
-	struct platform_device *pdev;
 
 	/* if OCMEM is used for GMEM: */
 	uint32_t ocmem_base;
diff --git a/drivers/gpu/drm/msm/adreno/a5xx_gpu.c b/drivers/gpu/drm/msm/adreno/a5xx_gpu.c
index b4b54f1c24bc..17c59d839e6f 100644
--- a/drivers/gpu/drm/msm/adreno/a5xx_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/a5xx_gpu.c
@@ -15,7 +15,7 @@
 #include <linux/cpumask.h>
 #include <linux/qcom_scm.h>
 #include <linux/dma-mapping.h>
-#include <linux/of_reserved_mem.h>
+#include <linux/of_address.h>
 #include <linux/soc/qcom/mdt_loader.h>
 #include "msm_gem.h"
 #include "msm_mmu.h"
@@ -26,16 +26,34 @@ static void a5xx_dump(struct msm_gpu *gpu);
 
 #define GPU_PAS_ID 13
 
-#if IS_ENABLED(CONFIG_QCOM_MDT_LOADER)
-
 static int zap_shader_load_mdt(struct device *dev, const char *fwname)
 {
 	const struct firmware *fw;
+	struct device_node *np;
+	struct resource r;
 	phys_addr_t mem_phys;
 	ssize_t mem_size;
 	void *mem_region = NULL;
 	int ret;
 
+	if (!IS_ENABLED(CONFIG_ARCH_QCOM))
+		return -EINVAL;
+
+	np = of_get_child_by_name(dev->of_node, "zap-shader");
+	if (!np)
+		return -ENODEV;
+
+	np = of_parse_phandle(np, "memory-region", 0);
+	if (!np)
+		return -EINVAL;
+
+	ret = of_address_to_resource(np, 0, &r);
+	if (ret)
+		return ret;
+
+	mem_phys = r.start;
+	mem_size = resource_size(&r);
+
 	/* Request the MDT file for the firmware */
 	ret = request_firmware(&fw, fwname, dev);
 	if (ret) {
@@ -51,7 +69,7 @@ static int zap_shader_load_mdt(struct device *dev, const char *fwname)
 	}
 
 	/* Allocate memory for the firmware image */
-	mem_region = dmam_alloc_coherent(dev, mem_size, &mem_phys, GFP_KERNEL);
+	mem_region = memremap(mem_phys, mem_size,  MEMREMAP_WC);
 	if (!mem_region) {
 		ret = -ENOMEM;
 		goto out;
@@ -69,16 +87,13 @@ static int zap_shader_load_mdt(struct device *dev, const char *fwname)
 		DRM_DEV_ERROR(dev, "Unable to authorize the image\n");
 
 out:
+	if (mem_region)
+		memunmap(mem_region);
+
 	release_firmware(fw);
 
 	return ret;
 }
-#else
-static int zap_shader_load_mdt(struct device *dev, const char *fwname)
-{
-	return -ENODEV;
-}
-#endif
 
 static void a5xx_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit,
 	struct msm_file_private *ctx)
@@ -117,12 +132,10 @@ static void a5xx_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit,
 	gpu->funcs->flush(gpu);
 }
 
-struct a5xx_hwcg {
+static const struct {
 	u32 offset;
 	u32 value;
-};
-
-static const struct a5xx_hwcg a530_hwcg[] = {
+} a5xx_hwcg[] = {
 	{REG_A5XX_RBBM_CLOCK_CNTL_SP0, 0x02222222},
 	{REG_A5XX_RBBM_CLOCK_CNTL_SP1, 0x02222222},
 	{REG_A5XX_RBBM_CLOCK_CNTL_SP2, 0x02222222},
@@ -217,38 +230,16 @@ static const struct a5xx_hwcg a530_hwcg[] = {
 	{REG_A5XX_RBBM_CLOCK_DELAY_VFD, 0x00002222}
 };
 
-static const struct {
-	int (*test)(struct adreno_gpu *gpu);
-	const struct a5xx_hwcg *regs;
-	unsigned int count;
-} a5xx_hwcg_regs[] = {
-	{ adreno_is_a530, a530_hwcg, ARRAY_SIZE(a530_hwcg), },
-};
-
-static void _a5xx_enable_hwcg(struct msm_gpu *gpu,
-		const struct a5xx_hwcg *regs, unsigned int count)
+void a5xx_set_hwcg(struct msm_gpu *gpu, bool state)
 {
 	unsigned int i;
 
-	for (i = 0; i < count; i++)
-		gpu_write(gpu, regs[i].offset, regs[i].value);
+	for (i = 0; i < ARRAY_SIZE(a5xx_hwcg); i++)
+		gpu_write(gpu, a5xx_hwcg[i].offset,
+			state ? a5xx_hwcg[i].value : 0);
 
-	gpu_write(gpu, REG_A5XX_RBBM_CLOCK_CNTL, 0xAAA8AA00);
-	gpu_write(gpu, REG_A5XX_RBBM_ISDB_CNT, 0x182);
-}
-
-static void a5xx_enable_hwcg(struct msm_gpu *gpu)
-{
-	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
-	unsigned int i;
-
-	for (i = 0; i < ARRAY_SIZE(a5xx_hwcg_regs); i++) {
-		if (a5xx_hwcg_regs[i].test(adreno_gpu)) {
-			_a5xx_enable_hwcg(gpu, a5xx_hwcg_regs[i].regs,
-				a5xx_hwcg_regs[i].count);
-			return;
-		}
-	}
+	gpu_write(gpu, REG_A5XX_RBBM_CLOCK_CNTL, state ? 0xAAA8AA00 : 0);
+	gpu_write(gpu, REG_A5XX_RBBM_ISDB_CNT, state ? 0x182 : 0x180);
 }
 
 static int a5xx_me_init(struct msm_gpu *gpu)
@@ -293,28 +284,14 @@ static int a5xx_me_init(struct msm_gpu *gpu)
 static struct drm_gem_object *a5xx_ucode_load_bo(struct msm_gpu *gpu,
 		const struct firmware *fw, u64 *iova)
 {
-	struct drm_device *drm = gpu->dev;
 	struct drm_gem_object *bo;
 	void *ptr;
 
-	bo = msm_gem_new_locked(drm, fw->size - 4, MSM_BO_UNCACHED);
-	if (IS_ERR(bo))
-		return bo;
-
-	ptr = msm_gem_get_vaddr(bo);
-	if (!ptr) {
-		drm_gem_object_unreference(bo);
-		return ERR_PTR(-ENOMEM);
-	}
-
-	if (iova) {
-		int ret = msm_gem_get_iova(bo, gpu->aspace, iova);
+	ptr = msm_gem_kernel_new_locked(gpu->dev, fw->size - 4,
+		MSM_BO_UNCACHED | MSM_BO_GPU_READONLY, gpu->aspace, &bo, iova);
 
-		if (ret) {
-			drm_gem_object_unreference(bo);
-			return ERR_PTR(ret);
-		}
-	}
+	if (IS_ERR(ptr))
+		return ERR_CAST(ptr);
 
 	memcpy(ptr, &fw->data[4], fw->size - 4);
 
@@ -377,51 +354,11 @@ static int a5xx_zap_shader_resume(struct msm_gpu *gpu)
 	return ret;
 }
 
-/* Set up a child device to "own" the zap shader */
-static int a5xx_zap_shader_dev_init(struct device *parent, struct device *dev)
-{
-	struct device_node *node;
-	int ret;
-
-	if (dev->parent)
-		return 0;
-
-	/* Find the sub-node for the zap shader */
-	node = of_get_child_by_name(parent->of_node, "zap-shader");
-	if (!node) {
-		DRM_DEV_ERROR(parent, "zap-shader not found in device tree\n");
-		return -ENODEV;
-	}
-
-	dev->parent = parent;
-	dev->of_node = node;
-	dev_set_name(dev, "adreno_zap_shader");
-
-	ret = device_register(dev);
-	if (ret) {
-		DRM_DEV_ERROR(parent, "Couldn't register zap shader device\n");
-		goto out;
-	}
-
-	ret = of_reserved_mem_device_init(dev);
-	if (ret) {
-		DRM_DEV_ERROR(parent, "Unable to set up the reserved memory\n");
-		device_unregister(dev);
-	}
-
-out:
-	if (ret)
-		dev->parent = NULL;
-
-	return ret;
-}
-
 static int a5xx_zap_shader_init(struct msm_gpu *gpu)
 {
 	static bool loaded;
 	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
-	struct a5xx_gpu *a5xx_gpu = to_a5xx_gpu(adreno_gpu);
-	struct platform_device *pdev = a5xx_gpu->pdev;
+	struct platform_device *pdev = gpu->pdev;
 	int ret;
 
 	/*
@@ -444,11 +381,7 @@ static int a5xx_zap_shader_init(struct msm_gpu *gpu)
 		return -ENODEV;
 	}
 
-	ret = a5xx_zap_shader_dev_init(&pdev->dev, &a5xx_gpu->zap_dev);
-
-	if (!ret)
-		ret = zap_shader_load_mdt(&a5xx_gpu->zap_dev,
-			adreno_gpu->info->zapfw);
+	ret = zap_shader_load_mdt(&pdev->dev, adreno_gpu->info->zapfw);
 
 	loaded = !ret;
 
@@ -462,6 +395,7 @@ static int a5xx_zap_shader_init(struct msm_gpu *gpu)
 	  A5XX_RBBM_INT_0_MASK_RBBM_ETS_MS_TIMEOUT | \
 	  A5XX_RBBM_INT_0_MASK_RBBM_ATB_ASYNC_OVERFLOW | \
 	  A5XX_RBBM_INT_0_MASK_CP_HW_ERROR | \
+	  A5XX_RBBM_INT_0_MASK_MISC_HANG_DETECT | \
 	  A5XX_RBBM_INT_0_MASK_CP_CACHE_FLUSH_TS | \
 	  A5XX_RBBM_INT_0_MASK_UCHE_OOB_ACCESS | \
 	  A5XX_RBBM_INT_0_MASK_GPMU_VOLTAGE_DROOP)
@@ -545,7 +479,7 @@ static int a5xx_hw_init(struct msm_gpu *gpu)
 	gpu_write(gpu, REG_A5XX_RBBM_AHB_CNTL1, 0xA6FFFFFF);
 
 	/* Enable HWCG */
-	a5xx_enable_hwcg(gpu);
+	a5xx_set_hwcg(gpu, true);
 
 	gpu_write(gpu, REG_A5XX_RBBM_AHB_CNTL2, 0x0000003F);
 
@@ -691,9 +625,6 @@ static void a5xx_destroy(struct msm_gpu *gpu)
 
 	DBG("%s", gpu->name);
 
-	if (a5xx_gpu->zap_dev.parent)
-		device_unregister(&a5xx_gpu->zap_dev);
-
 	if (a5xx_gpu->pm4_bo) {
 		if (a5xx_gpu->pm4_iova)
 			msm_gem_put_iova(a5xx_gpu->pm4_bo, gpu->aspace);
@@ -867,6 +798,27 @@ static void a5xx_gpmu_err_irq(struct msm_gpu *gpu)
 	dev_err_ratelimited(gpu->dev->dev, "GPMU | voltage droop\n");
 }
 
+static void a5xx_fault_detect_irq(struct msm_gpu *gpu)
+{
+	struct drm_device *dev = gpu->dev;
+	struct msm_drm_private *priv = dev->dev_private;
+
+	dev_err(dev->dev, "gpu fault fence %x status %8.8X rb %4.4x/%4.4x ib1 %16.16llX/%4.4x ib2 %16.16llX/%4.4x\n",
+		gpu->funcs->last_fence(gpu),
+		gpu_read(gpu, REG_A5XX_RBBM_STATUS),
+		gpu_read(gpu, REG_A5XX_CP_RB_RPTR),
+		gpu_read(gpu, REG_A5XX_CP_RB_WPTR),
+		gpu_read64(gpu, REG_A5XX_CP_IB1_BASE, REG_A5XX_CP_IB1_BASE_HI),
+		gpu_read(gpu, REG_A5XX_CP_IB1_BUFSZ),
+		gpu_read64(gpu, REG_A5XX_CP_IB2_BASE, REG_A5XX_CP_IB2_BASE_HI),
+		gpu_read(gpu, REG_A5XX_CP_IB2_BUFSZ));
+
+	/* Turn off the hangcheck timer to keep it from bothering us */
+	del_timer(&gpu->hangcheck_timer);
+
+	queue_work(priv->wq, &gpu->recover_work);
+}
+
 #define RBBM_ERROR_MASK \
 	(A5XX_RBBM_INT_0_MASK_RBBM_AHB_ERROR | \
 	A5XX_RBBM_INT_0_MASK_RBBM_TRANSFER_TIMEOUT | \
@@ -893,6 +845,9 @@ static irqreturn_t a5xx_irq(struct msm_gpu *gpu)
 	if (status & A5XX_RBBM_INT_0_MASK_CP_HW_ERROR)
 		a5xx_cp_err_irq(gpu);
 
+	if (status & A5XX_RBBM_INT_0_MASK_MISC_HANG_DETECT)
+		a5xx_fault_detect_irq(gpu);
+
 	if (status & A5XX_RBBM_INT_0_MASK_UCHE_OOB_ACCESS)
 		a5xx_uche_err_irq(gpu);
 
@@ -920,31 +875,30 @@ static const u32 a5xx_registers[] = {
 	0x0000, 0x0002, 0x0004, 0x0020, 0x0022, 0x0026, 0x0029, 0x002B,
 	0x002E, 0x0035, 0x0038, 0x0042, 0x0044, 0x0044, 0x0047, 0x0095,
 	0x0097, 0x00BB, 0x03A0, 0x0464, 0x0469, 0x046F, 0x04D2, 0x04D3,
-	0x04E0, 0x0533, 0x0540, 0x0555, 0xF400, 0xF400, 0xF800, 0xF807,
-	0x0800, 0x081A, 0x081F, 0x0841, 0x0860, 0x0860, 0x0880, 0x08A0,
-	0x0B00, 0x0B12, 0x0B15, 0x0B28, 0x0B78, 0x0B7F, 0x0BB0, 0x0BBD,
-	0x0BC0, 0x0BC6, 0x0BD0, 0x0C53, 0x0C60, 0x0C61, 0x0C80, 0x0C82,
-	0x0C84, 0x0C85, 0x0C90, 0x0C98, 0x0CA0, 0x0CA0, 0x0CB0, 0x0CB2,
-	0x2180, 0x2185, 0x2580, 0x2585, 0x0CC1, 0x0CC1, 0x0CC4, 0x0CC7,
-	0x0CCC, 0x0CCC, 0x0CD0, 0x0CD8, 0x0CE0, 0x0CE5, 0x0CE8, 0x0CE8,
-	0x0CEC, 0x0CF1, 0x0CFB, 0x0D0E, 0x2100, 0x211E, 0x2140, 0x2145,
-	0x2500, 0x251E, 0x2540, 0x2545, 0x0D10, 0x0D17, 0x0D20, 0x0D23,
-	0x0D30, 0x0D30, 0x20C0, 0x20C0, 0x24C0, 0x24C0, 0x0E40, 0x0E43,
-	0x0E4A, 0x0E4A, 0x0E50, 0x0E57, 0x0E60, 0x0E7C, 0x0E80, 0x0E8E,
-	0x0E90, 0x0E96, 0x0EA0, 0x0EA8, 0x0EB0, 0x0EB2, 0xE140, 0xE147,
-	0xE150, 0xE187, 0xE1A0, 0xE1A9, 0xE1B0, 0xE1B6, 0xE1C0, 0xE1C7,
-	0xE1D0, 0xE1D1, 0xE200, 0xE201, 0xE210, 0xE21C, 0xE240, 0xE268,
-	0xE000, 0xE006, 0xE010, 0xE09A, 0xE0A0, 0xE0A4, 0xE0AA, 0xE0EB,
-	0xE100, 0xE105, 0xE380, 0xE38F, 0xE3B0, 0xE3B0, 0xE400, 0xE405,
-	0xE408, 0xE4E9, 0xE4F0, 0xE4F0, 0xE280, 0xE280, 0xE282, 0xE2A3,
-	0xE2A5, 0xE2C2, 0xE940, 0xE947, 0xE950, 0xE987, 0xE9A0, 0xE9A9,
-	0xE9B0, 0xE9B6, 0xE9C0, 0xE9C7, 0xE9D0, 0xE9D1, 0xEA00, 0xEA01,
-	0xEA10, 0xEA1C, 0xEA40, 0xEA68, 0xE800, 0xE806, 0xE810, 0xE89A,
-	0xE8A0, 0xE8A4, 0xE8AA, 0xE8EB, 0xE900, 0xE905, 0xEB80, 0xEB8F,
-	0xEBB0, 0xEBB0, 0xEC00, 0xEC05, 0xEC08, 0xECE9, 0xECF0, 0xECF0,
-	0xEA80, 0xEA80, 0xEA82, 0xEAA3, 0xEAA5, 0xEAC2, 0xA800, 0xA8FF,
-	0xAC60, 0xAC60, 0xB000, 0xB97F, 0xB9A0, 0xB9BF,
-	~0
+	0x04E0, 0x0533, 0x0540, 0x0555, 0x0800, 0x081A, 0x081F, 0x0841,
+	0x0860, 0x0860, 0x0880, 0x08A0, 0x0B00, 0x0B12, 0x0B15, 0x0B28,
+	0x0B78, 0x0B7F, 0x0BB0, 0x0BBD, 0x0BC0, 0x0BC6, 0x0BD0, 0x0C53,
+	0x0C60, 0x0C61, 0x0C80, 0x0C82, 0x0C84, 0x0C85, 0x0C90, 0x0C98,
+	0x0CA0, 0x0CA0, 0x0CB0, 0x0CB2, 0x2180, 0x2185, 0x2580, 0x2585,
+	0x0CC1, 0x0CC1, 0x0CC4, 0x0CC7, 0x0CCC, 0x0CCC, 0x0CD0, 0x0CD8,
+	0x0CE0, 0x0CE5, 0x0CE8, 0x0CE8, 0x0CEC, 0x0CF1, 0x0CFB, 0x0D0E,
+	0x2100, 0x211E, 0x2140, 0x2145, 0x2500, 0x251E, 0x2540, 0x2545,
+	0x0D10, 0x0D17, 0x0D20, 0x0D23, 0x0D30, 0x0D30, 0x20C0, 0x20C0,
+	0x24C0, 0x24C0, 0x0E40, 0x0E43, 0x0E4A, 0x0E4A, 0x0E50, 0x0E57,
+	0x0E60, 0x0E7C, 0x0E80, 0x0E8E, 0x0E90, 0x0E96, 0x0EA0, 0x0EA8,
+	0x0EB0, 0x0EB2, 0xE140, 0xE147, 0xE150, 0xE187, 0xE1A0, 0xE1A9,
+	0xE1B0, 0xE1B6, 0xE1C0, 0xE1C7, 0xE1D0, 0xE1D1, 0xE200, 0xE201,
+	0xE210, 0xE21C, 0xE240, 0xE268, 0xE000, 0xE006, 0xE010, 0xE09A,
+	0xE0A0, 0xE0A4, 0xE0AA, 0xE0EB, 0xE100, 0xE105, 0xE380, 0xE38F,
+	0xE3B0, 0xE3B0, 0xE400, 0xE405, 0xE408, 0xE4E9, 0xE4F0, 0xE4F0,
+	0xE280, 0xE280, 0xE282, 0xE2A3, 0xE2A5, 0xE2C2, 0xE940, 0xE947,
+	0xE950, 0xE987, 0xE9A0, 0xE9A9, 0xE9B0, 0xE9B6, 0xE9C0, 0xE9C7,
+	0xE9D0, 0xE9D1, 0xEA00, 0xEA01, 0xEA10, 0xEA1C, 0xEA40, 0xEA68,
+	0xE800, 0xE806, 0xE810, 0xE89A, 0xE8A0, 0xE8A4, 0xE8AA, 0xE8EB,
+	0xE900, 0xE905, 0xEB80, 0xEB8F, 0xEBB0, 0xEBB0, 0xEC00, 0xEC05,
+	0xEC08, 0xECE9, 0xECF0, 0xECF0, 0xEA80, 0xEA80, 0xEA82, 0xEAA3,
+	0xEAA5, 0xEAC2, 0xA800, 0xA8FF, 0xAC60, 0xAC60, 0xB000, 0xB97F,
+	0xB9A0, 0xB9BF, ~0
 };
 
 static void a5xx_dump(struct msm_gpu *gpu)
@@ -1020,7 +974,14 @@ static void a5xx_show(struct msm_gpu *gpu, struct seq_file *m)
 {
 	seq_printf(m, "status:   %08x\n",
 			gpu_read(gpu, REG_A5XX_RBBM_STATUS));
+
+	/*
+	 * Temporarily disable hardware clock gating before going into
+	 * adreno_show to avoid issues while reading the registers
+	 */
+	a5xx_set_hwcg(gpu, false);
 	adreno_show(gpu, m);
+	a5xx_set_hwcg(gpu, true);
 }
 #endif
 
@@ -1064,7 +1025,6 @@ struct msm_gpu *a5xx_gpu_init(struct drm_device *dev)
 	adreno_gpu = &a5xx_gpu->base;
 	gpu = &adreno_gpu->base;
 
-	a5xx_gpu->pdev = pdev;
 	adreno_gpu->registers = a5xx_registers;
 	adreno_gpu->reg_offsets = a5xx_register_offsets;
 
diff --git a/drivers/gpu/drm/msm/adreno/a5xx_gpu.h b/drivers/gpu/drm/msm/adreno/a5xx_gpu.h
index 6638bc85645d..e94451685bf8 100644
--- a/drivers/gpu/drm/msm/adreno/a5xx_gpu.h
+++ b/drivers/gpu/drm/msm/adreno/a5xx_gpu.h
@@ -23,7 +23,6 @@
 
 struct a5xx_gpu {
 	struct adreno_gpu base;
-	struct platform_device *pdev;
 
 	struct drm_gem_object *pm4_bo;
 	uint64_t pm4_iova;
@@ -36,8 +35,6 @@ struct a5xx_gpu {
 	uint32_t gpmu_dwords;
 
 	uint32_t lm_leakage;
-
-	struct device zap_dev;
 };
 
 #define to_a5xx_gpu(x) container_of(x, struct a5xx_gpu, base)
@@ -59,5 +56,6 @@ static inline int spin_usecs(struct msm_gpu *gpu, uint32_t usecs,
 }
 
 bool a5xx_idle(struct msm_gpu *gpu);
+void a5xx_set_hwcg(struct msm_gpu *gpu, bool state);
 
 #endif /* __A5XX_GPU_H__ */
diff --git a/drivers/gpu/drm/msm/adreno/a5xx_power.c b/drivers/gpu/drm/msm/adreno/a5xx_power.c
index 87af6eea0483..04aab1dcae2b 100644
--- a/drivers/gpu/drm/msm/adreno/a5xx_power.c
+++ b/drivers/gpu/drm/msm/adreno/a5xx_power.c
@@ -294,16 +294,10 @@ void a5xx_gpmu_ucode_init(struct msm_gpu *gpu)
 	 */
 	bosize = (cmds_size + (cmds_size / TYPE4_MAX_PAYLOAD) + 1) << 2;
 
-	a5xx_gpu->gpmu_bo = msm_gem_new_locked(drm, bosize, MSM_BO_UNCACHED);
-	if (IS_ERR(a5xx_gpu->gpmu_bo))
-		goto err;
-
-	if (msm_gem_get_iova(a5xx_gpu->gpmu_bo, gpu->aspace,
-			&a5xx_gpu->gpmu_iova))
-		goto err;
-
-	ptr = msm_gem_get_vaddr(a5xx_gpu->gpmu_bo);
-	if (!ptr)
+	ptr = msm_gem_kernel_new_locked(drm, bosize,
+		MSM_BO_UNCACHED | MSM_BO_GPU_READONLY, gpu->aspace,
+		&a5xx_gpu->gpmu_bo, &a5xx_gpu->gpmu_iova);
+	if (IS_ERR(ptr))
 		goto err;
 
 	while (cmds_size > 0) {
diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.c b/drivers/gpu/drm/msm/adreno/adreno_gpu.c
index f1ab2703674a..c8b4ac254bb5 100644
--- a/drivers/gpu/drm/msm/adreno/adreno_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.c
@@ -48,8 +48,15 @@ int adreno_get_param(struct msm_gpu *gpu, uint32_t param, uint64_t *value)
 		*value = adreno_gpu->base.fast_rate;
 		return 0;
 	case MSM_PARAM_TIMESTAMP:
-		if (adreno_gpu->funcs->get_timestamp)
-			return adreno_gpu->funcs->get_timestamp(gpu, value);
+		if (adreno_gpu->funcs->get_timestamp) {
+			int ret;
+
+			pm_runtime_get_sync(&gpu->pdev->dev);
+			ret = adreno_gpu->funcs->get_timestamp(gpu, value);
+			pm_runtime_put_autosuspend(&gpu->pdev->dev);
+
+			return ret;
+		}
 		return -EINVAL;
 	default:
 		DBG("%s: invalid param: %u", gpu->name, param);
@@ -330,11 +337,6 @@ void adreno_wait_ring(struct msm_gpu *gpu, uint32_t ndwords)
 		DRM_ERROR("%s: timeout waiting for ringbuffer space\n", gpu->name);
 }
 
-static const char *iommu_ports[] = {
-		"gfx3d_user", "gfx3d_priv",
-		"gfx3d1_user", "gfx3d1_priv",
-};
-
 int adreno_gpu_init(struct drm_device *drm, struct platform_device *pdev,
 		struct adreno_gpu *adreno_gpu, const struct adreno_gpu_funcs *funcs)
 {
@@ -366,15 +368,15 @@ int adreno_gpu_init(struct drm_device *drm, struct platform_device *pdev,
 
 	adreno_gpu_config.ringsz = RB_SIZE;
 
+	pm_runtime_set_autosuspend_delay(&pdev->dev, DRM_MSM_INACTIVE_PERIOD);
+	pm_runtime_use_autosuspend(&pdev->dev);
+	pm_runtime_enable(&pdev->dev);
+
 	ret = msm_gpu_init(drm, pdev, &adreno_gpu->base, &funcs->base,
 			adreno_gpu->info->name, &adreno_gpu_config);
 	if (ret)
 		return ret;
 
-	pm_runtime_set_autosuspend_delay(&pdev->dev, DRM_MSM_INACTIVE_PERIOD);
-	pm_runtime_use_autosuspend(&pdev->dev);
-	pm_runtime_enable(&pdev->dev);
-
 	ret = request_firmware(&adreno_gpu->pm4, adreno_gpu->info->pm4fw, drm->dev);
 	if (ret) {
 		dev_err(drm->dev, "failed to load %s PM4 firmware: %d\n",
@@ -389,37 +391,17 @@ int adreno_gpu_init(struct drm_device *drm, struct platform_device *pdev,
 		return ret;
 	}
 
-	if (gpu->aspace && gpu->aspace->mmu) {
-		struct msm_mmu *mmu = gpu->aspace->mmu;
-		ret = mmu->funcs->attach(mmu, iommu_ports,
-				ARRAY_SIZE(iommu_ports));
-		if (ret)
-			return ret;
-	}
+	adreno_gpu->memptrs = msm_gem_kernel_new(drm,
+		sizeof(*adreno_gpu->memptrs), MSM_BO_UNCACHED, gpu->aspace,
+		&adreno_gpu->memptrs_bo, &adreno_gpu->memptrs_iova);
 
-	adreno_gpu->memptrs_bo = msm_gem_new(drm, sizeof(*adreno_gpu->memptrs),
-			MSM_BO_UNCACHED);
-	if (IS_ERR(adreno_gpu->memptrs_bo)) {
-		ret = PTR_ERR(adreno_gpu->memptrs_bo);
-		adreno_gpu->memptrs_bo = NULL;
-		dev_err(drm->dev, "could not allocate memptrs: %d\n", ret);
-		return ret;
-	}
-
-	adreno_gpu->memptrs = msm_gem_get_vaddr(adreno_gpu->memptrs_bo);
 	if (IS_ERR(adreno_gpu->memptrs)) {
-		dev_err(drm->dev, "could not vmap memptrs\n");
-		return -ENOMEM;
-	}
-
-	ret = msm_gem_get_iova(adreno_gpu->memptrs_bo, gpu->aspace,
-			&adreno_gpu->memptrs_iova);
-	if (ret) {
-		dev_err(drm->dev, "could not map memptrs: %d\n", ret);
-		return ret;
+		ret = PTR_ERR(adreno_gpu->memptrs);
+		adreno_gpu->memptrs = NULL;
+		dev_err(drm->dev, "could not allocate memptrs: %d\n", ret);
 	}
 
-	return 0;
+	return ret;
 }
 
 void adreno_gpu_cleanup(struct adreno_gpu *adreno_gpu)
@@ -439,10 +421,4 @@ void adreno_gpu_cleanup(struct adreno_gpu *adreno_gpu)
 	release_firmware(adreno_gpu->pfp);
 
 	msm_gpu_cleanup(gpu);
-
-	if (gpu->aspace) {
-		gpu->aspace->mmu->funcs->detach(gpu->aspace->mmu,
-			iommu_ports, ARRAY_SIZE(iommu_ports));
-		msm_gem_address_space_put(gpu->aspace);
-	}
 }
diff --git a/drivers/gpu/drm/msm/dsi/dsi.c b/drivers/gpu/drm/msm/dsi/dsi.c
index 311c1c1e7d6c..98742d7af6dc 100644
--- a/drivers/gpu/drm/msm/dsi/dsi.c
+++ b/drivers/gpu/drm/msm/dsi/dsi.c
@@ -161,12 +161,17 @@ static const struct of_device_id dt_match[] = {
 	{}
 };
 
+static const struct dev_pm_ops dsi_pm_ops = {
+	SET_RUNTIME_PM_OPS(msm_dsi_runtime_suspend, msm_dsi_runtime_resume, NULL)
+};
+
 static struct platform_driver dsi_driver = {
 	.probe = dsi_dev_probe,
 	.remove = dsi_dev_remove,
 	.driver = {
 		.name = "msm_dsi",
 		.of_match_table = dt_match,
+		.pm = &dsi_pm_ops,
 	},
 };
 
diff --git a/drivers/gpu/drm/msm/dsi/dsi.h b/drivers/gpu/drm/msm/dsi/dsi.h
index 9e6017387efb..2302046197a8 100644
--- a/drivers/gpu/drm/msm/dsi/dsi.h
+++ b/drivers/gpu/drm/msm/dsi/dsi.h
@@ -179,6 +179,8 @@ void msm_dsi_host_destroy(struct mipi_dsi_host *host);
 int msm_dsi_host_modeset_init(struct mipi_dsi_host *host,
 					struct drm_device *dev);
 int msm_dsi_host_init(struct msm_dsi *msm_dsi);
+int msm_dsi_runtime_suspend(struct device *dev);
+int msm_dsi_runtime_resume(struct device *dev);
 
 /* dsi phy */
 struct msm_dsi_phy;
diff --git a/drivers/gpu/drm/msm/dsi/dsi_host.c b/drivers/gpu/drm/msm/dsi/dsi_host.c
index 9e9c5696bc03..dbb31a014419 100644
--- a/drivers/gpu/drm/msm/dsi/dsi_host.c
+++ b/drivers/gpu/drm/msm/dsi/dsi_host.c
@@ -135,7 +135,6 @@ struct msm_dsi_host {
 	struct completion video_comp;
 	struct mutex dev_mutex;
 	struct mutex cmd_mutex;
-	struct mutex clk_mutex;
 	spinlock_t intr_lock; /* Protect interrupt ctrl register */
 
 	u32 err_work_state;
@@ -221,6 +220,8 @@ static const struct msm_dsi_cfg_handler *dsi_get_config(
 		goto put_gdsc;
 	}
 
+	pm_runtime_get_sync(dev);
+
 	ret = regulator_enable(gdsc_reg);
 	if (ret) {
 		pr_err("%s: unable to enable gdsc\n", __func__);
@@ -247,6 +248,7 @@ disable_clks:
 	clk_disable_unprepare(ahb_clk);
 disable_gdsc:
 	regulator_disable(gdsc_reg);
+	pm_runtime_put_autosuspend(dev);
 put_clk:
 	clk_put(ahb_clk);
 put_gdsc:
@@ -455,6 +457,34 @@ static void dsi_bus_clk_disable(struct msm_dsi_host *msm_host)
 		clk_disable_unprepare(msm_host->bus_clks[i]);
 }
 
+int msm_dsi_runtime_suspend(struct device *dev)
+{
+	struct platform_device *pdev = to_platform_device(dev);
+	struct msm_dsi *msm_dsi = platform_get_drvdata(pdev);
+	struct mipi_dsi_host *host = msm_dsi->host;
+	struct msm_dsi_host *msm_host = to_msm_dsi_host(host);
+
+	if (!msm_host->cfg_hnd)
+		return 0;
+
+	dsi_bus_clk_disable(msm_host);
+
+	return 0;
+}
+
+int msm_dsi_runtime_resume(struct device *dev)
+{
+	struct platform_device *pdev = to_platform_device(dev);
+	struct msm_dsi *msm_dsi = platform_get_drvdata(pdev);
+	struct mipi_dsi_host *host = msm_dsi->host;
+	struct msm_dsi_host *msm_host = to_msm_dsi_host(host);
+
+	if (!msm_host->cfg_hnd)
+		return 0;
+
+	return dsi_bus_clk_enable(msm_host);
+}
+
 static int dsi_link_clk_enable_6g(struct msm_dsi_host *msm_host)
 {
 	int ret;
@@ -596,35 +626,6 @@ static void dsi_link_clk_disable(struct msm_dsi_host *msm_host)
 	}
 }
 
-static int dsi_clk_ctrl(struct msm_dsi_host *msm_host, bool enable)
-{
-	int ret = 0;
-
-	mutex_lock(&msm_host->clk_mutex);
-	if (enable) {
-		ret = dsi_bus_clk_enable(msm_host);
-		if (ret) {
-			pr_err("%s: Can not enable bus clk, %d\n",
-				__func__, ret);
-			goto unlock_ret;
-		}
-		ret = dsi_link_clk_enable(msm_host);
-		if (ret) {
-			pr_err("%s: Can not enable link clk, %d\n",
-				__func__, ret);
-			dsi_bus_clk_disable(msm_host);
-			goto unlock_ret;
-		}
-	} else {
-		dsi_link_clk_disable(msm_host);
-		dsi_bus_clk_disable(msm_host);
-	}
-
-unlock_ret:
-	mutex_unlock(&msm_host->clk_mutex);
-	return ret;
-}
-
 static int dsi_calc_clk_rate(struct msm_dsi_host *msm_host)
 {
 	struct drm_display_mode *mode = msm_host->mode;
@@ -1699,6 +1700,7 @@ int msm_dsi_host_init(struct msm_dsi *msm_dsi)
 	}
 
 	msm_host->pdev = pdev;
+	msm_dsi->host = &msm_host->base;
 
 	ret = dsi_host_parse_dt(msm_host);
 	if (ret) {
@@ -1713,6 +1715,8 @@ int msm_dsi_host_init(struct msm_dsi *msm_dsi)
 		goto fail;
 	}
 
+	pm_runtime_enable(&pdev->dev);
+
 	msm_host->cfg_hnd = dsi_get_config(msm_host);
 	if (!msm_host->cfg_hnd) {
 		ret = -EINVAL;
@@ -1753,7 +1757,6 @@ int msm_dsi_host_init(struct msm_dsi *msm_dsi)
 	init_completion(&msm_host->video_comp);
 	mutex_init(&msm_host->dev_mutex);
 	mutex_init(&msm_host->cmd_mutex);
-	mutex_init(&msm_host->clk_mutex);
 	spin_lock_init(&msm_host->intr_lock);
 
 	/* setup workqueue */
@@ -1761,7 +1764,6 @@ int msm_dsi_host_init(struct msm_dsi *msm_dsi)
 	INIT_WORK(&msm_host->err_work, dsi_err_worker);
 	INIT_WORK(&msm_host->hpd_work, dsi_hpd_worker);
 
-	msm_dsi->host = &msm_host->base;
 	msm_dsi->id = msm_host->id;
 
 	DBG("Dsi Host %d initialized", msm_host->id);
@@ -1783,9 +1785,10 @@ void msm_dsi_host_destroy(struct mipi_dsi_host *host)
 		msm_host->workqueue = NULL;
 	}
 
-	mutex_destroy(&msm_host->clk_mutex);
 	mutex_destroy(&msm_host->cmd_mutex);
 	mutex_destroy(&msm_host->dev_mutex);
+
+	pm_runtime_disable(&msm_host->pdev->dev);
 }
 
 int msm_dsi_host_modeset_init(struct mipi_dsi_host *host,
@@ -1881,7 +1884,8 @@ int msm_dsi_host_xfer_prepare(struct mipi_dsi_host *host,
 	 * mdss interrupt is generated in mdp core clock domain
 	 * mdp clock need to be enabled to receive dsi interrupt
 	 */
-	dsi_clk_ctrl(msm_host, 1);
+	pm_runtime_get_sync(&msm_host->pdev->dev);
+	dsi_link_clk_enable(msm_host);
 
 	/* TODO: vote for bus bandwidth */
 
@@ -1911,7 +1915,8 @@ void msm_dsi_host_xfer_restore(struct mipi_dsi_host *host,
 
 	/* TODO: unvote for bus bandwidth */
 
-	dsi_clk_ctrl(msm_host, 0);
+	dsi_link_clk_disable(msm_host);
+	pm_runtime_put_autosuspend(&msm_host->pdev->dev);
 }
 
 int msm_dsi_host_cmd_tx(struct mipi_dsi_host *host,
@@ -2137,6 +2142,13 @@ void msm_dsi_host_get_phy_clk_req(struct mipi_dsi_host *host,
 	struct msm_dsi_phy_clk_request *clk_req)
 {
 	struct msm_dsi_host *msm_host = to_msm_dsi_host(host);
+	int ret;
+
+	ret = dsi_calc_clk_rate(msm_host);
+	if (ret) {
+		pr_err("%s: unable to calc clk rate, %d\n", __func__, ret);
+		return;
+	}
 
 	clk_req->bitclk_rate = msm_host->byte_clk_rate * 8;
 	clk_req->escclk_rate = msm_host->esc_clk_rate;
@@ -2153,8 +2165,11 @@ int msm_dsi_host_enable(struct mipi_dsi_host *host)
 	 * and only turned on before MDP START.
 	 * This part of code should be enabled once mdp driver support it.
 	 */
-	/* if (msm_panel->mode == MSM_DSI_CMD_MODE)
-		dsi_clk_ctrl(msm_host, 0); */
+	/* if (msm_panel->mode == MSM_DSI_CMD_MODE) {
+	 *	dsi_link_clk_disable(msm_host);
+	 *	pm_runtime_put_autosuspend(&msm_host->pdev->dev);
+	 * }
+	 */
 
 	return 0;
 }
@@ -2210,9 +2225,11 @@ int msm_dsi_host_power_on(struct mipi_dsi_host *host,
 		goto unlock_ret;
 	}
 
-	ret = dsi_clk_ctrl(msm_host, 1);
+	pm_runtime_get_sync(&msm_host->pdev->dev);
+	ret = dsi_link_clk_enable(msm_host);
 	if (ret) {
-		pr_err("%s: failed to enable clocks. ret=%d\n", __func__, ret);
+		pr_err("%s: failed to enable link clocks. ret=%d\n",
+		       __func__, ret);
 		goto fail_disable_reg;
 	}
 
@@ -2236,7 +2253,8 @@ int msm_dsi_host_power_on(struct mipi_dsi_host *host,
 	return 0;
 
 fail_disable_clk:
-	dsi_clk_ctrl(msm_host, 0);
+	dsi_link_clk_disable(msm_host);
+	pm_runtime_put_autosuspend(&msm_host->pdev->dev);
 fail_disable_reg:
 	dsi_host_regulator_disable(msm_host);
 unlock_ret:
@@ -2261,7 +2279,8 @@ int msm_dsi_host_power_off(struct mipi_dsi_host *host)
 
 	pinctrl_pm_select_sleep_state(&msm_host->pdev->dev);
 
-	dsi_clk_ctrl(msm_host, 0);
+	dsi_link_clk_disable(msm_host);
+	pm_runtime_put_autosuspend(&msm_host->pdev->dev);
 
 	dsi_host_regulator_disable(msm_host);
 
@@ -2280,7 +2299,6 @@ int msm_dsi_host_set_display_mode(struct mipi_dsi_host *host,
 					struct drm_display_mode *mode)
 {
 	struct msm_dsi_host *msm_host = to_msm_dsi_host(host);
-	int ret;
 
 	if (msm_host->mode) {
 		drm_mode_destroy(msm_host->dev, msm_host->mode);
@@ -2293,12 +2311,6 @@ int msm_dsi_host_set_display_mode(struct mipi_dsi_host *host,
 		return -ENOMEM;
 	}
 
-	ret = dsi_calc_clk_rate(msm_host);
-	if (ret) {
-		pr_err("%s: unable to calc clk rate, %d\n", __func__, ret);
-		return ret;
-	}
-
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c b/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c
index 0c2eb9c9a1fc..7c9bf91bc22b 100644
--- a/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c
+++ b/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c
@@ -373,7 +373,7 @@ static int dsi_phy_enable_resource(struct msm_dsi_phy *phy)
 static void dsi_phy_disable_resource(struct msm_dsi_phy *phy)
 {
 	clk_disable_unprepare(phy->ahb_clk);
-	pm_runtime_put_sync(&phy->pdev->dev);
+	pm_runtime_put_autosuspend(&phy->pdev->dev);
 }
 
 static const struct of_device_id dsi_phy_dt_match[] = {
diff --git a/drivers/gpu/drm/msm/hdmi/hdmi.c b/drivers/gpu/drm/msm/hdmi/hdmi.c
index a968cad509c2..17e069a133a4 100644
--- a/drivers/gpu/drm/msm/hdmi/hdmi.c
+++ b/drivers/gpu/drm/msm/hdmi/hdmi.c
@@ -239,6 +239,8 @@ static struct hdmi *msm_hdmi_init(struct platform_device *pdev)
 		hdmi->pwr_clks[i] = clk;
 	}
 
+	pm_runtime_enable(&pdev->dev);
+
 	hdmi->workq = alloc_ordered_workqueue("msm_hdmi", 0);
 
 	hdmi->i2c = msm_hdmi_i2c_init(hdmi);
diff --git a/drivers/gpu/drm/msm/hdmi/hdmi_bridge.c b/drivers/gpu/drm/msm/hdmi/hdmi_bridge.c
index 13ac822dee5d..7e357077ed26 100644
--- a/drivers/gpu/drm/msm/hdmi/hdmi_bridge.c
+++ b/drivers/gpu/drm/msm/hdmi/hdmi_bridge.c
@@ -35,6 +35,8 @@ static void msm_hdmi_power_on(struct drm_bridge *bridge)
 	const struct hdmi_platform_config *config = hdmi->config;
 	int i, ret;
 
+	pm_runtime_get_sync(&hdmi->pdev->dev);
+
 	for (i = 0; i < config->pwr_reg_cnt; i++) {
 		ret = regulator_enable(hdmi->pwr_regs[i]);
 		if (ret) {
@@ -84,6 +86,8 @@ static void power_off(struct drm_bridge *bridge)
 					config->pwr_reg_names[i], ret);
 		}
 	}
+
+	pm_runtime_put_autosuspend(&hdmi->pdev->dev);
 }
 
 #define AVI_IFRAME_LINE_NUMBER 1
diff --git a/drivers/gpu/drm/msm/hdmi/hdmi_connector.c b/drivers/gpu/drm/msm/hdmi/hdmi_connector.c
index 71536d9c7fe8..c0848dfedd50 100644
--- a/drivers/gpu/drm/msm/hdmi/hdmi_connector.c
+++ b/drivers/gpu/drm/msm/hdmi/hdmi_connector.c
@@ -137,6 +137,36 @@ err:
 	return ret;
 }
 
+static void enable_hpd_clocks(struct hdmi *hdmi, bool enable)
+{
+	const struct hdmi_platform_config *config = hdmi->config;
+	struct device *dev = &hdmi->pdev->dev;
+	int i, ret;
+
+	if (enable) {
+		for (i = 0; i < config->hpd_clk_cnt; i++) {
+			if (config->hpd_freq && config->hpd_freq[i]) {
+				ret = clk_set_rate(hdmi->hpd_clks[i],
+						   config->hpd_freq[i]);
+				if (ret)
+					dev_warn(dev,
+						 "failed to set clk %s (%d)\n",
+						 config->hpd_clk_names[i], ret);
+			}
+
+			ret = clk_prepare_enable(hdmi->hpd_clks[i]);
+			if (ret) {
+				dev_err(dev,
+					"failed to enable hpd clk: %s (%d)\n",
+					config->hpd_clk_names[i], ret);
+			}
+		}
+	} else {
+		for (i = config->hpd_clk_cnt - 1; i >= 0; i--)
+			clk_disable_unprepare(hdmi->hpd_clks[i]);
+	}
+}
+
 static int hpd_enable(struct hdmi_connector *hdmi_connector)
 {
 	struct hdmi *hdmi = hdmi_connector->hdmi;
@@ -167,22 +197,8 @@ static int hpd_enable(struct hdmi_connector *hdmi_connector)
 		goto fail;
 	}
 
-	for (i = 0; i < config->hpd_clk_cnt; i++) {
-		if (config->hpd_freq && config->hpd_freq[i]) {
-			ret = clk_set_rate(hdmi->hpd_clks[i],
-					config->hpd_freq[i]);
-			if (ret)
-				dev_warn(dev, "failed to set clk %s (%d)\n",
-						config->hpd_clk_names[i], ret);
-		}
-
-		ret = clk_prepare_enable(hdmi->hpd_clks[i]);
-		if (ret) {
-			dev_err(dev, "failed to enable hpd clk: %s (%d)\n",
-					config->hpd_clk_names[i], ret);
-			goto fail;
-		}
-	}
+	pm_runtime_get_sync(dev);
+	enable_hpd_clocks(hdmi, true);
 
 	msm_hdmi_set_mode(hdmi, false);
 	msm_hdmi_phy_reset(hdmi);
@@ -225,8 +241,8 @@ static void hdp_disable(struct hdmi_connector *hdmi_connector)
 
 	msm_hdmi_set_mode(hdmi, false);
 
-	for (i = 0; i < config->hpd_clk_cnt; i++)
-		clk_disable_unprepare(hdmi->hpd_clks[i]);
+	enable_hpd_clocks(hdmi, false);
+	pm_runtime_put_autosuspend(dev);
 
 	ret = gpio_config(hdmi, false);
 	if (ret)
@@ -285,7 +301,16 @@ void msm_hdmi_connector_irq(struct drm_connector *connector)
 
 static enum drm_connector_status detect_reg(struct hdmi *hdmi)
 {
-	uint32_t hpd_int_status = hdmi_read(hdmi, REG_HDMI_HPD_INT_STATUS);
+	uint32_t hpd_int_status;
+
+	pm_runtime_get_sync(&hdmi->pdev->dev);
+	enable_hpd_clocks(hdmi, true);
+
+	hpd_int_status = hdmi_read(hdmi, REG_HDMI_HPD_INT_STATUS);
+
+	enable_hpd_clocks(hdmi, false);
+	pm_runtime_put_autosuspend(&hdmi->pdev->dev);
+
 	return (hpd_int_status & HDMI_HPD_INT_STATUS_CABLE_DETECTED) ?
 			connector_status_connected : connector_status_disconnected;
 }
diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_cmd_encoder.c b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_cmd_encoder.c
index aa7402e03f67..60790df91bfa 100644
--- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_cmd_encoder.c
+++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_cmd_encoder.c
@@ -192,6 +192,7 @@ int mdp5_cmd_encoder_set_split_display(struct drm_encoder *encoder,
 {
 	struct mdp5_encoder *mdp5_cmd_enc = to_mdp5_encoder(encoder);
 	struct mdp5_kms *mdp5_kms;
+	struct device *dev;
 	int intf_num;
 	u32 data = 0;
 
@@ -214,14 +215,16 @@ int mdp5_cmd_encoder_set_split_display(struct drm_encoder *encoder,
 	/* Smart Panel, Sync mode */
 	data |= MDP5_SPLIT_DPL_UPPER_SMART_PANEL;
 
+	dev = &mdp5_kms->pdev->dev;
+
 	/* Make sure clocks are on when connectors calling this function. */
-	mdp5_enable(mdp5_kms);
+	pm_runtime_get_sync(dev);
 	mdp5_write(mdp5_kms, REG_MDP5_SPLIT_DPL_UPPER, data);
 
 	mdp5_write(mdp5_kms, REG_MDP5_SPLIT_DPL_LOWER,
 		   MDP5_SPLIT_DPL_LOWER_SMART_PANEL);
 	mdp5_write(mdp5_kms, REG_MDP5_SPLIT_DPL_EN, 1);
-	mdp5_disable(mdp5_kms);
+	pm_runtime_put_autosuspend(dev);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_crtc.c b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_crtc.c
index 5e3bc7224eee..6fcb58ab718c 100644
--- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_crtc.c
+++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_crtc.c
@@ -221,8 +221,8 @@ static void blend_setup(struct drm_crtc *crtc)
 	struct mdp5_ctl *ctl = mdp5_cstate->ctl;
 	uint32_t blend_op, fg_alpha, bg_alpha, ctl_blend_flags = 0;
 	unsigned long flags;
-	enum mdp5_pipe stage[STAGE_MAX + 1][MAX_PIPE_STAGE] = { SSPP_NONE };
-	enum mdp5_pipe r_stage[STAGE_MAX + 1][MAX_PIPE_STAGE] = { SSPP_NONE };
+	enum mdp5_pipe stage[STAGE_MAX + 1][MAX_PIPE_STAGE] = { { SSPP_NONE } };
+	enum mdp5_pipe r_stage[STAGE_MAX + 1][MAX_PIPE_STAGE] = { { SSPP_NONE } };
 	int i, plane_cnt = 0;
 	bool bg_alpha_enabled = false;
 	u32 mixer_op_mode = 0;
@@ -415,6 +415,7 @@ static void mdp5_crtc_atomic_disable(struct drm_crtc *crtc,
 	struct mdp5_crtc *mdp5_crtc = to_mdp5_crtc(crtc);
 	struct mdp5_crtc_state *mdp5_cstate = to_mdp5_crtc_state(crtc->state);
 	struct mdp5_kms *mdp5_kms = get_kms(crtc);
+	struct device *dev = &mdp5_kms->pdev->dev;
 
 	DBG("%s", crtc->name);
 
@@ -425,7 +426,7 @@ static void mdp5_crtc_atomic_disable(struct drm_crtc *crtc,
 		mdp_irq_unregister(&mdp5_kms->base, &mdp5_crtc->pp_done);
 
 	mdp_irq_unregister(&mdp5_kms->base, &mdp5_crtc->err);
-	mdp5_disable(mdp5_kms);
+	pm_runtime_put_autosuspend(dev);
 
 	mdp5_crtc->enabled = false;
 }
@@ -436,13 +437,17 @@ static void mdp5_crtc_atomic_enable(struct drm_crtc *crtc,
 	struct mdp5_crtc *mdp5_crtc = to_mdp5_crtc(crtc);
 	struct mdp5_crtc_state *mdp5_cstate = to_mdp5_crtc_state(crtc->state);
 	struct mdp5_kms *mdp5_kms = get_kms(crtc);
+	struct device *dev = &mdp5_kms->pdev->dev;
 
 	DBG("%s", crtc->name);
 
 	if (WARN_ON(mdp5_crtc->enabled))
 		return;
 
-	mdp5_enable(mdp5_kms);
+	pm_runtime_get_sync(dev);
+
+	mdp5_crtc_mode_set_nofb(crtc);
+
 	mdp_irq_register(&mdp5_kms->base, &mdp5_crtc->err);
 
 	if (mdp5_cstate->cmd_mode)
@@ -533,7 +538,7 @@ static bool is_fullscreen(struct drm_crtc_state *cstate,
 		((pstate->crtc_y + pstate->crtc_h) >= cstate->mode.vdisplay);
 }
 
-enum mdp_mixer_stage_id get_start_stage(struct drm_crtc *crtc,
+static enum mdp_mixer_stage_id get_start_stage(struct drm_crtc *crtc,
 					struct drm_crtc_state *new_crtc_state,
 					struct drm_plane_state *bpstate)
 {
@@ -727,6 +732,7 @@ static int mdp5_crtc_cursor_set(struct drm_crtc *crtc,
 	struct mdp5_pipeline *pipeline = &mdp5_cstate->pipeline;
 	struct drm_device *dev = crtc->dev;
 	struct mdp5_kms *mdp5_kms = get_kms(crtc);
+	struct platform_device *pdev = mdp5_kms->pdev;
 	struct msm_kms *kms = &mdp5_kms->base.base;
 	struct drm_gem_object *cursor_bo, *old_bo = NULL;
 	uint32_t blendcfg, stride;
@@ -755,6 +761,7 @@ static int mdp5_crtc_cursor_set(struct drm_crtc *crtc,
 	if (!handle) {
 		DBG("Cursor off");
 		cursor_enable = false;
+		pm_runtime_get_sync(&pdev->dev);
 		goto set_cursor;
 	}
 
@@ -769,6 +776,8 @@ static int mdp5_crtc_cursor_set(struct drm_crtc *crtc,
 	lm = mdp5_cstate->pipeline.mixer->lm;
 	stride = width * drm_format_plane_cpp(DRM_FORMAT_ARGB8888, 0);
 
+	pm_runtime_get_sync(&pdev->dev);
+
 	spin_lock_irqsave(&mdp5_crtc->cursor.lock, flags);
 	old_bo = mdp5_crtc->cursor.scanout_bo;
 
@@ -795,6 +804,8 @@ static int mdp5_crtc_cursor_set(struct drm_crtc *crtc,
 
 	spin_unlock_irqrestore(&mdp5_crtc->cursor.lock, flags);
 
+	pm_runtime_put_autosuspend(&pdev->dev);
+
 set_cursor:
 	ret = mdp5_ctl_set_cursor(ctl, pipeline, 0, cursor_enable);
 	if (ret) {
@@ -806,6 +817,7 @@ set_cursor:
 	crtc_flush(crtc, flush_mask);
 
 end:
+	pm_runtime_put_autosuspend(&pdev->dev);
 	if (old_bo) {
 		drm_flip_work_queue(&mdp5_crtc->unref_cursor_work, old_bo);
 		/* enable vblank to complete cursor work: */
@@ -838,6 +850,8 @@ static int mdp5_crtc_cursor_move(struct drm_crtc *crtc, int x, int y)
 
 	get_roi(crtc, &roi_w, &roi_h);
 
+	pm_runtime_get_sync(&mdp5_kms->pdev->dev);
+
 	spin_lock_irqsave(&mdp5_crtc->cursor.lock, flags);
 	mdp5_write(mdp5_kms, REG_MDP5_LM_CURSOR_SIZE(lm),
 			MDP5_LM_CURSOR_SIZE_ROI_H(roi_h) |
@@ -849,6 +863,8 @@ static int mdp5_crtc_cursor_move(struct drm_crtc *crtc, int x, int y)
 
 	crtc_flush(crtc, flush_mask);
 
+	pm_runtime_put_autosuspend(&mdp5_kms->pdev->dev);
+
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_encoder.c b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_encoder.c
index 97f3294fbfc6..5b851380d3f2 100644
--- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_encoder.c
+++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_encoder.c
@@ -297,9 +297,13 @@ static void mdp5_encoder_enable(struct drm_encoder *encoder)
 {
 	struct mdp5_encoder *mdp5_encoder = to_mdp5_encoder(encoder);
 	struct mdp5_interface *intf = mdp5_encoder->intf;
+	/* this isn't right I think */
+	struct drm_crtc_state *cstate = encoder->crtc->state;
+
+	mdp5_encoder_mode_set(encoder, &cstate->mode, &cstate->adjusted_mode);
 
 	if (intf->mode == MDP5_INTF_DSI_MODE_COMMAND)
-		mdp5_cmd_encoder_disable(encoder);
+		mdp5_cmd_encoder_enable(encoder);
 	else
 		mdp5_vid_encoder_enable(encoder);
 }
@@ -320,7 +324,6 @@ static int mdp5_encoder_atomic_check(struct drm_encoder *encoder,
 }
 
 static const struct drm_encoder_helper_funcs mdp5_encoder_helper_funcs = {
-	.mode_set = mdp5_encoder_mode_set,
 	.disable = mdp5_encoder_disable,
 	.enable = mdp5_encoder_enable,
 	.atomic_check = mdp5_encoder_atomic_check,
@@ -350,6 +353,7 @@ int mdp5_vid_encoder_set_split_display(struct drm_encoder *encoder,
 	struct mdp5_encoder *mdp5_encoder = to_mdp5_encoder(encoder);
 	struct mdp5_encoder *mdp5_slave_enc = to_mdp5_encoder(slave_encoder);
 	struct mdp5_kms *mdp5_kms;
+	struct device *dev;
 	int intf_num;
 	u32 data = 0;
 
@@ -369,8 +373,10 @@ int mdp5_vid_encoder_set_split_display(struct drm_encoder *encoder,
 	else
 		return -EINVAL;
 
+	dev = &mdp5_kms->pdev->dev;
 	/* Make sure clocks are on when connectors calling this function. */
-	mdp5_enable(mdp5_kms);
+	pm_runtime_get_sync(dev);
+
 	/* Dumb Panel, Sync mode */
 	mdp5_write(mdp5_kms, REG_MDP5_SPLIT_DPL_UPPER, 0);
 	mdp5_write(mdp5_kms, REG_MDP5_SPLIT_DPL_LOWER, data);
@@ -378,7 +384,7 @@ int mdp5_vid_encoder_set_split_display(struct drm_encoder *encoder,
 
 	mdp5_ctl_pair(mdp5_encoder->ctl, mdp5_slave_enc->ctl, true);
 
-	mdp5_disable(mdp5_kms);
+	pm_runtime_put_autosuspend(dev);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_irq.c b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_irq.c
index 3ce8b9dec9c1..bb5deb00c899 100644
--- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_irq.c
+++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_irq.c
@@ -49,16 +49,19 @@ static void mdp5_irq_error_handler(struct mdp_irq *irq, uint32_t irqstatus)
 void mdp5_irq_preinstall(struct msm_kms *kms)
 {
 	struct mdp5_kms *mdp5_kms = to_mdp5_kms(to_mdp_kms(kms));
-	mdp5_enable(mdp5_kms);
+	struct device *dev = &mdp5_kms->pdev->dev;
+
+	pm_runtime_get_sync(dev);
 	mdp5_write(mdp5_kms, REG_MDP5_INTR_CLEAR, 0xffffffff);
 	mdp5_write(mdp5_kms, REG_MDP5_INTR_EN, 0x00000000);
-	mdp5_disable(mdp5_kms);
+	pm_runtime_put_autosuspend(dev);
 }
 
 int mdp5_irq_postinstall(struct msm_kms *kms)
 {
 	struct mdp_kms *mdp_kms = to_mdp_kms(kms);
 	struct mdp5_kms *mdp5_kms = to_mdp5_kms(mdp_kms);
+	struct device *dev = &mdp5_kms->pdev->dev;
 	struct mdp_irq *error_handler = &mdp5_kms->error_handler;
 
 	error_handler->irq = mdp5_irq_error_handler;
@@ -67,9 +70,9 @@ int mdp5_irq_postinstall(struct msm_kms *kms)
 			MDP5_IRQ_INTF2_UNDER_RUN |
 			MDP5_IRQ_INTF3_UNDER_RUN;
 
-	mdp5_enable(mdp5_kms);
+	pm_runtime_get_sync(dev);
 	mdp_irq_register(mdp_kms, error_handler);
-	mdp5_disable(mdp5_kms);
+	pm_runtime_put_autosuspend(dev);
 
 	return 0;
 }
@@ -77,9 +80,11 @@ int mdp5_irq_postinstall(struct msm_kms *kms)
 void mdp5_irq_uninstall(struct msm_kms *kms)
 {
 	struct mdp5_kms *mdp5_kms = to_mdp5_kms(to_mdp_kms(kms));
-	mdp5_enable(mdp5_kms);
+	struct device *dev = &mdp5_kms->pdev->dev;
+
+	pm_runtime_get_sync(dev);
 	mdp5_write(mdp5_kms, REG_MDP5_INTR_EN, 0x00000000);
-	mdp5_disable(mdp5_kms);
+	pm_runtime_put_autosuspend(dev);
 }
 
 irqreturn_t mdp5_irq(struct msm_kms *kms)
@@ -109,11 +114,12 @@ irqreturn_t mdp5_irq(struct msm_kms *kms)
 int mdp5_enable_vblank(struct msm_kms *kms, struct drm_crtc *crtc)
 {
 	struct mdp5_kms *mdp5_kms = to_mdp5_kms(to_mdp_kms(kms));
+	struct device *dev = &mdp5_kms->pdev->dev;
 
-	mdp5_enable(mdp5_kms);
+	pm_runtime_get_sync(dev);
 	mdp_update_vblank_mask(to_mdp_kms(kms),
 			mdp5_crtc_vblank(crtc), true);
-	mdp5_disable(mdp5_kms);
+	pm_runtime_put_autosuspend(dev);
 
 	return 0;
 }
@@ -121,9 +127,10 @@ int mdp5_enable_vblank(struct msm_kms *kms, struct drm_crtc *crtc)
 void mdp5_disable_vblank(struct msm_kms *kms, struct drm_crtc *crtc)
 {
 	struct mdp5_kms *mdp5_kms = to_mdp5_kms(to_mdp_kms(kms));
+	struct device *dev = &mdp5_kms->pdev->dev;
 
-	mdp5_enable(mdp5_kms);
+	pm_runtime_get_sync(dev);
 	mdp_update_vblank_mask(to_mdp_kms(kms),
 			mdp5_crtc_vblank(crtc), false);
-	mdp5_disable(mdp5_kms);
+	pm_runtime_put_autosuspend(dev);
 }
diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c
index 5d13fa5381ee..f7c0698fec40 100644
--- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c
+++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c
@@ -30,11 +30,10 @@ static const char *iommu_ports[] = {
 static int mdp5_hw_init(struct msm_kms *kms)
 {
 	struct mdp5_kms *mdp5_kms = to_mdp5_kms(to_mdp_kms(kms));
-	struct platform_device *pdev = mdp5_kms->pdev;
+	struct device *dev = &mdp5_kms->pdev->dev;
 	unsigned long flags;
 
-	pm_runtime_get_sync(&pdev->dev);
-	mdp5_enable(mdp5_kms);
+	pm_runtime_get_sync(dev);
 
 	/* Magic unknown register writes:
 	 *
@@ -66,8 +65,7 @@ static int mdp5_hw_init(struct msm_kms *kms)
 
 	mdp5_ctlm_hw_reset(mdp5_kms->ctlm);
 
-	mdp5_disable(mdp5_kms);
-	pm_runtime_put_sync(&pdev->dev);
+	pm_runtime_put_sync(dev);
 
 	return 0;
 }
@@ -111,8 +109,9 @@ static void mdp5_swap_state(struct msm_kms *kms, struct drm_atomic_state *state)
 static void mdp5_prepare_commit(struct msm_kms *kms, struct drm_atomic_state *state)
 {
 	struct mdp5_kms *mdp5_kms = to_mdp5_kms(to_mdp_kms(kms));
+	struct device *dev = &mdp5_kms->pdev->dev;
 
-	mdp5_enable(mdp5_kms);
+	pm_runtime_get_sync(dev);
 
 	if (mdp5_kms->smp)
 		mdp5_smp_prepare_commit(mdp5_kms->smp, &mdp5_kms->state->smp);
@@ -121,11 +120,12 @@ static void mdp5_prepare_commit(struct msm_kms *kms, struct drm_atomic_state *st
 static void mdp5_complete_commit(struct msm_kms *kms, struct drm_atomic_state *state)
 {
 	struct mdp5_kms *mdp5_kms = to_mdp5_kms(to_mdp_kms(kms));
+	struct device *dev = &mdp5_kms->pdev->dev;
 
 	if (mdp5_kms->smp)
 		mdp5_smp_complete_commit(mdp5_kms->smp, &mdp5_kms->state->smp);
 
-	mdp5_disable(mdp5_kms);
+	pm_runtime_put_autosuspend(dev);
 }
 
 static void mdp5_wait_for_crtc_commit_done(struct msm_kms *kms,
@@ -249,6 +249,9 @@ int mdp5_disable(struct mdp5_kms *mdp5_kms)
 {
 	DBG("");
 
+	mdp5_kms->enable_count--;
+	WARN_ON(mdp5_kms->enable_count < 0);
+
 	clk_disable_unprepare(mdp5_kms->ahb_clk);
 	clk_disable_unprepare(mdp5_kms->axi_clk);
 	clk_disable_unprepare(mdp5_kms->core_clk);
@@ -262,6 +265,8 @@ int mdp5_enable(struct mdp5_kms *mdp5_kms)
 {
 	DBG("");
 
+	mdp5_kms->enable_count++;
+
 	clk_prepare_enable(mdp5_kms->ahb_clk);
 	clk_prepare_enable(mdp5_kms->axi_clk);
 	clk_prepare_enable(mdp5_kms->core_clk);
@@ -486,11 +491,12 @@ fail:
 static void read_mdp_hw_revision(struct mdp5_kms *mdp5_kms,
 				 u32 *major, u32 *minor)
 {
+	struct device *dev = &mdp5_kms->pdev->dev;
 	u32 version;
 
-	mdp5_enable(mdp5_kms);
+	pm_runtime_get_sync(dev);
 	version = mdp5_read(mdp5_kms, REG_MDP5_HW_VERSION);
-	mdp5_disable(mdp5_kms);
+	pm_runtime_put_autosuspend(dev);
 
 	*major = FIELD(version, MDP5_HW_VERSION_MAJOR);
 	*minor = FIELD(version, MDP5_HW_VERSION_MINOR);
@@ -502,7 +508,7 @@ static int get_clk(struct platform_device *pdev, struct clk **clkp,
 		const char *name, bool mandatory)
 {
 	struct device *dev = &pdev->dev;
-	struct clk *clk = devm_clk_get(dev, name);
+	struct clk *clk = msm_clk_get(pdev, name);
 	if (IS_ERR(clk) && mandatory) {
 		dev_err(dev, "failed to get %s (%ld)\n", name, PTR_ERR(clk));
 		return PTR_ERR(clk);
@@ -643,7 +649,7 @@ struct msm_kms *mdp5_kms_init(struct drm_device *dev)
 	 * have left things on, in which case we'll start getting faults if
 	 * we don't disable):
 	 */
-	mdp5_enable(mdp5_kms);
+	pm_runtime_get_sync(&pdev->dev);
 	for (i = 0; i < MDP5_INTF_NUM_MAX; i++) {
 		if (mdp5_cfg_intf_is_virtual(config->hw->intf.connect[i]) ||
 		    !config->hw->intf.base[i])
@@ -652,7 +658,6 @@ struct msm_kms *mdp5_kms_init(struct drm_device *dev)
 
 		mdp5_write(mdp5_kms, REG_MDP5_INTF_FRAME_LINE_COUNT_EN(i), 0x3);
 	}
-	mdp5_disable(mdp5_kms);
 	mdelay(16);
 
 	if (config->platform.iommu) {
@@ -678,6 +683,8 @@ struct msm_kms *mdp5_kms_init(struct drm_device *dev)
 		aspace = NULL;;
 	}
 
+	pm_runtime_put_autosuspend(&pdev->dev);
+
 	ret = modeset_init(mdp5_kms);
 	if (ret) {
 		dev_err(&pdev->dev, "modeset_init failed: %d\n", ret);
@@ -887,21 +894,21 @@ static int mdp5_init(struct platform_device *pdev, struct drm_device *dev)
 	}
 
 	/* mandatory clocks: */
-	ret = get_clk(pdev, &mdp5_kms->axi_clk, "bus_clk", true);
+	ret = get_clk(pdev, &mdp5_kms->axi_clk, "bus", true);
 	if (ret)
 		goto fail;
-	ret = get_clk(pdev, &mdp5_kms->ahb_clk, "iface_clk", true);
+	ret = get_clk(pdev, &mdp5_kms->ahb_clk, "iface", true);
 	if (ret)
 		goto fail;
-	ret = get_clk(pdev, &mdp5_kms->core_clk, "core_clk", true);
+	ret = get_clk(pdev, &mdp5_kms->core_clk, "core", true);
 	if (ret)
 		goto fail;
-	ret = get_clk(pdev, &mdp5_kms->vsync_clk, "vsync_clk", true);
+	ret = get_clk(pdev, &mdp5_kms->vsync_clk, "vsync", true);
 	if (ret)
 		goto fail;
 
 	/* optional clocks: */
-	get_clk(pdev, &mdp5_kms->lut_clk, "lut_clk", false);
+	get_clk(pdev, &mdp5_kms->lut_clk, "lut", false);
 
 	/* we need to set a default rate before enabling.  Set a safe
 	 * rate first, then figure out hw revision, and then set a
@@ -1005,6 +1012,30 @@ static int mdp5_dev_remove(struct platform_device *pdev)
 	return 0;
 }
 
+static __maybe_unused int mdp5_runtime_suspend(struct device *dev)
+{
+	struct platform_device *pdev = to_platform_device(dev);
+	struct mdp5_kms *mdp5_kms = platform_get_drvdata(pdev);
+
+	DBG("");
+
+	return mdp5_disable(mdp5_kms);
+}
+
+static __maybe_unused int mdp5_runtime_resume(struct device *dev)
+{
+	struct platform_device *pdev = to_platform_device(dev);
+	struct mdp5_kms *mdp5_kms = platform_get_drvdata(pdev);
+
+	DBG("");
+
+	return mdp5_enable(mdp5_kms);
+}
+
+static const struct dev_pm_ops mdp5_pm_ops = {
+	SET_RUNTIME_PM_OPS(mdp5_runtime_suspend, mdp5_runtime_resume, NULL)
+};
+
 static const struct of_device_id mdp5_dt_match[] = {
 	{ .compatible = "qcom,mdp5", },
 	/* to support downstream DT files */
@@ -1019,6 +1050,7 @@ static struct platform_driver mdp5_driver = {
 	.driver = {
 		.name = "msm_mdp",
 		.of_match_table = mdp5_dt_match,
+		.pm = &mdp5_pm_ops,
 	},
 };
 
diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.h b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.h
index 17caa0e8c8ae..9b3fe01089d1 100644
--- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.h
+++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.h
@@ -76,6 +76,8 @@ struct mdp5_kms {
 	bool rpm_enabled;
 
 	struct mdp_irq error_handler;
+
+	int enable_count;
 };
 #define to_mdp5_kms(x) container_of(x, struct mdp5_kms, base)
 
@@ -167,11 +169,13 @@ struct mdp5_encoder {
 
 static inline void mdp5_write(struct mdp5_kms *mdp5_kms, u32 reg, u32 data)
 {
+	WARN_ON(mdp5_kms->enable_count <= 0);
 	msm_writel(data, mdp5_kms->mmio + reg);
 }
 
 static inline u32 mdp5_read(struct mdp5_kms *mdp5_kms, u32 reg)
 {
+	WARN_ON(mdp5_kms->enable_count <= 0);
 	return msm_readl(mdp5_kms->mmio + reg);
 }
 
@@ -255,9 +259,6 @@ static inline uint32_t lm2ppdone(struct mdp5_hw_mixer *mixer)
 	return MDP5_IRQ_PING_PONG_0_DONE << mixer->pp;
 }
 
-int mdp5_disable(struct mdp5_kms *mdp5_kms);
-int mdp5_enable(struct mdp5_kms *mdp5_kms);
-
 void mdp5_set_irqmask(struct mdp_kms *mdp_kms, uint32_t irqmask,
 		uint32_t old_irqmask);
 void mdp5_irq_preinstall(struct msm_kms *kms);
diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_mdss.c b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_mdss.c
index 9c34d7824988..f2a0db7a8a03 100644
--- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_mdss.c
+++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_mdss.c
@@ -31,6 +31,10 @@ struct msm_mdss {
 
 	struct regulator *vdd;
 
+	struct clk *ahb_clk;
+	struct clk *axi_clk;
+	struct clk *vsync_clk;
+
 	struct {
 		volatile unsigned long enabled_mask;
 		struct irq_domain *domain;
@@ -140,6 +144,51 @@ static int mdss_irq_domain_init(struct msm_mdss *mdss)
 	return 0;
 }
 
+int msm_mdss_enable(struct msm_mdss *mdss)
+{
+	DBG("");
+
+	clk_prepare_enable(mdss->ahb_clk);
+	if (mdss->axi_clk)
+		clk_prepare_enable(mdss->axi_clk);
+	if (mdss->vsync_clk)
+		clk_prepare_enable(mdss->vsync_clk);
+
+	return 0;
+}
+
+int msm_mdss_disable(struct msm_mdss *mdss)
+{
+	DBG("");
+
+	if (mdss->vsync_clk)
+		clk_disable_unprepare(mdss->vsync_clk);
+	if (mdss->axi_clk)
+		clk_disable_unprepare(mdss->axi_clk);
+	clk_disable_unprepare(mdss->ahb_clk);
+
+	return 0;
+}
+
+static int msm_mdss_get_clocks(struct msm_mdss *mdss)
+{
+	struct platform_device *pdev = to_platform_device(mdss->dev->dev);
+
+	mdss->ahb_clk = msm_clk_get(pdev, "iface");
+	if (IS_ERR(mdss->ahb_clk))
+		mdss->ahb_clk = NULL;
+
+	mdss->axi_clk = msm_clk_get(pdev, "bus");
+	if (IS_ERR(mdss->axi_clk))
+		mdss->axi_clk = NULL;
+
+	mdss->vsync_clk = msm_clk_get(pdev, "vsync");
+	if (IS_ERR(mdss->vsync_clk))
+		mdss->vsync_clk = NULL;
+
+	return 0;
+}
+
 void msm_mdss_destroy(struct drm_device *dev)
 {
 	struct msm_drm_private *priv = dev->dev_private;
@@ -153,8 +202,6 @@ void msm_mdss_destroy(struct drm_device *dev)
 
 	regulator_disable(mdss->vdd);
 
-	pm_runtime_put_sync(dev->dev);
-
 	pm_runtime_disable(dev->dev);
 }
 
@@ -190,6 +237,12 @@ int msm_mdss_init(struct drm_device *dev)
 		goto fail;
 	}
 
+	ret = msm_mdss_get_clocks(mdss);
+	if (ret) {
+		dev_err(dev->dev, "failed to get clocks: %d\n", ret);
+		goto fail;
+	}
+
 	/* Regulator to enable GDSCs in downstream kernels */
 	mdss->vdd = devm_regulator_get(dev->dev, "vdd");
 	if (IS_ERR(mdss->vdd)) {
@@ -221,12 +274,6 @@ int msm_mdss_init(struct drm_device *dev)
 
 	pm_runtime_enable(dev->dev);
 
-	/*
-	 * TODO: This is needed as the MDSS GDSC is only tied to MDSS's power
-	 * domain. Remove this once runtime PM is adapted for all the devices.
-	 */
-	pm_runtime_get_sync(dev->dev);
-
 	return 0;
 fail_irq:
 	regulator_disable(mdss->vdd);
diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_plane.c b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_plane.c
index 818244ac4a4b..4b22ac3413a1 100644
--- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_plane.c
+++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_plane.c
@@ -888,8 +888,8 @@ static int mdp5_plane_mode_set(struct drm_plane *plane,
 	struct mdp5_hw_pipe *right_hwpipe;
 	const struct mdp_format *format;
 	uint32_t nplanes, config = 0;
-	struct phase_step step = { 0 };
-	struct pixel_ext pe = { 0 };
+	struct phase_step step = { { 0 } };
+	struct pixel_ext pe = { { 0 } };
 	uint32_t hdecm = 0, vdecm = 0;
 	uint32_t pix_format;
 	unsigned int rotation;
diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_smp.c b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_smp.c
index 58f712d37e7f..ae4983d9d0a5 100644
--- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_smp.c
+++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_smp.c
@@ -28,6 +28,13 @@ struct mdp5_smp {
 
 	int blk_cnt;
 	int blk_size;
+
+	/* register cache */
+	u32 alloc_w[22];
+	u32 alloc_r[22];
+	u32 pipe_reqprio_fifo_wm0[SSPP_MAX];
+	u32 pipe_reqprio_fifo_wm1[SSPP_MAX];
+	u32 pipe_reqprio_fifo_wm2[SSPP_MAX];
 };
 
 static inline
@@ -98,16 +105,15 @@ static int smp_request_block(struct mdp5_smp *smp,
 static void set_fifo_thresholds(struct mdp5_smp *smp,
 		enum mdp5_pipe pipe, int nblks)
 {
-	struct mdp5_kms *mdp5_kms = get_kms(smp);
 	u32 smp_entries_per_blk = smp->blk_size / (128 / BITS_PER_BYTE);
 	u32 val;
 
 	/* 1/4 of SMP pool that is being fetched */
 	val = (nblks * smp_entries_per_blk) / 4;
 
-	mdp5_write(mdp5_kms, REG_MDP5_PIPE_REQPRIO_FIFO_WM_0(pipe), val * 1);
-	mdp5_write(mdp5_kms, REG_MDP5_PIPE_REQPRIO_FIFO_WM_1(pipe), val * 2);
-	mdp5_write(mdp5_kms, REG_MDP5_PIPE_REQPRIO_FIFO_WM_2(pipe), val * 3);
+	smp->pipe_reqprio_fifo_wm0[pipe] = val * 1;
+	smp->pipe_reqprio_fifo_wm1[pipe] = val * 2;
+	smp->pipe_reqprio_fifo_wm2[pipe] = val * 3;
 }
 
 /*
@@ -222,7 +228,6 @@ void mdp5_smp_release(struct mdp5_smp *smp, struct mdp5_smp_state *state,
 static unsigned update_smp_state(struct mdp5_smp *smp,
 		u32 cid, mdp5_smp_state_t *assigned)
 {
-	struct mdp5_kms *mdp5_kms = get_kms(smp);
 	int cnt = smp->blk_cnt;
 	unsigned nblks = 0;
 	u32 blk, val;
@@ -231,7 +236,7 @@ static unsigned update_smp_state(struct mdp5_smp *smp,
 		int idx = blk / 3;
 		int fld = blk % 3;
 
-		val = mdp5_read(mdp5_kms, REG_MDP5_SMP_ALLOC_W_REG(idx));
+		val = smp->alloc_w[idx];
 
 		switch (fld) {
 		case 0:
@@ -248,8 +253,8 @@ static unsigned update_smp_state(struct mdp5_smp *smp,
 			break;
 		}
 
-		mdp5_write(mdp5_kms, REG_MDP5_SMP_ALLOC_W_REG(idx), val);
-		mdp5_write(mdp5_kms, REG_MDP5_SMP_ALLOC_R_REG(idx), val);
+		smp->alloc_w[idx] = val;
+		smp->alloc_r[idx] = val;
 
 		nblks++;
 	}
@@ -257,6 +262,39 @@ static unsigned update_smp_state(struct mdp5_smp *smp,
 	return nblks;
 }
 
+static void write_smp_alloc_regs(struct mdp5_smp *smp)
+{
+	struct mdp5_kms *mdp5_kms = get_kms(smp);
+	int i, num_regs;
+
+	num_regs = smp->blk_cnt / 3 + 1;
+
+	for (i = 0; i < num_regs; i++) {
+		mdp5_write(mdp5_kms, REG_MDP5_SMP_ALLOC_W_REG(i),
+			   smp->alloc_w[i]);
+		mdp5_write(mdp5_kms, REG_MDP5_SMP_ALLOC_R_REG(i),
+			   smp->alloc_r[i]);
+	}
+}
+
+static void write_smp_fifo_regs(struct mdp5_smp *smp)
+{
+	struct mdp5_kms *mdp5_kms = get_kms(smp);
+	int i;
+
+	for (i = 0; i < mdp5_kms->num_hwpipes; i++) {
+		struct mdp5_hw_pipe *hwpipe = mdp5_kms->hwpipes[i];
+		enum mdp5_pipe pipe = hwpipe->pipe;
+
+		mdp5_write(mdp5_kms, REG_MDP5_PIPE_REQPRIO_FIFO_WM_0(pipe),
+			   smp->pipe_reqprio_fifo_wm0[pipe]);
+		mdp5_write(mdp5_kms, REG_MDP5_PIPE_REQPRIO_FIFO_WM_1(pipe),
+			   smp->pipe_reqprio_fifo_wm1[pipe]);
+		mdp5_write(mdp5_kms, REG_MDP5_PIPE_REQPRIO_FIFO_WM_2(pipe),
+			   smp->pipe_reqprio_fifo_wm2[pipe]);
+	}
+}
+
 void mdp5_smp_prepare_commit(struct mdp5_smp *smp, struct mdp5_smp_state *state)
 {
 	enum mdp5_pipe pipe;
@@ -277,6 +315,9 @@ void mdp5_smp_prepare_commit(struct mdp5_smp *smp, struct mdp5_smp_state *state)
 		set_fifo_thresholds(smp, pipe, nblks);
 	}
 
+	write_smp_alloc_regs(smp);
+	write_smp_fifo_regs(smp);
+
 	state->assigned = 0;
 }
 
@@ -289,6 +330,8 @@ void mdp5_smp_complete_commit(struct mdp5_smp *smp, struct mdp5_smp_state *state
 		set_fifo_thresholds(smp, pipe, 0);
 	}
 
+	write_smp_fifo_regs(smp);
+
 	state->released = 0;
 }
 
diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c
index f49f6ac5585c..606df7bea97b 100644
--- a/drivers/gpu/drm/msm/msm_drv.c
+++ b/drivers/gpu/drm/msm/msm_drv.c
@@ -73,6 +73,10 @@ bool dumpstate = false;
 MODULE_PARM_DESC(dumpstate, "Dump KMS state on errors");
 module_param(dumpstate, bool, 0600);
 
+static bool modeset = true;
+MODULE_PARM_DESC(modeset, "Use kernel modesetting [KMS] (1=on (default), 0=disable)");
+module_param(modeset, bool, 0600);
+
 /*
  * Util/helpers:
  */
@@ -832,7 +836,6 @@ static struct drm_driver msm_driver = {
 	.gem_vm_ops         = &vm_ops,
 	.dumb_create        = msm_gem_dumb_create,
 	.dumb_map_offset    = msm_gem_dumb_map_offset,
-	.dumb_destroy       = drm_gem_dumb_destroy,
 	.prime_handle_to_fd = drm_gem_prime_handle_to_fd,
 	.prime_fd_to_handle = drm_gem_prime_fd_to_handle,
 	.gem_prime_export   = drm_gem_prime_export,
@@ -879,8 +882,37 @@ static int msm_pm_resume(struct device *dev)
 }
 #endif
 
+#ifdef CONFIG_PM
+static int msm_runtime_suspend(struct device *dev)
+{
+	struct drm_device *ddev = dev_get_drvdata(dev);
+	struct msm_drm_private *priv = ddev->dev_private;
+
+	DBG("");
+
+	if (priv->mdss)
+		return msm_mdss_disable(priv->mdss);
+
+	return 0;
+}
+
+static int msm_runtime_resume(struct device *dev)
+{
+	struct drm_device *ddev = dev_get_drvdata(dev);
+	struct msm_drm_private *priv = ddev->dev_private;
+
+	DBG("");
+
+	if (priv->mdss)
+		return msm_mdss_enable(priv->mdss);
+
+	return 0;
+}
+#endif
+
 static const struct dev_pm_ops msm_pm_ops = {
 	SET_SYSTEM_SLEEP_PM_OPS(msm_pm_suspend, msm_pm_resume)
+	SET_RUNTIME_PM_OPS(msm_runtime_suspend, msm_runtime_resume, NULL)
 };
 
 /*
@@ -1104,6 +1136,9 @@ static struct platform_driver msm_platform_driver = {
 
 static int __init msm_drm_register(void)
 {
+	if (!modeset)
+		return -EINVAL;
+
 	DBG("init");
 	msm_mdp_register();
 	msm_dsi_register();
diff --git a/drivers/gpu/drm/msm/msm_drv.h b/drivers/gpu/drm/msm/msm_drv.h
index fc8d24f7c084..5e8109c07560 100644
--- a/drivers/gpu/drm/msm/msm_drv.h
+++ b/drivers/gpu/drm/msm/msm_drv.h
@@ -55,8 +55,6 @@ struct msm_fence_cb;
 struct msm_gem_address_space;
 struct msm_gem_vma;
 
-#define NUM_DOMAINS 2    /* one for KMS, then one per gpu core (?) */
-
 struct msm_file_private {
 	/* currently we don't do anything useful with this.. but when
 	 * per-context address spaces are supported we'd keep track of
@@ -237,6 +235,12 @@ struct drm_gem_object *msm_gem_new(struct drm_device *dev,
 		uint32_t size, uint32_t flags);
 struct drm_gem_object *msm_gem_new_locked(struct drm_device *dev,
 		uint32_t size, uint32_t flags);
+void *msm_gem_kernel_new(struct drm_device *dev, uint32_t size,
+		uint32_t flags, struct msm_gem_address_space *aspace,
+		struct drm_gem_object **bo, uint64_t *iova);
+void *msm_gem_kernel_new_locked(struct drm_device *dev, uint32_t size,
+		uint32_t flags, struct msm_gem_address_space *aspace,
+		struct drm_gem_object **bo, uint64_t *iova);
 struct drm_gem_object *msm_gem_import(struct drm_device *dev,
 		struct dma_buf *dmabuf, struct sg_table *sgt);
 
@@ -248,10 +252,10 @@ uint32_t msm_framebuffer_iova(struct drm_framebuffer *fb,
 		struct msm_gem_address_space *aspace, int plane);
 struct drm_gem_object *msm_framebuffer_bo(struct drm_framebuffer *fb, int plane);
 const struct msm_format *msm_framebuffer_format(struct drm_framebuffer *fb);
-struct drm_framebuffer *msm_framebuffer_init(struct drm_device *dev,
-		const struct drm_mode_fb_cmd2 *mode_cmd, struct drm_gem_object **bos);
 struct drm_framebuffer *msm_framebuffer_create(struct drm_device *dev,
 		struct drm_file *file, const struct drm_mode_fb_cmd2 *mode_cmd);
+struct drm_framebuffer * msm_alloc_stolen_fb(struct drm_device *dev,
+		int w, int h, int p, uint32_t format);
 
 struct drm_fb_helper *msm_fbdev_init(struct drm_device *dev);
 void msm_fbdev_free(struct drm_device *dev);
diff --git a/drivers/gpu/drm/msm/msm_fb.c b/drivers/gpu/drm/msm/msm_fb.c
index 6ecb7b170316..fc175e724ad6 100644
--- a/drivers/gpu/drm/msm/msm_fb.c
+++ b/drivers/gpu/drm/msm/msm_fb.c
@@ -20,6 +20,7 @@
 
 #include "msm_drv.h"
 #include "msm_kms.h"
+#include "msm_gem.h"
 
 struct msm_framebuffer {
 	struct drm_framebuffer base;
@@ -28,6 +29,8 @@ struct msm_framebuffer {
 };
 #define to_msm_framebuffer(x) container_of(x, struct msm_framebuffer, base)
 
+static struct drm_framebuffer *msm_framebuffer_init(struct drm_device *dev,
+		const struct drm_mode_fb_cmd2 *mode_cmd, struct drm_gem_object **bos);
 
 static int msm_framebuffer_create_handle(struct drm_framebuffer *fb,
 		struct drm_file *file_priv,
@@ -161,7 +164,7 @@ out_unref:
 	return ERR_PTR(ret);
 }
 
-struct drm_framebuffer *msm_framebuffer_init(struct drm_device *dev,
+static struct drm_framebuffer *msm_framebuffer_init(struct drm_device *dev,
 		const struct drm_mode_fb_cmd2 *mode_cmd, struct drm_gem_object **bos)
 {
 	struct msm_drm_private *priv = dev->dev_private;
@@ -237,3 +240,43 @@ fail:
 
 	return ERR_PTR(ret);
 }
+
+struct drm_framebuffer *
+msm_alloc_stolen_fb(struct drm_device *dev, int w, int h, int p, uint32_t format)
+{
+	struct drm_mode_fb_cmd2 mode_cmd = {
+		.pixel_format = format,
+		.width = w,
+		.height = h,
+		.pitches = { p },
+	};
+	struct drm_gem_object *bo;
+	struct drm_framebuffer *fb;
+	int size;
+
+	/* allocate backing bo */
+	size = mode_cmd.pitches[0] * mode_cmd.height;
+	DBG("allocating %d bytes for fb %d", size, dev->primary->index);
+	bo = msm_gem_new(dev, size, MSM_BO_SCANOUT | MSM_BO_WC | MSM_BO_STOLEN);
+	if (IS_ERR(bo)) {
+		dev_warn(dev->dev, "could not allocate stolen bo\n");
+		/* try regular bo: */
+		bo = msm_gem_new(dev, size, MSM_BO_SCANOUT | MSM_BO_WC);
+	}
+	if (IS_ERR(bo)) {
+		dev_err(dev->dev, "failed to allocate buffer object\n");
+		return ERR_CAST(bo);
+	}
+
+	fb = msm_framebuffer_init(dev, &mode_cmd, &bo);
+	if (IS_ERR(fb)) {
+		dev_err(dev->dev, "failed to allocate fb\n");
+		/* note: if fb creation failed, we can't rely on fb destroy
+		 * to unref the bo:
+		 */
+		drm_gem_object_unreference_unlocked(bo);
+		return ERR_CAST(fb);
+	}
+
+	return fb;
+}
diff --git a/drivers/gpu/drm/msm/msm_fbdev.c b/drivers/gpu/drm/msm/msm_fbdev.c
index 9c00fedfc741..c178563fcd4d 100644
--- a/drivers/gpu/drm/msm/msm_fbdev.c
+++ b/drivers/gpu/drm/msm/msm_fbdev.c
@@ -19,7 +19,6 @@
 #include <drm/drm_fb_helper.h>
 
 #include "msm_drv.h"
-#include "msm_gem.h"
 #include "msm_kms.h"
 
 extern int msm_gem_mmap_obj(struct drm_gem_object *obj,
@@ -35,7 +34,6 @@ static int msm_fbdev_mmap(struct fb_info *info, struct vm_area_struct *vma);
 struct msm_fbdev {
 	struct drm_fb_helper base;
 	struct drm_framebuffer *fb;
-	struct drm_gem_object *bo;
 };
 
 static struct fb_ops msm_fb_ops = {
@@ -57,16 +55,16 @@ static int msm_fbdev_mmap(struct fb_info *info, struct vm_area_struct *vma)
 {
 	struct drm_fb_helper *helper = (struct drm_fb_helper *)info->par;
 	struct msm_fbdev *fbdev = to_msm_fbdev(helper);
-	struct drm_gem_object *drm_obj = fbdev->bo;
+	struct drm_gem_object *bo = msm_framebuffer_bo(fbdev->fb, 0);
 	int ret = 0;
 
-	ret = drm_gem_mmap_obj(drm_obj, drm_obj->size, vma);
+	ret = drm_gem_mmap_obj(bo, bo->size, vma);
 	if (ret) {
 		pr_err("%s:drm_gem_mmap_obj fail\n", __func__);
 		return ret;
 	}
 
-	return msm_gem_mmap_obj(drm_obj, vma);
+	return msm_gem_mmap_obj(bo, vma);
 }
 
 static int msm_fbdev_create(struct drm_fb_helper *helper,
@@ -76,47 +74,30 @@ static int msm_fbdev_create(struct drm_fb_helper *helper,
 	struct drm_device *dev = helper->dev;
 	struct msm_drm_private *priv = dev->dev_private;
 	struct drm_framebuffer *fb = NULL;
+	struct drm_gem_object *bo;
 	struct fb_info *fbi = NULL;
-	struct drm_mode_fb_cmd2 mode_cmd = {0};
 	uint64_t paddr;
-	int ret, size;
+	uint32_t format;
+	int ret, pitch;
+
+	format = drm_mode_legacy_fb_format(sizes->surface_bpp, sizes->surface_depth);
 
 	DBG("create fbdev: %dx%d@%d (%dx%d)", sizes->surface_width,
 			sizes->surface_height, sizes->surface_bpp,
 			sizes->fb_width, sizes->fb_height);
 
-	mode_cmd.pixel_format = drm_mode_legacy_fb_format(sizes->surface_bpp,
-			sizes->surface_depth);
-
-	mode_cmd.width = sizes->surface_width;
-	mode_cmd.height = sizes->surface_height;
-
-	mode_cmd.pitches[0] = align_pitch(
-			mode_cmd.width, sizes->surface_bpp);
+	pitch = align_pitch(sizes->surface_width, sizes->surface_bpp);
+	fb = msm_alloc_stolen_fb(dev, sizes->surface_width,
+			sizes->surface_height, pitch, format);
 
-	/* allocate backing bo */
-	size = mode_cmd.pitches[0] * mode_cmd.height;
-	DBG("allocating %d bytes for fb %d", size, dev->primary->index);
-	fbdev->bo = msm_gem_new(dev, size, MSM_BO_SCANOUT |
-			MSM_BO_WC | MSM_BO_STOLEN);
-	if (IS_ERR(fbdev->bo)) {
-		ret = PTR_ERR(fbdev->bo);
-		fbdev->bo = NULL;
-		dev_err(dev->dev, "failed to allocate buffer object: %d\n", ret);
-		goto fail;
-	}
-
-	fb = msm_framebuffer_init(dev, &mode_cmd, &fbdev->bo);
 	if (IS_ERR(fb)) {
 		dev_err(dev->dev, "failed to allocate fb\n");
-		/* note: if fb creation failed, we can't rely on fb destroy
-		 * to unref the bo:
-		 */
-		drm_gem_object_unreference_unlocked(fbdev->bo);
 		ret = PTR_ERR(fb);
 		goto fail;
 	}
 
+	bo = msm_framebuffer_bo(fb, 0);
+
 	mutex_lock(&dev->struct_mutex);
 
 	/*
@@ -124,7 +105,7 @@ static int msm_fbdev_create(struct drm_fb_helper *helper,
 	 * in panic (ie. lock-safe, etc) we could avoid pinning the
 	 * buffer now:
 	 */
-	ret = msm_gem_get_iova(fbdev->bo, priv->kms->aspace, &paddr);
+	ret = msm_gem_get_iova(bo, priv->kms->aspace, &paddr);
 	if (ret) {
 		dev_err(dev->dev, "failed to get buffer obj iova: %d\n", ret);
 		goto fail_unlock;
@@ -152,14 +133,14 @@ static int msm_fbdev_create(struct drm_fb_helper *helper,
 
 	dev->mode_config.fb_base = paddr;
 
-	fbi->screen_base = msm_gem_get_vaddr(fbdev->bo);
+	fbi->screen_base = msm_gem_get_vaddr(bo);
 	if (IS_ERR(fbi->screen_base)) {
 		ret = PTR_ERR(fbi->screen_base);
 		goto fail_unlock;
 	}
-	fbi->screen_size = fbdev->bo->size;
+	fbi->screen_size = bo->size;
 	fbi->fix.smem_start = paddr;
-	fbi->fix.smem_len = fbdev->bo->size;
+	fbi->fix.smem_len = bo->size;
 
 	DBG("par=%p, %dx%d", fbi->par, fbi->var.xres, fbi->var.yres);
 	DBG("allocated %dx%d fb", fbdev->fb->width, fbdev->fb->height);
@@ -241,7 +222,9 @@ void msm_fbdev_free(struct drm_device *dev)
 
 	/* this will free the backing object */
 	if (fbdev->fb) {
-		msm_gem_put_vaddr(fbdev->bo);
+		struct drm_gem_object *bo =
+			msm_framebuffer_bo(fbdev->fb, 0);
+		msm_gem_put_vaddr(bo);
 		drm_framebuffer_remove(fbdev->fb);
 	}
 
diff --git a/drivers/gpu/drm/msm/msm_gem.c b/drivers/gpu/drm/msm/msm_gem.c
index 65f35544c1ec..f15821a0d900 100644
--- a/drivers/gpu/drm/msm/msm_gem.c
+++ b/drivers/gpu/drm/msm/msm_gem.c
@@ -383,8 +383,10 @@ int msm_gem_get_iova(struct drm_gem_object *obj,
 		struct page **pages;
 
 		vma = add_vma(obj, aspace);
-		if (IS_ERR(vma))
-			return PTR_ERR(vma);
+		if (IS_ERR(vma)) {
+			ret = PTR_ERR(vma);
+			goto unlock;
+		}
 
 		pages = get_pages(obj);
 		if (IS_ERR(pages)) {
@@ -405,7 +407,7 @@ int msm_gem_get_iova(struct drm_gem_object *obj,
 
 fail:
 	del_vma(vma);
-
+unlock:
 	mutex_unlock(&msm_obj->lock);
 	return ret;
 }
@@ -928,8 +930,12 @@ static struct drm_gem_object *_msm_gem_new(struct drm_device *dev,
 	if (use_vram) {
 		struct msm_gem_vma *vma;
 		struct page **pages;
+		struct msm_gem_object *msm_obj = to_msm_bo(obj);
+
+		mutex_lock(&msm_obj->lock);
 
 		vma = add_vma(obj, NULL);
+		mutex_unlock(&msm_obj->lock);
 		if (IS_ERR(vma)) {
 			ret = PTR_ERR(vma);
 			goto fail;
@@ -1018,3 +1024,49 @@ fail:
 	drm_gem_object_unreference_unlocked(obj);
 	return ERR_PTR(ret);
 }
+
+static void *_msm_gem_kernel_new(struct drm_device *dev, uint32_t size,
+		uint32_t flags, struct msm_gem_address_space *aspace,
+		struct drm_gem_object **bo, uint64_t *iova, bool locked)
+{
+	void *vaddr;
+	struct drm_gem_object *obj = _msm_gem_new(dev, size, flags, locked);
+	int ret;
+
+	if (IS_ERR(obj))
+		return ERR_CAST(obj);
+
+	if (iova) {
+		ret = msm_gem_get_iova(obj, aspace, iova);
+		if (ret) {
+			drm_gem_object_unreference(obj);
+			return ERR_PTR(ret);
+		}
+	}
+
+	vaddr = msm_gem_get_vaddr(obj);
+	if (!vaddr) {
+		msm_gem_put_iova(obj, aspace);
+		drm_gem_object_unreference(obj);
+		return ERR_PTR(-ENOMEM);
+	}
+
+	if (bo)
+		*bo = obj;
+
+	return vaddr;
+}
+
+void *msm_gem_kernel_new(struct drm_device *dev, uint32_t size,
+		uint32_t flags, struct msm_gem_address_space *aspace,
+		struct drm_gem_object **bo, uint64_t *iova)
+{
+	return _msm_gem_kernel_new(dev, size, flags, aspace, bo, iova, false);
+}
+
+void *msm_gem_kernel_new_locked(struct drm_device *dev, uint32_t size,
+		uint32_t flags, struct msm_gem_address_space *aspace,
+		struct drm_gem_object **bo, uint64_t *iova)
+{
+	return _msm_gem_kernel_new(dev, size, flags, aspace, bo, iova, true);
+}
diff --git a/drivers/gpu/drm/msm/msm_gem_submit.c b/drivers/gpu/drm/msm/msm_gem_submit.c
index 6bfca7470141..8a75c0bd8a78 100644
--- a/drivers/gpu/drm/msm/msm_gem_submit.c
+++ b/drivers/gpu/drm/msm/msm_gem_submit.c
@@ -34,8 +34,8 @@ static struct msm_gem_submit *submit_create(struct drm_device *dev,
 		struct msm_gpu *gpu, uint32_t nr_bos, uint32_t nr_cmds)
 {
 	struct msm_gem_submit *submit;
-	uint64_t sz = sizeof(*submit) + (nr_bos * sizeof(submit->bos[0])) +
-		(nr_cmds * sizeof(submit->cmd[0]));
+	uint64_t sz = sizeof(*submit) + ((u64)nr_bos * sizeof(submit->bos[0])) +
+		((u64)nr_cmds * sizeof(submit->cmd[0]));
 
 	if (sz > SIZE_MAX)
 		return NULL;
@@ -451,7 +451,7 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data,
 	if (ret)
 		goto out;
 
-	if (!(args->fence & MSM_SUBMIT_NO_IMPLICIT)) {
+	if (!(args->flags & MSM_SUBMIT_NO_IMPLICIT)) {
 		ret = submit_fence_sync(submit);
 		if (ret)
 			goto out;
diff --git a/drivers/gpu/drm/msm/msm_gem_vma.c b/drivers/gpu/drm/msm/msm_gem_vma.c
index c36321bc8714..d34e331554f3 100644
--- a/drivers/gpu/drm/msm/msm_gem_vma.c
+++ b/drivers/gpu/drm/msm/msm_gem_vma.c
@@ -42,7 +42,7 @@ void
 msm_gem_unmap_vma(struct msm_gem_address_space *aspace,
 		struct msm_gem_vma *vma, struct sg_table *sgt)
 {
-	if (!vma->iova)
+	if (!aspace || !vma->iova)
 		return;
 
 	if (aspace->mmu) {
diff --git a/drivers/gpu/drm/msm/msm_gpu.c b/drivers/gpu/drm/msm/msm_gpu.c
index 9f3dbc236ab3..ffbff27600e0 100644
--- a/drivers/gpu/drm/msm/msm_gpu.c
+++ b/drivers/gpu/drm/msm/msm_gpu.c
@@ -562,11 +562,49 @@ static int get_clocks(struct platform_device *pdev, struct msm_gpu *gpu)
 	return 0;
 }
 
+static struct msm_gem_address_space *
+msm_gpu_create_address_space(struct msm_gpu *gpu, struct platform_device *pdev,
+		uint64_t va_start, uint64_t va_end)
+{
+	struct iommu_domain *iommu;
+	struct msm_gem_address_space *aspace;
+	int ret;
+
+	/*
+	 * Setup IOMMU.. eventually we will (I think) do this once per context
+	 * and have separate page tables per context.  For now, to keep things
+	 * simple and to get something working, just use a single address space:
+	 */
+	iommu = iommu_domain_alloc(&platform_bus_type);
+	if (!iommu)
+		return NULL;
+
+	iommu->geometry.aperture_start = va_start;
+	iommu->geometry.aperture_end = va_end;
+
+	dev_info(gpu->dev->dev, "%s: using IOMMU\n", gpu->name);
+
+	aspace = msm_gem_address_space_create(&pdev->dev, iommu, "gpu");
+	if (IS_ERR(aspace)) {
+		dev_err(gpu->dev->dev, "failed to init iommu: %ld\n",
+			PTR_ERR(aspace));
+		iommu_domain_free(iommu);
+		return ERR_CAST(aspace);
+	}
+
+	ret = aspace->mmu->funcs->attach(aspace->mmu, NULL, 0);
+	if (ret) {
+		msm_gem_address_space_put(aspace);
+		return ERR_PTR(ret);
+	}
+
+	return aspace;
+}
+
 int msm_gpu_init(struct drm_device *drm, struct platform_device *pdev,
 		struct msm_gpu *gpu, const struct msm_gpu_funcs *funcs,
 		const char *name, struct msm_gpu_config *config)
 {
-	struct iommu_domain *iommu;
 	int ret;
 
 	if (WARN_ON(gpu->num_perfcntrs > ARRAY_SIZE(gpu->last_cntrs)))
@@ -636,28 +674,19 @@ int msm_gpu_init(struct drm_device *drm, struct platform_device *pdev,
 	if (IS_ERR(gpu->gpu_cx))
 		gpu->gpu_cx = NULL;
 
-	/* Setup IOMMU.. eventually we will (I think) do this once per context
-	 * and have separate page tables per context.  For now, to keep things
-	 * simple and to get something working, just use a single address space:
-	 */
-	iommu = iommu_domain_alloc(&platform_bus_type);
-	if (iommu) {
-		iommu->geometry.aperture_start = config->va_start;
-		iommu->geometry.aperture_end = config->va_end;
-
-		dev_info(drm->dev, "%s: using IOMMU\n", name);
-		gpu->aspace = msm_gem_address_space_create(&pdev->dev,
-				iommu, "gpu");
-		if (IS_ERR(gpu->aspace)) {
-			ret = PTR_ERR(gpu->aspace);
-			dev_err(drm->dev, "failed to init iommu: %d\n", ret);
-			gpu->aspace = NULL;
-			iommu_domain_free(iommu);
-			goto fail;
-		}
+	gpu->pdev = pdev;
+	platform_set_drvdata(pdev, gpu);
+
+	bs_init(gpu);
 
-	} else {
+	gpu->aspace = msm_gpu_create_address_space(gpu, pdev,
+		config->va_start, config->va_end);
+
+	if (gpu->aspace == NULL)
 		dev_info(drm->dev, "%s: no IOMMU, fallback to VRAM carveout!\n", name);
+	else if (IS_ERR(gpu->aspace)) {
+		ret = PTR_ERR(gpu->aspace);
+		goto fail;
 	}
 
 	/* Create ringbuffer: */
@@ -669,14 +698,10 @@ int msm_gpu_init(struct drm_device *drm, struct platform_device *pdev,
 		goto fail;
 	}
 
-	gpu->pdev = pdev;
-	platform_set_drvdata(pdev, gpu);
-
-	bs_init(gpu);
-
 	return 0;
 
 fail:
+	platform_set_drvdata(pdev, NULL);
 	return ret;
 }
 
@@ -693,7 +718,9 @@ void msm_gpu_cleanup(struct msm_gpu *gpu)
 			msm_gem_put_iova(gpu->rb->bo, gpu->aspace);
 		msm_ringbuffer_destroy(gpu->rb);
 	}
-
-	if (gpu->fctx)
-		msm_fence_context_free(gpu->fctx);
+	if (gpu->aspace) {
+		gpu->aspace->mmu->funcs->detach(gpu->aspace->mmu,
+			NULL, 0);
+		msm_gem_address_space_put(gpu->aspace);
+	}
 }
diff --git a/drivers/gpu/drm/msm/msm_kms.h b/drivers/gpu/drm/msm/msm_kms.h
index a8f2ba5e5f07..17d5824417ad 100644
--- a/drivers/gpu/drm/msm/msm_kms.h
+++ b/drivers/gpu/drm/msm/msm_kms.h
@@ -99,5 +99,7 @@ struct msm_kms *mdp4_kms_init(struct drm_device *dev);
 struct msm_kms *mdp5_kms_init(struct drm_device *dev);
 int msm_mdss_init(struct drm_device *dev);
 void msm_mdss_destroy(struct drm_device *dev);
+int msm_mdss_enable(struct msm_mdss *mdss);
+int msm_mdss_disable(struct msm_mdss *mdss);
 
 #endif /* __MSM_KMS_H__ */
diff --git a/drivers/gpu/drm/msm/msm_ringbuffer.c b/drivers/gpu/drm/msm/msm_ringbuffer.c
index 791bca3c6a9c..bf065a540130 100644
--- a/drivers/gpu/drm/msm/msm_ringbuffer.c
+++ b/drivers/gpu/drm/msm/msm_ringbuffer.c
@@ -33,16 +33,14 @@ struct msm_ringbuffer *msm_ringbuffer_new(struct msm_gpu *gpu, int size)
 	}
 
 	ring->gpu = gpu;
-	ring->bo = msm_gem_new(gpu->dev, size, MSM_BO_WC);
-	if (IS_ERR(ring->bo)) {
-		ret = PTR_ERR(ring->bo);
-		ring->bo = NULL;
-		goto fail;
-	}
 
-	ring->start = msm_gem_get_vaddr(ring->bo);
+	/* Pass NULL for the iova pointer - we will map it later */
+	ring->start = msm_gem_kernel_new(gpu->dev, size, MSM_BO_WC,
+		gpu->aspace, &ring->bo, NULL);
+
 	if (IS_ERR(ring->start)) {
 		ret = PTR_ERR(ring->start);
+		ring->start = 0;
 		goto fail;
 	}
 	ring->end   = ring->start + (size / 4);
diff --git a/drivers/gpu/drm/mxsfb/mxsfb_drv.c b/drivers/gpu/drm/mxsfb/mxsfb_drv.c
index 93c38eb6d187..7fbad9cb656e 100644
--- a/drivers/gpu/drm/mxsfb/mxsfb_drv.c
+++ b/drivers/gpu/drm/mxsfb/mxsfb_drv.c
@@ -337,8 +337,6 @@ static struct drm_driver mxsfb_driver = {
 	.gem_free_object_unlocked = drm_gem_cma_free_object,
 	.gem_vm_ops		= &drm_gem_cma_vm_ops,
 	.dumb_create		= drm_gem_cma_dumb_create,
-	.dumb_map_offset	= drm_gem_cma_dumb_map_offset,
-	.dumb_destroy		= drm_gem_dumb_destroy,
 	.prime_handle_to_fd	= drm_gem_prime_handle_to_fd,
 	.prime_fd_to_handle	= drm_gem_prime_fd_to_handle,
 	.gem_prime_export	= drm_gem_prime_export,
diff --git a/drivers/gpu/drm/nouveau/dispnv04/crtc.c b/drivers/gpu/drm/nouveau/dispnv04/crtc.c
index 8f689f1f6122..6aa6ee16dcbd 100644
--- a/drivers/gpu/drm/nouveau/dispnv04/crtc.c
+++ b/drivers/gpu/drm/nouveau/dispnv04/crtc.c
@@ -1096,6 +1096,38 @@ static const struct drm_crtc_helper_funcs nv04_crtc_helper_funcs = {
 	.disable = nv_crtc_disable,
 };
 
+static const uint32_t modeset_formats[] = {
+        DRM_FORMAT_XRGB8888,
+        DRM_FORMAT_RGB565,
+        DRM_FORMAT_XRGB1555,
+};
+
+static struct drm_plane *
+create_primary_plane(struct drm_device *dev)
+{
+        struct drm_plane *primary;
+        int ret;
+
+        primary = kzalloc(sizeof(*primary), GFP_KERNEL);
+        if (primary == NULL) {
+                DRM_DEBUG_KMS("Failed to allocate primary plane\n");
+                return NULL;
+        }
+
+        /* possible_crtc's will be filled in later by crtc_init */
+        ret = drm_universal_plane_init(dev, primary, 0,
+                                       &drm_primary_helper_funcs,
+                                       modeset_formats,
+                                       ARRAY_SIZE(modeset_formats), NULL,
+                                       DRM_PLANE_TYPE_PRIMARY, NULL);
+        if (ret) {
+                kfree(primary);
+                primary = NULL;
+        }
+
+        return primary;
+}
+
 int
 nv04_crtc_create(struct drm_device *dev, int crtc_num)
 {
@@ -1114,7 +1146,9 @@ nv04_crtc_create(struct drm_device *dev, int crtc_num)
 	nv_crtc->save = nv_crtc_save;
 	nv_crtc->restore = nv_crtc_restore;
 
-	drm_crtc_init(dev, &nv_crtc->base, &nv04_crtc_funcs);
+	drm_crtc_init_with_planes(dev, &nv_crtc->base,
+                                  create_primary_plane(dev), NULL,
+                                  &nv04_crtc_funcs, NULL);
 	drm_crtc_helper_add(&nv_crtc->base, &nv04_crtc_helper_funcs);
 	drm_mode_crtc_set_gamma_size(&nv_crtc->base, 256);
 
diff --git a/drivers/gpu/drm/nouveau/dispnv04/overlay.c b/drivers/gpu/drm/nouveau/dispnv04/overlay.c
index e54944d23268..c8c2333f24ee 100644
--- a/drivers/gpu/drm/nouveau/dispnv04/overlay.c
+++ b/drivers/gpu/drm/nouveau/dispnv04/overlay.c
@@ -63,6 +63,7 @@ static uint32_t formats[] = {
 	DRM_FORMAT_YUYV,
 	DRM_FORMAT_UYVY,
 	DRM_FORMAT_NV12,
+	DRM_FORMAT_NV21,
 };
 
 /* Sine can be approximated with
@@ -90,6 +91,26 @@ cos_mul(int degrees, int factor)
 }
 
 static int
+verify_scaling(const struct drm_framebuffer *fb, uint8_t shift,
+               uint32_t src_x, uint32_t src_y, uint32_t src_w, uint32_t src_h,
+               uint32_t crtc_w, uint32_t crtc_h)
+{
+	if (crtc_w < (src_w >> shift) || crtc_h < (src_h >> shift)) {
+		DRM_DEBUG_KMS("Unsuitable framebuffer scaling: %dx%d -> %dx%d\n",
+			      src_w, src_h, crtc_w, crtc_h);
+		return -ERANGE;
+	}
+
+	if (src_x != 0 || src_y != 0) {
+		DRM_DEBUG_KMS("Unsuitable framebuffer offset: %d,%d\n",
+                              src_x, src_y);
+		return -ERANGE;
+	}
+
+	return 0;
+}
+
+static int
 nv10_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 		  struct drm_framebuffer *fb, int crtc_x, int crtc_y,
 		  unsigned int crtc_w, unsigned int crtc_h,
@@ -107,7 +128,9 @@ nv10_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 	bool flip = nv_plane->flip;
 	int soff = NV_PCRTC0_SIZE * nv_crtc->index;
 	int soff2 = NV_PCRTC0_SIZE * !nv_crtc->index;
-	int format, ret;
+	unsigned shift = drm->client.device.info.chipset >= 0x30 ? 1 : 3;
+	unsigned format = 0;
+	int ret;
 
 	/* Source parameters given in 16.16 fixed point, ignore fractional. */
 	src_x >>= 16;
@@ -115,18 +138,9 @@ nv10_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 	src_w >>= 16;
 	src_h >>= 16;
 
-	format = ALIGN(src_w * 4, 0x100);
-
-	if (format > 0xffff)
-		return -ERANGE;
-
-	if (drm->client.device.info.chipset >= 0x30) {
-		if (crtc_w < (src_w >> 1) || crtc_h < (src_h >> 1))
-			return -ERANGE;
-	} else {
-		if (crtc_w < (src_w >> 3) || crtc_h < (src_h >> 3))
-			return -ERANGE;
-	}
+	ret = verify_scaling(fb, shift, 0, 0, src_w, src_h, crtc_w, crtc_h);
+	if (ret)
+		return ret;
 
 	ret = nouveau_bo_pin(nv_fb->nvbo, TTM_PL_FLAG_VRAM, false);
 	if (ret)
@@ -146,21 +160,23 @@ nv10_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 	nvif_wr32(dev, NV_PVIDEO_POINT_OUT(flip), crtc_y << 16 | crtc_x);
 	nvif_wr32(dev, NV_PVIDEO_SIZE_OUT(flip), crtc_h << 16 | crtc_w);
 
-	if (fb->format->format != DRM_FORMAT_UYVY)
+	if (fb->format->format == DRM_FORMAT_YUYV ||
+	    fb->format->format == DRM_FORMAT_NV12)
 		format |= NV_PVIDEO_FORMAT_COLOR_LE_CR8YB8CB8YA8;
-	if (fb->format->format == DRM_FORMAT_NV12)
+	if (fb->format->format == DRM_FORMAT_NV12 ||
+	    fb->format->format == DRM_FORMAT_NV21)
 		format |= NV_PVIDEO_FORMAT_PLANAR;
 	if (nv_plane->iturbt_709)
 		format |= NV_PVIDEO_FORMAT_MATRIX_ITURBT709;
 	if (nv_plane->colorkey & (1 << 24))
 		format |= NV_PVIDEO_FORMAT_DISPLAY_COLOR_KEY;
 
-	if (fb->format->format == DRM_FORMAT_NV12) {
+	if (format & NV_PVIDEO_FORMAT_PLANAR) {
 		nvif_wr32(dev, NV_PVIDEO_UVPLANE_BASE(flip), 0);
 		nvif_wr32(dev, NV_PVIDEO_UVPLANE_OFFSET_BUFF(flip),
 			nv_fb->nvbo->bo.offset + fb->offsets[1]);
 	}
-	nvif_wr32(dev, NV_PVIDEO_FORMAT(flip), format);
+	nvif_wr32(dev, NV_PVIDEO_FORMAT(flip), format | fb->pitches[0]);
 	nvif_wr32(dev, NV_PVIDEO_STOP, 0);
 	/* TODO: wait for vblank? */
 	nvif_wr32(dev, NV_PVIDEO_BUFFER, flip ? 0x10 : 0x1);
@@ -357,7 +373,7 @@ nv04_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 	struct nouveau_bo *cur = nv_plane->cur;
 	uint32_t overlay = 1;
 	int brightness = (nv_plane->brightness - 512) * 62 / 512;
-	int pitch, ret, i;
+	int ret, i;
 
 	/* Source parameters given in 16.16 fixed point, ignore fractional. */
 	src_x >>= 16;
@@ -365,17 +381,9 @@ nv04_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 	src_w >>= 16;
 	src_h >>= 16;
 
-	pitch = ALIGN(src_w * 4, 0x100);
-
-	if (pitch > 0xffff)
-		return -ERANGE;
-
-	/* TODO: Compute an offset? Not sure how to do this for YUYV. */
-	if (src_x != 0 || src_y != 0)
-		return -ERANGE;
-
-	if (crtc_w < src_w || crtc_h < src_h)
-		return -ERANGE;
+	ret = verify_scaling(fb, 0, src_x, src_y, src_w, src_h, crtc_w, crtc_h);
+	if (ret)
+		return ret;
 
 	ret = nouveau_bo_pin(nv_fb->nvbo, TTM_PL_FLAG_VRAM, false);
 	if (ret)
@@ -389,8 +397,9 @@ nv04_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 
 	for (i = 0; i < 2; i++) {
 		nvif_wr32(dev, NV_PVIDEO_BUFF0_START_ADDRESS + 4 * i,
-			nv_fb->nvbo->bo.offset);
-		nvif_wr32(dev, NV_PVIDEO_BUFF0_PITCH_LENGTH + 4 * i, pitch);
+			  nv_fb->nvbo->bo.offset);
+		nvif_wr32(dev, NV_PVIDEO_BUFF0_PITCH_LENGTH + 4 * i,
+			  fb->pitches[0]);
 		nvif_wr32(dev, NV_PVIDEO_BUFF0_OFFSET + 4 * i, 0);
 	}
 	nvif_wr32(dev, NV_PVIDEO_WINDOW_START, crtc_y << 16 | crtc_x);
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/bios/conn.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/bios/conn.h
index e8e77ee24776..deb477282dde 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/bios/conn.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/bios/conn.h
@@ -18,6 +18,7 @@ enum dcb_connector_type {
 	DCB_CONNECTOR_HDMI_C = 0x63,
 	DCB_CONNECTOR_DMS59_DP0 = 0x64,
 	DCB_CONNECTOR_DMS59_DP1 = 0x65,
+	DCB_CONNECTOR_WFD	= 0x70,
 	DCB_CONNECTOR_NONE = 0xff
 };
 
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/bios/dcb.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/bios/dcb.h
index 4892a65ddd48..903d117603d8 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/bios/dcb.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/bios/dcb.h
@@ -6,6 +6,7 @@ enum dcb_output_type {
 	DCB_OUTPUT_TMDS		= 0x2,
 	DCB_OUTPUT_LVDS		= 0x3,
 	DCB_OUTPUT_DP		= 0x6,
+	DCB_OUTPUT_WFD		= 0x8,
 	DCB_OUTPUT_EOL		= 0xe,
 	DCB_OUTPUT_UNUSED	= 0xf,
 	DCB_OUTPUT_ANY = -1,
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/therm.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/therm.h
index b268b96faece..1bfd93b85575 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/therm.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/therm.h
@@ -96,4 +96,5 @@ int g84_therm_new(struct nvkm_device *, int, struct nvkm_therm **);
 int gt215_therm_new(struct nvkm_device *, int, struct nvkm_therm **);
 int gf119_therm_new(struct nvkm_device *, int, struct nvkm_therm **);
 int gm107_therm_new(struct nvkm_device *, int, struct nvkm_therm **);
+int gm200_therm_new(struct nvkm_device *, int, struct nvkm_therm **);
 #endif
diff --git a/drivers/gpu/drm/nouveau/nouveau_bios.c b/drivers/gpu/drm/nouveau/nouveau_bios.c
index b998c33af18a..dd6fba55ad5d 100644
--- a/drivers/gpu/drm/nouveau/nouveau_bios.c
+++ b/drivers/gpu/drm/nouveau/nouveau_bios.c
@@ -351,11 +351,8 @@ static int parse_fp_mode_table(struct drm_device *dev, struct nvbios *bios)
 	struct lvdstableheader lth;
 
 	if (bios->fp.fptablepointer == 0x0) {
-		/* Apple cards don't have the fp table; the laptops use DDC */
-		/* The table is also missing on some x86 IGPs */
-#ifndef __powerpc__
-		NV_ERROR(drm, "Pointer to flat panel table invalid\n");
-#endif
+		/* Most laptop cards lack an fp table. They use DDC. */
+		NV_DEBUG(drm, "Pointer to flat panel table invalid\n");
 		bios->digital_min_front_porch = 0x4b;
 		return 0;
 	}
diff --git a/drivers/gpu/drm/nouveau/nouveau_connector.c b/drivers/gpu/drm/nouveau/nouveau_connector.c
index 196eb668d30d..70d8e0d69ad5 100644
--- a/drivers/gpu/drm/nouveau/nouveau_connector.c
+++ b/drivers/gpu/drm/nouveau/nouveau_connector.c
@@ -1147,8 +1147,6 @@ nouveau_connector_aux_xfer(struct drm_dp_aux *obj, struct drm_dp_aux_msg *msg)
 		return -ENODEV;
 	if (WARN_ON(msg->size > 16))
 		return -E2BIG;
-	if (msg->size == 0)
-		return msg->size;
 
 	ret = nvkm_i2c_aux_acquire(aux);
 	if (ret)
@@ -1186,6 +1184,7 @@ drm_conntype_from_dcb(enum dcb_connector_type dcb)
 	case DCB_CONNECTOR_HDMI_0   :
 	case DCB_CONNECTOR_HDMI_1   :
 	case DCB_CONNECTOR_HDMI_C   : return DRM_MODE_CONNECTOR_HDMIA;
+	case DCB_CONNECTOR_WFD	    : return DRM_MODE_CONNECTOR_VIRTUAL;
 	default:
 		break;
 	}
diff --git a/drivers/gpu/drm/nouveau/nouveau_display.c b/drivers/gpu/drm/nouveau/nouveau_display.c
index b9a109be989c..2e7785f49e6d 100644
--- a/drivers/gpu/drm/nouveau/nouveau_display.c
+++ b/drivers/gpu/drm/nouveau/nouveau_display.c
@@ -231,9 +231,30 @@ nouveau_framebuffer_new(struct drm_device *dev,
 			struct nouveau_bo *nvbo,
 			struct nouveau_framebuffer **pfb)
 {
+	struct nouveau_drm *drm = nouveau_drm(dev);
 	struct nouveau_framebuffer *fb;
 	int ret;
 
+        /* YUV overlays have special requirements pre-NV50 */
+	if (drm->client.device.info.family < NV_DEVICE_INFO_V0_TESLA &&
+
+	    (mode_cmd->pixel_format == DRM_FORMAT_YUYV ||
+	     mode_cmd->pixel_format == DRM_FORMAT_UYVY ||
+	     mode_cmd->pixel_format == DRM_FORMAT_NV12 ||
+	     mode_cmd->pixel_format == DRM_FORMAT_NV21) &&
+	    (mode_cmd->pitches[0] & 0x3f || /* align 64 */
+	     mode_cmd->pitches[0] >= 0x10000 || /* at most 64k pitch */
+	     (mode_cmd->pitches[1] && /* pitches for planes must match */
+	      mode_cmd->pitches[0] != mode_cmd->pitches[1]))) {
+		struct drm_format_name_buf format_name;
+		DRM_DEBUG_KMS("Unsuitable framebuffer: format: %s; pitches: 0x%x\n 0x%x\n",
+			      drm_get_format_name(mode_cmd->pixel_format,
+						  &format_name),
+			      mode_cmd->pitches[0],
+			      mode_cmd->pitches[1]);
+		return -EINVAL;
+	}
+
 	if (!(fb = *pfb = kzalloc(sizeof(*fb), GFP_KERNEL)))
 		return -ENOMEM;
 
@@ -407,7 +428,6 @@ 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_crtc *crtc;
 
 	if (!suspend) {
 		if (drm_drv_uses_atomic_modeset(dev))
@@ -416,10 +436,6 @@ nouveau_display_fini(struct drm_device *dev, bool suspend)
 			drm_crtc_force_disable_all(dev);
 	}
 
-	/* Make sure that drm and hw vblank irqs get properly disabled. */
-	drm_for_each_crtc(crtc, dev)
-		drm_crtc_vblank_off(crtc);
-
 	/* disable flip completion events */
 	nvif_notify_put(&drm->flip);
 
diff --git a/drivers/gpu/drm/nouveau/nouveau_drm.c b/drivers/gpu/drm/nouveau/nouveau_drm.c
index df7e2037031a..595630d1fb9e 100644
--- a/drivers/gpu/drm/nouveau/nouveau_drm.c
+++ b/drivers/gpu/drm/nouveau/nouveau_drm.c
@@ -585,18 +585,18 @@ nouveau_do_suspend(struct drm_device *dev, bool runtime)
 	nouveau_led_suspend(dev);
 
 	if (dev->mode_config.num_crtc) {
-		NV_INFO(drm, "suspending console...\n");
+		NV_DEBUG(drm, "suspending console...\n");
 		nouveau_fbcon_set_suspend(dev, 1);
-		NV_INFO(drm, "suspending display...\n");
+		NV_DEBUG(drm, "suspending display...\n");
 		ret = nouveau_display_suspend(dev, runtime);
 		if (ret)
 			return ret;
 	}
 
-	NV_INFO(drm, "evicting buffers...\n");
+	NV_DEBUG(drm, "evicting buffers...\n");
 	ttm_bo_evict_mm(&drm->ttm.bdev, TTM_PL_VRAM);
 
-	NV_INFO(drm, "waiting for kernel channels to go idle...\n");
+	NV_DEBUG(drm, "waiting for kernel channels to go idle...\n");
 	if (drm->cechan) {
 		ret = nouveau_channel_idle(drm->cechan);
 		if (ret)
@@ -609,7 +609,7 @@ nouveau_do_suspend(struct drm_device *dev, bool runtime)
 			goto fail_display;
 	}
 
-	NV_INFO(drm, "suspending fence...\n");
+	NV_DEBUG(drm, "suspending fence...\n");
 	if (drm->fence && nouveau_fence(drm)->suspend) {
 		if (!nouveau_fence(drm)->suspend(drm)) {
 			ret = -ENOMEM;
@@ -617,7 +617,7 @@ nouveau_do_suspend(struct drm_device *dev, bool runtime)
 		}
 	}
 
-	NV_INFO(drm, "suspending object tree...\n");
+	NV_DEBUG(drm, "suspending object tree...\n");
 	ret = nvif_client_suspend(&drm->client.base);
 	if (ret)
 		goto fail_client;
@@ -630,7 +630,7 @@ fail_client:
 
 fail_display:
 	if (dev->mode_config.num_crtc) {
-		NV_INFO(drm, "resuming display...\n");
+		NV_DEBUG(drm, "resuming display...\n");
 		nouveau_display_resume(dev, runtime);
 	}
 	return ret;
@@ -641,19 +641,19 @@ nouveau_do_resume(struct drm_device *dev, bool runtime)
 {
 	struct nouveau_drm *drm = nouveau_drm(dev);
 
-	NV_INFO(drm, "resuming object tree...\n");
+	NV_DEBUG(drm, "resuming object tree...\n");
 	nvif_client_resume(&drm->client.base);
 
-	NV_INFO(drm, "resuming fence...\n");
+	NV_DEBUG(drm, "resuming fence...\n");
 	if (drm->fence && nouveau_fence(drm)->resume)
 		nouveau_fence(drm)->resume(drm);
 
 	nouveau_run_vbios_init(dev);
 
 	if (dev->mode_config.num_crtc) {
-		NV_INFO(drm, "resuming display...\n");
+		NV_DEBUG(drm, "resuming display...\n");
 		nouveau_display_resume(dev, runtime);
-		NV_INFO(drm, "resuming console...\n");
+		NV_DEBUG(drm, "resuming console...\n");
 		nouveau_fbcon_set_suspend(dev, 0);
 	}
 
diff --git a/drivers/gpu/drm/nouveau/nouveau_ttm.c b/drivers/gpu/drm/nouveau/nouveau_ttm.c
index 999c35a25498..b0ad7fcefcf5 100644
--- a/drivers/gpu/drm/nouveau/nouveau_ttm.c
+++ b/drivers/gpu/drm/nouveau/nouveau_ttm.c
@@ -179,7 +179,8 @@ nouveau_gart_manager_new(struct ttm_mem_type_manager *man,
 }
 
 static void
-nouveau_gart_manager_debug(struct ttm_mem_type_manager *man, const char *prefix)
+nouveau_gart_manager_debug(struct ttm_mem_type_manager *man,
+			   struct drm_printer *printer)
 {
 }
 
@@ -252,7 +253,8 @@ nv04_gart_manager_new(struct ttm_mem_type_manager *man,
 }
 
 static void
-nv04_gart_manager_debug(struct ttm_mem_type_manager *man, const char *prefix)
+nv04_gart_manager_debug(struct ttm_mem_type_manager *man,
+			struct drm_printer *printer)
 {
 }
 
diff --git a/drivers/gpu/drm/nouveau/nv50_display.c b/drivers/gpu/drm/nouveau/nv50_display.c
index 6dee4071bb3f..2dbf62a2ac41 100644
--- a/drivers/gpu/drm/nouveau/nv50_display.c
+++ b/drivers/gpu/drm/nouveau/nv50_display.c
@@ -3141,7 +3141,7 @@ nv50_mstc_new(struct nv50_mstm *mstm, struct drm_dp_mst_port *port,
 	mstc->connector.funcs->reset(&mstc->connector);
 	nouveau_conn_attach_properties(&mstc->connector);
 
-	for (i = 0; i < ARRAY_SIZE(mstm->msto) && mstm->msto; i++)
+	for (i = 0; i < ARRAY_SIZE(mstm->msto) && mstm->msto[i]; i++)
 		drm_mode_connector_attach_encoder(&mstc->connector, &mstm->msto[i]->encoder);
 
 	drm_object_attach_property(&mstc->connector.base, dev->mode_config.path_property, 0);
@@ -3658,15 +3658,24 @@ nv50_sor_create(struct drm_connector *connector, struct dcb_output *dcbe)
 	drm_mode_connector_attach_encoder(connector, encoder);
 
 	if (dcbe->type == DCB_OUTPUT_DP) {
+		struct nv50_disp *disp = nv50_disp(encoder->dev);
 		struct nvkm_i2c_aux *aux =
 			nvkm_i2c_aux_find(i2c, dcbe->i2c_index);
 		if (aux) {
-			nv_encoder->i2c = &nv_connector->aux.ddc;
+			if (disp->disp->oclass < GF110_DISP) {
+				/* HW has no support for address-only
+				 * transactions, so we're required to
+				 * use custom I2C-over-AUX code.
+				 */
+				nv_encoder->i2c = &aux->i2c;
+			} else {
+				nv_encoder->i2c = &nv_connector->aux.ddc;
+			}
 			nv_encoder->aux = aux;
 		}
 
 		/*TODO: Use DP Info Table to check for support. */
-		if (nv50_disp(encoder->dev)->disp->oclass >= GF110_DISP) {
+		if (disp->disp->oclass >= GF110_DISP) {
 			ret = nv50_mstm_new(nv_encoder, &nv_connector->aux, 16,
 					    nv_connector->base.base.id,
 					    &nv_encoder->dp.mstm);
@@ -3888,7 +3897,7 @@ static void
 nv50_disp_atomic_commit_tail(struct drm_atomic_state *state)
 {
 	struct drm_device *dev = state->dev;
-	struct drm_crtc_state *new_crtc_state;
+	struct drm_crtc_state *new_crtc_state, *old_crtc_state;
 	struct drm_crtc *crtc;
 	struct drm_plane_state *new_plane_state;
 	struct drm_plane *plane;
@@ -3909,12 +3918,14 @@ nv50_disp_atomic_commit_tail(struct drm_atomic_state *state)
 		mutex_lock(&disp->mutex);
 
 	/* Disable head(s). */
-	for_each_new_crtc_in_state(state, crtc, new_crtc_state, i) {
+	for_each_oldnew_crtc_in_state(state, crtc, old_crtc_state, new_crtc_state, i) {
 		struct nv50_head_atom *asyh = nv50_head_atom(new_crtc_state);
 		struct nv50_head *head = nv50_head(crtc);
 
 		NV_ATOMIC(drm, "%s: clr %04x (set %04x)\n", crtc->name,
 			  asyh->clr.mask, asyh->set.mask);
+		if (old_crtc_state->active && !new_crtc_state->active)
+			drm_crtc_vblank_off(crtc);
 
 		if (asyh->clr.mask) {
 			nv50_head_flush_clr(head, asyh, atom->flush_disable);
@@ -3989,7 +4000,7 @@ nv50_disp_atomic_commit_tail(struct drm_atomic_state *state)
 	}
 
 	/* Update head(s). */
-	for_each_new_crtc_in_state(state, crtc, new_crtc_state, i) {
+	for_each_oldnew_crtc_in_state(state, crtc, old_crtc_state, new_crtc_state, i) {
 		struct nv50_head_atom *asyh = nv50_head_atom(new_crtc_state);
 		struct nv50_head *head = nv50_head(crtc);
 
@@ -4000,11 +4011,13 @@ nv50_disp_atomic_commit_tail(struct drm_atomic_state *state)
 			nv50_head_flush_set(head, asyh);
 			interlock_core = 1;
 		}
-	}
 
-	for_each_new_crtc_in_state(state, crtc, new_crtc_state, i) {
-		if (new_crtc_state->event)
-			drm_crtc_vblank_get(crtc);
+		if (new_crtc_state->active) {
+			if (!old_crtc_state->active)
+				drm_crtc_vblank_on(crtc);
+			if (new_crtc_state->event)
+				drm_crtc_vblank_get(crtc);
+		}
 	}
 
 	/* Update plane(s). */
@@ -4051,12 +4064,15 @@ nv50_disp_atomic_commit_tail(struct drm_atomic_state *state)
 		if (new_crtc_state->event) {
 			unsigned long flags;
 			/* Get correct count/ts if racing with vblank irq */
-			drm_crtc_accurate_vblank_count(crtc);
+			if (new_crtc_state->active)
+				drm_crtc_accurate_vblank_count(crtc);
 			spin_lock_irqsave(&crtc->dev->event_lock, flags);
 			drm_crtc_send_vblank_event(crtc, new_crtc_state->event);
 			spin_unlock_irqrestore(&crtc->dev->event_lock, flags);
+
 			new_crtc_state->event = NULL;
-			drm_crtc_vblank_put(crtc);
+			if (new_crtc_state->active)
+				drm_crtc_vblank_put(crtc);
 		}
 	}
 
@@ -4435,11 +4451,13 @@ nv50_display_create(struct drm_device *dev)
 
 	/* create crtc objects to represent the hw heads */
 	if (disp->disp->oclass >= GF110_DISP)
-		crtcs = nvif_rd32(&device->object, 0x022448);
+		crtcs = nvif_rd32(&device->object, 0x612004) & 0xf;
 	else
-		crtcs = 2;
+		crtcs = 0x3;
 
-	for (i = 0; i < crtcs; i++) {
+	for (i = 0; i < fls(crtcs); i++) {
+		if (!(crtcs & (1 << i)))
+			continue;
 		ret = nv50_head_create(dev, i);
 		if (ret)
 			goto out;
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c b/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c
index 7bdc7a5ae723..e096a5d9c292 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c
@@ -2043,6 +2043,7 @@ nv120_chipset = {
 	.mxm = nv50_mxm_new,
 	.pci = gk104_pci_new,
 	.pmu = gm107_pmu_new,
+	.therm = gm200_therm_new,
 	.secboot = gm200_secboot_new,
 	.timer = gk20a_timer_new,
 	.top = gk104_top_new,
@@ -2077,6 +2078,7 @@ nv124_chipset = {
 	.mxm = nv50_mxm_new,
 	.pci = gk104_pci_new,
 	.pmu = gm107_pmu_new,
+	.therm = gm200_therm_new,
 	.secboot = gm200_secboot_new,
 	.timer = gk20a_timer_new,
 	.top = gk104_top_new,
@@ -2111,6 +2113,7 @@ nv126_chipset = {
 	.mxm = nv50_mxm_new,
 	.pci = gk104_pci_new,
 	.pmu = gm107_pmu_new,
+	.therm = gm200_therm_new,
 	.secboot = gm200_secboot_new,
 	.timer = gk20a_timer_new,
 	.top = gk104_top_new,
@@ -2321,6 +2324,35 @@ nv137_chipset = {
 };
 
 static const struct nvkm_device_chip
+nv138_chipset = {
+	.name = "GP108",
+	.bar = gf100_bar_new,
+	.bios = nvkm_bios_new,
+	.bus = gf100_bus_new,
+	.devinit = gm200_devinit_new,
+	.fb = gp102_fb_new,
+	.fuse = gm107_fuse_new,
+	.gpio = gk104_gpio_new,
+	.i2c = gm200_i2c_new,
+	.ibus = gm200_ibus_new,
+	.imem = nv50_instmem_new,
+	.ltc = gp100_ltc_new,
+	.mc = gp100_mc_new,
+	.mmu = gf100_mmu_new,
+	.pci = gp100_pci_new,
+	.pmu = gp102_pmu_new,
+	.timer = gk20a_timer_new,
+	.top = gk104_top_new,
+	.ce[0] = gp102_ce_new,
+	.ce[1] = gp102_ce_new,
+	.ce[2] = gp102_ce_new,
+	.ce[3] = gp102_ce_new,
+	.disp = gp102_disp_new,
+	.dma = gf119_dma_new,
+	.fifo = gp100_fifo_new,
+};
+
+static const struct nvkm_device_chip
 nv13b_chipset = {
 	.name = "GP10B",
 	.bar = gk20a_bar_new,
@@ -2782,6 +2814,7 @@ nvkm_device_ctor(const struct nvkm_device_func *func,
 		case 0x134: device->chip = &nv134_chipset; break;
 		case 0x136: device->chip = &nv136_chipset; break;
 		case 0x137: device->chip = &nv137_chipset; break;
+		case 0x138: device->chip = &nv138_chipset; break;
 		case 0x13b: device->chip = &nv13b_chipset; break;
 		default:
 			nvdev_error(device, "unknown chipset (%08x)\n", boot0);
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/base.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/base.c
index c7c84d34d97e..93a75e5b2791 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/base.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/base.c
@@ -267,6 +267,8 @@ nvkm_disp_oneinit(struct nvkm_engine *engine)
 	/* Create output path objects for each VBIOS display path. */
 	i = -1;
 	while ((data = dcb_outp_parse(bios, ++i, &ver, &hdr, &dcbE))) {
+		if (ver < 0x40) /* No support for chipsets prior to NV50. */
+			break;
 		if (dcbE.type == DCB_OUTPUT_UNUSED)
 			continue;
 		if (dcbE.type == DCB_OUTPUT_EOL)
@@ -283,6 +285,10 @@ nvkm_disp_oneinit(struct nvkm_engine *engine)
 		case DCB_OUTPUT_DP:
 			ret = nvkm_dp_new(disp, i, &dcbE, &outp);
 			break;
+		case DCB_OUTPUT_WFD:
+			/* No support for WFD yet. */
+			ret = -ENODEV;
+			continue;
 		default:
 			nvkm_warn(subdev, "dcb %d type %d unknown\n",
 				  i, dcbE.type);
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/headgf119.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/headgf119.c
index b33552757647..9fd7ae331308 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/headgf119.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/headgf119.c
@@ -92,5 +92,8 @@ gf119_head = {
 int
 gf119_head_new(struct nvkm_disp *disp, int id)
 {
+	struct nvkm_device *device = disp->engine.subdev.device;
+	if (!(nvkm_rd32(device, 0x612004) & (0x00000001 << id)))
+		return 0;
 	return nvkm_head_new_(&gf119_head, disp, id);
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/ior.h b/drivers/gpu/drm/nouveau/nvkm/engine/disp/ior.h
index a24312fb0228..a1e8bf48b778 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/ior.h
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/ior.h
@@ -22,6 +22,7 @@ struct nvkm_ior {
 		unsigned proto_evo:4;
 		enum nvkm_ior_proto {
 			CRT,
+			TV,
 			TMDS,
 			LVDS,
 			DP,
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/nv50.h b/drivers/gpu/drm/nouveau/nvkm/engine/disp/nv50.h
index 19c635663399..6ea19466f436 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/nv50.h
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/nv50.h
@@ -22,7 +22,7 @@ struct nv50_disp {
 		u8 type[3];
 	} pior;
 
-	struct nv50_disp_chan *chan[17];
+	struct nv50_disp_chan *chan[21];
 };
 
 void nv50_disp_super_1(struct nv50_disp *);
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/outp.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/outp.c
index 85aff85394ac..be9e7f8c3b23 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/outp.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/outp.c
@@ -62,6 +62,7 @@ nvkm_outp_xlat(struct nvkm_outp *outp, enum nvkm_ior_type *type)
 	case 0:
 		switch (outp->info.type) {
 		case DCB_OUTPUT_ANALOG: *type = DAC; return  CRT;
+		case DCB_OUTPUT_TV    : *type = DAC; return   TV;
 		case DCB_OUTPUT_TMDS  : *type = SOR; return TMDS;
 		case DCB_OUTPUT_LVDS  : *type = SOR; return LVDS;
 		case DCB_OUTPUT_DP    : *type = SOR; return   DP;
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/mpeg/nv31.c b/drivers/gpu/drm/nouveau/nvkm/engine/mpeg/nv31.c
index 8a8895246d26..7fea7d45202f 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/mpeg/nv31.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/mpeg/nv31.c
@@ -124,6 +124,8 @@ nv31_mpeg_tile(struct nvkm_engine *engine, int i, struct nvkm_fb_tile *tile)
 static bool
 nv31_mpeg_mthd_dma(struct nvkm_device *device, u32 mthd, u32 data)
 {
+	struct nv31_mpeg *mpeg = nv31_mpeg(device->mpeg);
+	struct nvkm_subdev *subdev = &mpeg->engine.subdev;
 	u32 inst = data << 4;
 	u32 dma0 = nvkm_rd32(device, 0x700000 + inst);
 	u32 dma1 = nvkm_rd32(device, 0x700004 + inst);
@@ -132,8 +134,11 @@ nv31_mpeg_mthd_dma(struct nvkm_device *device, u32 mthd, u32 data)
 	u32 size = dma1 + 1;
 
 	/* only allow linear DMA objects */
-	if (!(dma0 & 0x00002000))
+	if (!(dma0 & 0x00002000)) {
+		nvkm_error(subdev, "inst %08x dma0 %08x dma1 %08x dma2 %08x\n",
+			   inst, dma0, dma1, dma2);
 		return false;
+	}
 
 	if (mthd == 0x0190) {
 		/* DMA_CMD */
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/mpeg/nv40.c b/drivers/gpu/drm/nouveau/nvkm/engine/mpeg/nv40.c
index 16de5bd94b14..b5ec7c504dc6 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/mpeg/nv40.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/mpeg/nv40.c
@@ -31,6 +31,8 @@ bool
 nv40_mpeg_mthd_dma(struct nvkm_device *device, u32 mthd, u32 data)
 {
 	struct nvkm_instmem *imem = device->imem;
+	struct nv31_mpeg *mpeg = nv31_mpeg(device->mpeg);
+	struct nvkm_subdev *subdev = &mpeg->engine.subdev;
 	u32 inst = data << 4;
 	u32 dma0 = nvkm_instmem_rd32(imem, inst + 0);
 	u32 dma1 = nvkm_instmem_rd32(imem, inst + 4);
@@ -39,8 +41,11 @@ nv40_mpeg_mthd_dma(struct nvkm_device *device, u32 mthd, u32 data)
 	u32 size = dma1 + 1;
 
 	/* only allow linear DMA objects */
-	if (!(dma0 & 0x00002000))
+	if (!(dma0 & 0x00002000)) {
+		nvkm_error(subdev, "inst %08x dma0 %08x dma1 %08x dma2 %08x\n",
+			   inst, dma0, dma1, dma2);
 		return false;
+	}
 
 	if (mthd == 0x0190) {
 		/* DMA_CMD */
diff --git a/drivers/gpu/drm/nouveau/nvkm/falcon/msgqueue.c b/drivers/gpu/drm/nouveau/nvkm/falcon/msgqueue.c
index d45d7947a964..77273b53672c 100644
--- a/drivers/gpu/drm/nouveau/nvkm/falcon/msgqueue.c
+++ b/drivers/gpu/drm/nouveau/nvkm/falcon/msgqueue.c
@@ -251,7 +251,7 @@ cmd_write(struct nvkm_msgqueue *priv, struct nvkm_msgqueue_hdr *cmd,
 	  struct nvkm_msgqueue_queue *queue)
 {
 	const struct nvkm_subdev *subdev = priv->falcon->owner;
-	static unsigned long timeout = ~0;
+	static unsigned timeout = 2000;
 	unsigned long end_jiffies = jiffies + msecs_to_jiffies(timeout);
 	int ret = -EAGAIN;
 	bool commit = true;
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/bar/gf100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/bar/gf100.c
index c794b2c2d21e..676c167c95b9 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/bar/gf100.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/bar/gf100.c
@@ -24,6 +24,7 @@
 #include "gf100.h"
 
 #include <core/gpuobj.h>
+#include <core/option.h>
 #include <subdev/fb.h>
 #include <subdev/mmu.h>
 
@@ -59,6 +60,8 @@ gf100_bar_ctor_vm(struct gf100_bar *bar, struct gf100_bar_vm *bar_vm,
 		return ret;
 
 	bar_len = device->func->resource_size(device, bar_nr);
+	if (bar_nr == 3 && bar->bar2_halve)
+		bar_len >>= 1;
 
 	ret = nvkm_vm_new(device, 0, bar_len, 0, key, &vm);
 	if (ret)
@@ -129,7 +132,9 @@ gf100_bar_init(struct nvkm_bar *base)
 
 	if (bar->bar[0].mem) {
 		addr = nvkm_memory_addr(bar->bar[0].mem) >> 12;
-		nvkm_wr32(device, 0x001714, 0xc0000000 | addr);
+		if (bar->bar2_halve)
+			addr |= 0x40000000;
+		nvkm_wr32(device, 0x001714, 0x80000000 | addr);
 	}
 
 	return 0;
@@ -161,6 +166,7 @@ gf100_bar_new_(const struct nvkm_bar_func *func, struct nvkm_device *device,
 	if (!(bar = kzalloc(sizeof(*bar), GFP_KERNEL)))
 		return -ENOMEM;
 	nvkm_bar_ctor(func, device, index, &bar->base);
+	bar->bar2_halve = nvkm_boolopt(device->cfgopt, "NvBar2Halve", false);
 	*pbar = &bar->base;
 	return 0;
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/bar/gf100.h b/drivers/gpu/drm/nouveau/nvkm/subdev/bar/gf100.h
index f7dea69640d8..20a5255362ba 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/bar/gf100.h
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/bar/gf100.h
@@ -11,6 +11,7 @@ struct gf100_bar_vm {
 
 struct gf100_bar {
 	struct nvkm_bar base;
+	bool bar2_halve;
 	struct gf100_bar_vm bar[2];
 };
 
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/gf100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/gf100.c
index 3841ad6be99e..a239e73562c8 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/gf100.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/gf100.c
@@ -60,12 +60,12 @@ gf100_fb_oneinit(struct nvkm_fb *base)
 	size = min(size, 0x1000);
 
 	ret = nvkm_memory_new(device, NVKM_MEM_TARGET_INST, size, 0x1000,
-			      false, &fb->base.mmu_rd);
+			      true, &fb->base.mmu_rd);
 	if (ret)
 		return ret;
 
 	ret = nvkm_memory_new(device, NVKM_MEM_TARGET_INST, size, 0x1000,
-			      false, &fb->base.mmu_wr);
+			      true, &fb->base.mmu_wr);
 	if (ret)
 		return ret;
 
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/Kbuild b/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/Kbuild
index 48f01e40b8fc..b768e66a472b 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/Kbuild
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/Kbuild
@@ -25,6 +25,7 @@ nvkm-y += nvkm/subdev/i2c/bit.o
 
 nvkm-y += nvkm/subdev/i2c/aux.o
 nvkm-y += nvkm/subdev/i2c/auxg94.o
+nvkm-y += nvkm/subdev/i2c/auxgf119.o
 nvkm-y += nvkm/subdev/i2c/auxgm200.o
 
 nvkm-y += nvkm/subdev/i2c/anx9805.o
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/aux.c b/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/aux.c
index d172e42dd228..4c1f547da463 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/aux.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/aux.c
@@ -117,6 +117,10 @@ int
 nvkm_i2c_aux_xfer(struct nvkm_i2c_aux *aux, bool retry, u8 type,
 		  u32 addr, u8 *data, u8 *size)
 {
+	if (!*size && !aux->func->address_only) {
+		AUX_ERR(aux, "address-only transaction dropped");
+		return -ENOSYS;
+	}
 	return aux->func->xfer(aux, retry, type, addr, data, size);
 }
 
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/aux.h b/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/aux.h
index 27a4a39c87f0..9587ab456d9e 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/aux.h
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/aux.h
@@ -3,6 +3,7 @@
 #include "pad.h"
 
 struct nvkm_i2c_aux_func {
+	bool address_only;
 	int  (*xfer)(struct nvkm_i2c_aux *, bool retry, u8 type,
 		     u32 addr, u8 *data, u8 *size);
 	int  (*lnk_ctl)(struct nvkm_i2c_aux *, int link_nr, int link_bw,
@@ -17,7 +18,12 @@ void nvkm_i2c_aux_del(struct nvkm_i2c_aux **);
 int nvkm_i2c_aux_xfer(struct nvkm_i2c_aux *, bool retry, u8 type,
 		      u32 addr, u8 *data, u8 *size);
 
+int g94_i2c_aux_new_(const struct nvkm_i2c_aux_func *, struct nvkm_i2c_pad *,
+		     int, u8, struct nvkm_i2c_aux **);
+
 int g94_i2c_aux_new(struct nvkm_i2c_pad *, int, u8, struct nvkm_i2c_aux **);
+int g94_i2c_aux_xfer(struct nvkm_i2c_aux *, bool, u8, u32, u8 *, u8 *);
+int gf119_i2c_aux_new(struct nvkm_i2c_pad *, int, u8, struct nvkm_i2c_aux **);
 int gm200_i2c_aux_new(struct nvkm_i2c_pad *, int, u8, struct nvkm_i2c_aux **);
 
 #define AUX_MSG(b,l,f,a...) do {                                               \
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/auxg94.c b/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/auxg94.c
index ab8cb196c34e..c8ab1b5741a3 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/auxg94.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/auxg94.c
@@ -72,7 +72,7 @@ g94_i2c_aux_init(struct g94_i2c_aux *aux)
 	return 0;
 }
 
-static int
+int
 g94_i2c_aux_xfer(struct nvkm_i2c_aux *obj, bool retry,
 		 u8 type, u32 addr, u8 *data, u8 *size)
 {
@@ -105,9 +105,9 @@ g94_i2c_aux_xfer(struct nvkm_i2c_aux *obj, bool retry,
 	}
 
 	ctrl  = nvkm_rd32(device, 0x00e4e4 + base);
-	ctrl &= ~0x0001f0ff;
+	ctrl &= ~0x0001f1ff;
 	ctrl |= type << 12;
-	ctrl |= *size - 1;
+	ctrl |= (*size ? (*size - 1) : 0x00000100);
 	nvkm_wr32(device, 0x00e4e0 + base, addr);
 
 	/* (maybe) retry transaction a number of times on failure... */
@@ -160,14 +160,10 @@ out:
 	return ret < 0 ? ret : (stat & 0x000f0000) >> 16;
 }
 
-static const struct nvkm_i2c_aux_func
-g94_i2c_aux_func = {
-	.xfer = g94_i2c_aux_xfer,
-};
-
 int
-g94_i2c_aux_new(struct nvkm_i2c_pad *pad, int index, u8 drive,
-		struct nvkm_i2c_aux **paux)
+g94_i2c_aux_new_(const struct nvkm_i2c_aux_func *func,
+		 struct nvkm_i2c_pad *pad, int index, u8 drive,
+		 struct nvkm_i2c_aux **paux)
 {
 	struct g94_i2c_aux *aux;
 
@@ -175,8 +171,20 @@ g94_i2c_aux_new(struct nvkm_i2c_pad *pad, int index, u8 drive,
 		return -ENOMEM;
 	*paux = &aux->base;
 
-	nvkm_i2c_aux_ctor(&g94_i2c_aux_func, pad, index, &aux->base);
+	nvkm_i2c_aux_ctor(func, pad, index, &aux->base);
 	aux->ch = drive;
 	aux->base.intr = 1 << aux->ch;
 	return 0;
 }
+
+static const struct nvkm_i2c_aux_func
+g94_i2c_aux = {
+	.xfer = g94_i2c_aux_xfer,
+};
+
+int
+g94_i2c_aux_new(struct nvkm_i2c_pad *pad, int index, u8 drive,
+		struct nvkm_i2c_aux **paux)
+{
+	return g94_i2c_aux_new_(&g94_i2c_aux, pad, index, drive, paux);
+}
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/auxgf119.c b/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/auxgf119.c
new file mode 100644
index 000000000000..dab40cd8fe3a
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/auxgf119.c
@@ -0,0 +1,35 @@
+/*
+ * Copyright 2017 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ */
+#include "aux.h"
+
+static const struct nvkm_i2c_aux_func
+gf119_i2c_aux = {
+	.address_only = true,
+	.xfer = g94_i2c_aux_xfer,
+};
+
+int
+gf119_i2c_aux_new(struct nvkm_i2c_pad *pad, int index, u8 drive,
+		  struct nvkm_i2c_aux **paux)
+{
+	return g94_i2c_aux_new_(&gf119_i2c_aux, pad, index, drive, paux);
+}
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/auxgm200.c b/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/auxgm200.c
index ee091fa79628..7ef60895f43a 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/auxgm200.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/auxgm200.c
@@ -105,9 +105,9 @@ gm200_i2c_aux_xfer(struct nvkm_i2c_aux *obj, bool retry,
 	}
 
 	ctrl  = nvkm_rd32(device, 0x00d954 + base);
-	ctrl &= ~0x0001f0ff;
+	ctrl &= ~0x0001f1ff;
 	ctrl |= type << 12;
-	ctrl |= *size - 1;
+	ctrl |= (*size ? (*size - 1) : 0x00000100);
 	nvkm_wr32(device, 0x00d950 + base, addr);
 
 	/* (maybe) retry transaction a number of times on failure... */
@@ -162,6 +162,7 @@ out:
 
 static const struct nvkm_i2c_aux_func
 gm200_i2c_aux_func = {
+	.address_only = true,
 	.xfer = gm200_i2c_aux_xfer,
 };
 
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/padgf119.c b/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/padgf119.c
index d53212f1aa52..3bc4d0310076 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/padgf119.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/padgf119.c
@@ -28,7 +28,7 @@
 static const struct nvkm_i2c_pad_func
 gf119_i2c_pad_s_func = {
 	.bus_new_4 = gf119_i2c_bus_new,
-	.aux_new_6 = g94_i2c_aux_new,
+	.aux_new_6 = gf119_i2c_aux_new,
 	.mode = g94_i2c_pad_mode,
 };
 
@@ -41,7 +41,7 @@ gf119_i2c_pad_s_new(struct nvkm_i2c *i2c, int id, struct nvkm_i2c_pad **ppad)
 static const struct nvkm_i2c_pad_func
 gf119_i2c_pad_x_func = {
 	.bus_new_4 = gf119_i2c_bus_new,
-	.aux_new_6 = g94_i2c_aux_new,
+	.aux_new_6 = gf119_i2c_aux_new,
 };
 
 int
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mc/gf100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/mc/gf100.c
index d2c4d6033abb..f93766418056 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/mc/gf100.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mc/gf100.c
@@ -27,6 +27,7 @@ static const struct nvkm_mc_map
 gf100_mc_reset[] = {
 	{ 0x00020000, NVKM_ENGINE_MSPDEC },
 	{ 0x00008000, NVKM_ENGINE_MSVLD },
+	{ 0x00002000, NVKM_SUBDEV_PMU, true },
 	{ 0x00001000, NVKM_ENGINE_GR },
 	{ 0x00000100, NVKM_ENGINE_FIFO },
 	{ 0x00000080, NVKM_ENGINE_CE1 },
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/base.c b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/base.c
index eb9b278198b2..a4cb82495cee 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/base.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/base.c
@@ -192,6 +192,10 @@ nvkm_pci_new_(const struct nvkm_pci_func *func, struct nvkm_device *device,
 		}
 	}
 
+#ifdef __BIG_ENDIAN
+	pci->msi = false;
+#endif
+
 	pci->msi = nvkm_boolopt(device->cfgopt, "NvMSI", pci->msi);
 	if (pci->msi && func->msi_rearm) {
 		pci->msi = pci_enable_msi(pci->pdev) == 0;
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/base.c b/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/base.c
index 3306f9fe7140..ce70a193caa7 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/base.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/base.c
@@ -75,7 +75,7 @@ nvkm_pmu_reset(struct nvkm_pmu *pmu)
 {
 	struct nvkm_device *device = pmu->subdev.device;
 
-	if (!(nvkm_rd32(device, 0x000200) & 0x00002000))
+	if (!pmu->func->enabled(pmu))
 		return 0;
 
 	/* Inhibit interrupts, and wait for idle. */
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/gf100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/gf100.c
index 0e36d4cb7201..0b458656e870 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/gf100.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/gf100.c
@@ -24,13 +24,30 @@
 #include "priv.h"
 #include "fuc/gf100.fuc3.h"
 
+#include <subdev/mc.h>
+
+void
+gf100_pmu_reset(struct nvkm_pmu *pmu)
+{
+	struct nvkm_device *device = pmu->subdev.device;
+	nvkm_mc_disable(device, NVKM_SUBDEV_PMU);
+	nvkm_mc_enable(device, NVKM_SUBDEV_PMU);
+}
+
+bool
+gf100_pmu_enabled(struct nvkm_pmu *pmu)
+{
+	return nvkm_mc_enabled(pmu->subdev.device, NVKM_SUBDEV_PMU);
+}
+
 static const struct nvkm_pmu_func
 gf100_pmu = {
 	.code.data = gf100_pmu_code,
 	.code.size = sizeof(gf100_pmu_code),
 	.data.data = gf100_pmu_data,
 	.data.size = sizeof(gf100_pmu_data),
-	.reset = gt215_pmu_reset,
+	.enabled = gf100_pmu_enabled,
+	.reset = gf100_pmu_reset,
 	.init = gt215_pmu_init,
 	.fini = gt215_pmu_fini,
 	.intr = gt215_pmu_intr,
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/gf119.c b/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/gf119.c
index 0e4ba4248b15..3dfa79d4fb13 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/gf119.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/gf119.c
@@ -30,7 +30,8 @@ gf119_pmu = {
 	.code.size = sizeof(gf119_pmu_code),
 	.data.data = gf119_pmu_data,
 	.data.size = sizeof(gf119_pmu_data),
-	.reset = gt215_pmu_reset,
+	.enabled = gf100_pmu_enabled,
+	.reset = gf100_pmu_reset,
 	.init = gt215_pmu_init,
 	.fini = gt215_pmu_fini,
 	.intr = gt215_pmu_intr,
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/gk104.c b/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/gk104.c
index 2ad858d825ac..8f7ec10fd2a4 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/gk104.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/gk104.c
@@ -109,7 +109,8 @@ gk104_pmu = {
 	.code.size = sizeof(gk104_pmu_code),
 	.data.data = gk104_pmu_data,
 	.data.size = sizeof(gk104_pmu_data),
-	.reset = gt215_pmu_reset,
+	.enabled = gf100_pmu_enabled,
+	.reset = gf100_pmu_reset,
 	.init = gt215_pmu_init,
 	.fini = gt215_pmu_fini,
 	.intr = gt215_pmu_intr,
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/gk110.c b/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/gk110.c
index fc4b8ecfdaeb..345741d55a56 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/gk110.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/gk110.c
@@ -88,7 +88,8 @@ gk110_pmu = {
 	.code.size = sizeof(gk110_pmu_code),
 	.data.data = gk110_pmu_data,
 	.data.size = sizeof(gk110_pmu_data),
-	.reset = gt215_pmu_reset,
+	.enabled = gf100_pmu_enabled,
+	.reset = gf100_pmu_reset,
 	.init = gt215_pmu_init,
 	.fini = gt215_pmu_fini,
 	.intr = gt215_pmu_intr,
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/gk208.c b/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/gk208.c
index e9a91277683a..e4acf7876ea1 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/gk208.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/gk208.c
@@ -30,7 +30,8 @@ gk208_pmu = {
 	.code.size = sizeof(gk208_pmu_code),
 	.data.data = gk208_pmu_data,
 	.data.size = sizeof(gk208_pmu_data),
-	.reset = gt215_pmu_reset,
+	.enabled = gf100_pmu_enabled,
+	.reset = gf100_pmu_reset,
 	.init = gt215_pmu_init,
 	.fini = gt215_pmu_fini,
 	.intr = gt215_pmu_intr,
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/gk20a.c b/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/gk20a.c
index 978aae3c1001..05e81855c367 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/gk20a.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/gk20a.c
@@ -196,9 +196,10 @@ gk20a_dvfs_data= {
 
 static const struct nvkm_pmu_func
 gk20a_pmu = {
+	.enabled = gf100_pmu_enabled,
 	.init = gk20a_pmu_init,
 	.fini = gk20a_pmu_fini,
-	.reset = gt215_pmu_reset,
+	.reset = gf100_pmu_reset,
 };
 
 int
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/gm107.c b/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/gm107.c
index 9a248ed75f09..459df1ef9e70 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/gm107.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/gm107.c
@@ -32,7 +32,8 @@ gm107_pmu = {
 	.code.size = sizeof(gm107_pmu_code),
 	.data.data = gm107_pmu_data,
 	.data.size = sizeof(gm107_pmu_data),
-	.reset = gt215_pmu_reset,
+	.enabled = gf100_pmu_enabled,
+	.reset = gf100_pmu_reset,
 	.init = gt215_pmu_init,
 	.fini = gt215_pmu_fini,
 	.intr = gt215_pmu_intr,
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/gm20b.c b/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/gm20b.c
index 44bef22bce52..31c843145c7a 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/gm20b.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/gm20b.c
@@ -38,6 +38,7 @@ gm20b_pmu_recv(struct nvkm_pmu *pmu)
 
 static const struct nvkm_pmu_func
 gm20b_pmu = {
+	.enabled = gf100_pmu_enabled,
 	.intr = gt215_pmu_intr,
 	.recv = gm20b_pmu_recv,
 };
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/gp100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/gp100.c
index 6c41c20c85a7..e210cd6af816 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/gp100.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/gp100.c
@@ -25,7 +25,8 @@
 
 static const struct nvkm_pmu_func
 gp100_pmu = {
-	.reset = gt215_pmu_reset,
+	.enabled = gf100_pmu_enabled,
+	.reset = gf100_pmu_reset,
 };
 
 int
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/gp102.c b/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/gp102.c
index f017352206c9..98c7a2a8afc4 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/gp102.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/gp102.c
@@ -31,8 +31,15 @@ gp102_pmu_reset(struct nvkm_pmu *pmu)
 	nvkm_mask(device, 0x10a3c0, 0x00000001, 0x00000000);
 }
 
+static bool
+gp102_pmu_enabled(struct nvkm_pmu *pmu)
+{
+	return !(nvkm_rd32(pmu->subdev.device, 0x10a3c0) & 0x00000001);
+}
+
 static const struct nvkm_pmu_func
 gp102_pmu = {
+	.enabled = gp102_pmu_enabled,
 	.reset = gp102_pmu_reset,
 };
 
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/gt215.c b/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/gt215.c
index 90d428b3be97..e04216daea58 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/gt215.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/gt215.c
@@ -180,13 +180,19 @@ gt215_pmu_fini(struct nvkm_pmu *pmu)
 	nvkm_wr32(pmu->subdev.device, 0x10a014, 0x00000060);
 }
 
-void
+static void
 gt215_pmu_reset(struct nvkm_pmu *pmu)
 {
 	struct nvkm_device *device = pmu->subdev.device;
-	nvkm_mask(device, 0x000200, 0x00002000, 0x00000000);
-	nvkm_mask(device, 0x000200, 0x00002000, 0x00002000);
-	nvkm_rd32(device, 0x000200);
+	nvkm_mask(device, 0x022210, 0x00000001, 0x00000000);
+	nvkm_mask(device, 0x022210, 0x00000001, 0x00000001);
+	nvkm_rd32(device, 0x022210);
+}
+
+static bool
+gt215_pmu_enabled(struct nvkm_pmu *pmu)
+{
+	return nvkm_rd32(pmu->subdev.device, 0x022210) & 0x00000001;
 }
 
 int
@@ -241,6 +247,7 @@ gt215_pmu = {
 	.code.size = sizeof(gt215_pmu_code),
 	.data.data = gt215_pmu_data,
 	.data.size = sizeof(gt215_pmu_data),
+	.enabled = gt215_pmu_enabled,
 	.reset = gt215_pmu_reset,
 	.init = gt215_pmu_init,
 	.fini = gt215_pmu_fini,
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/priv.h b/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/priv.h
index 096cba069f72..a4c48a10cd47 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/priv.h
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/priv.h
@@ -20,6 +20,7 @@ struct nvkm_pmu_func {
 		u32  size;
 	} data;
 
+	bool (*enabled)(struct nvkm_pmu *);
 	void (*reset)(struct nvkm_pmu *);
 	int (*init)(struct nvkm_pmu *);
 	void (*fini)(struct nvkm_pmu *);
@@ -30,12 +31,14 @@ struct nvkm_pmu_func {
 	void (*pgob)(struct nvkm_pmu *, bool);
 };
 
-void gt215_pmu_reset(struct nvkm_pmu *);
 int gt215_pmu_init(struct nvkm_pmu *);
 void gt215_pmu_fini(struct nvkm_pmu *);
 void gt215_pmu_intr(struct nvkm_pmu *);
 void gt215_pmu_recv(struct nvkm_pmu *);
 int gt215_pmu_send(struct nvkm_pmu *, u32[2], u32, u32, u32, u32);
 
+bool gf100_pmu_enabled(struct nvkm_pmu *);
+void gf100_pmu_reset(struct nvkm_pmu *);
+
 void gk110_pmu_pgob(struct nvkm_pmu *, bool);
 #endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/therm/Kbuild b/drivers/gpu/drm/nouveau/nvkm/subdev/therm/Kbuild
index 135758ba3e28..2bafcc1d1818 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/therm/Kbuild
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/therm/Kbuild
@@ -11,3 +11,4 @@ nvkm-y += nvkm/subdev/therm/g84.o
 nvkm-y += nvkm/subdev/therm/gt215.o
 nvkm-y += nvkm/subdev/therm/gf119.o
 nvkm-y += nvkm/subdev/therm/gm107.o
+nvkm-y += nvkm/subdev/therm/gm200.o
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/therm/g84.c b/drivers/gpu/drm/nouveau/nvkm/subdev/therm/g84.c
index 86e81930d8ee..96f8da40ac82 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/therm/g84.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/therm/g84.c
@@ -203,7 +203,7 @@ g84_therm_fini(struct nvkm_therm *therm)
 	nvkm_wr32(device, 0x1100, 0x10000); /* PBUS */
 }
 
-static void
+void
 g84_therm_init(struct nvkm_therm *therm)
 {
 	g84_sensor_setup(therm);
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/therm/gm200.c b/drivers/gpu/drm/nouveau/nvkm/subdev/therm/gm200.c
new file mode 100644
index 000000000000..73dc78093d5d
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/therm/gm200.c
@@ -0,0 +1,39 @@
+/*
+ * Copyright 2017 Karol Herbst
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Karol Herbst
+ */
+#include "priv.h"
+
+static const struct nvkm_therm_func
+gm200_therm = {
+	.init = g84_therm_init,
+	.fini = g84_therm_fini,
+	.temp_get = g84_temp_get,
+	.program_alarms = nvkm_therm_program_alarms_polling,
+};
+
+int
+gm200_therm_new(struct nvkm_device *device, int index,
+		struct nvkm_therm **ptherm)
+{
+	return nvkm_therm_new_(&gm200_therm, device, index, ptherm);
+}
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/therm/priv.h b/drivers/gpu/drm/nouveau/nvkm/subdev/therm/priv.h
index 235a5d8daff6..1f46e371d7c4 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/therm/priv.h
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/therm/priv.h
@@ -111,6 +111,7 @@ void g84_therm_fini(struct nvkm_therm *);
 
 int gt215_therm_fan_sense(struct nvkm_therm *);
 
+void g84_therm_init(struct nvkm_therm *);
 void gf119_therm_init(struct nvkm_therm *);
 
 int nvkm_fanpwm_create(struct nvkm_therm *, struct dcb_gpio_func *);
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/therm/temp.c b/drivers/gpu/drm/nouveau/nvkm/subdev/therm/temp.c
index e93b2410c38b..ddb2b2c600ca 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/therm/temp.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/therm/temp.c
@@ -83,7 +83,7 @@ nvkm_therm_sensor_event(struct nvkm_therm *therm, enum nvkm_therm_thrs thrs,
 {
 	struct nvkm_subdev *subdev = &therm->subdev;
 	bool active;
-	const char *thresolds[] = {
+	static const char * const thresholds[] = {
 		"fanboost", "downclock", "critical", "shutdown"
 	};
 	int temperature = therm->func->temp_get(therm);
@@ -94,10 +94,10 @@ nvkm_therm_sensor_event(struct nvkm_therm *therm, enum nvkm_therm_thrs thrs,
 	if (dir == NVKM_THERM_THRS_FALLING)
 		nvkm_info(subdev,
 			  "temperature (%i C) went below the '%s' threshold\n",
-			  temperature, thresolds[thrs]);
+			  temperature, thresholds[thrs]);
 	else
 		nvkm_info(subdev, "temperature (%i C) hit the '%s' threshold\n",
-			  temperature, thresolds[thrs]);
+			  temperature, thresholds[thrs]);
 
 	active = (dir == NVKM_THERM_THRS_RISING);
 	switch (thrs) {
diff --git a/drivers/gpu/drm/omapdrm/displays/connector-analog-tv.c b/drivers/gpu/drm/omapdrm/displays/connector-analog-tv.c
index e1fa143a5625..542a76503fbd 100644
--- a/drivers/gpu/drm/omapdrm/displays/connector-analog-tv.c
+++ b/drivers/gpu/drm/omapdrm/displays/connector-analog-tv.c
@@ -198,6 +198,9 @@ static int tvc_probe(struct platform_device *pdev)
 	struct omap_dss_device *dssdev;
 	int r;
 
+	if (!pdev->dev.of_node)
+		return -ENODEV;
+
 	ddata = devm_kzalloc(&pdev->dev, sizeof(*ddata), GFP_KERNEL);
 	if (!ddata)
 		return -ENOMEM;
diff --git a/drivers/gpu/drm/omapdrm/displays/connector-hdmi.c b/drivers/gpu/drm/omapdrm/displays/connector-hdmi.c
index 79cb69f1acf5..d9d25df6fc1b 100644
--- a/drivers/gpu/drm/omapdrm/displays/connector-hdmi.c
+++ b/drivers/gpu/drm/omapdrm/displays/connector-hdmi.c
@@ -15,6 +15,7 @@
 #include <linux/platform_device.h>
 #include <linux/of.h>
 #include <linux/of_gpio.h>
+#include <linux/mutex.h>
 
 #include <drm/drm_edid.h>
 
@@ -37,6 +38,10 @@ static const struct videomode hdmic_default_vm = {
 struct panel_drv_data {
 	struct omap_dss_device dssdev;
 	struct omap_dss_device *in;
+	void (*hpd_cb)(void *cb_data, enum drm_connector_status status);
+	void *hpd_cb_data;
+	bool hpd_enabled;
+	struct mutex hpd_lock;
 
 	struct device *dev;
 
@@ -167,6 +172,70 @@ static bool hdmic_detect(struct omap_dss_device *dssdev)
 		return in->ops.hdmi->detect(in);
 }
 
+static int hdmic_register_hpd_cb(struct omap_dss_device *dssdev,
+				 void (*cb)(void *cb_data,
+					    enum drm_connector_status status),
+				 void *cb_data)
+{
+	struct panel_drv_data *ddata = to_panel_data(dssdev);
+	struct omap_dss_device *in = ddata->in;
+
+	if (gpio_is_valid(ddata->hpd_gpio)) {
+		mutex_lock(&ddata->hpd_lock);
+		ddata->hpd_cb = cb;
+		ddata->hpd_cb_data = cb_data;
+		mutex_unlock(&ddata->hpd_lock);
+		return 0;
+	} else if (in->ops.hdmi->register_hpd_cb) {
+		return in->ops.hdmi->register_hpd_cb(in, cb, cb_data);
+	}
+
+	return -ENOTSUPP;
+}
+
+static void hdmic_unregister_hpd_cb(struct omap_dss_device *dssdev)
+{
+	struct panel_drv_data *ddata = to_panel_data(dssdev);
+	struct omap_dss_device *in = ddata->in;
+
+	if (gpio_is_valid(ddata->hpd_gpio)) {
+		mutex_lock(&ddata->hpd_lock);
+		ddata->hpd_cb = NULL;
+		ddata->hpd_cb_data = NULL;
+		mutex_unlock(&ddata->hpd_lock);
+	} else if (in->ops.hdmi->unregister_hpd_cb) {
+		in->ops.hdmi->unregister_hpd_cb(in);
+	}
+}
+
+static void hdmic_enable_hpd(struct omap_dss_device *dssdev)
+{
+	struct panel_drv_data *ddata = to_panel_data(dssdev);
+	struct omap_dss_device *in = ddata->in;
+
+	if (gpio_is_valid(ddata->hpd_gpio)) {
+		mutex_lock(&ddata->hpd_lock);
+		ddata->hpd_enabled = true;
+		mutex_unlock(&ddata->hpd_lock);
+	} else if (in->ops.hdmi->enable_hpd) {
+		in->ops.hdmi->enable_hpd(in);
+	}
+}
+
+static void hdmic_disable_hpd(struct omap_dss_device *dssdev)
+{
+	struct panel_drv_data *ddata = to_panel_data(dssdev);
+	struct omap_dss_device *in = ddata->in;
+
+	if (gpio_is_valid(ddata->hpd_gpio)) {
+		mutex_lock(&ddata->hpd_lock);
+		ddata->hpd_enabled = false;
+		mutex_unlock(&ddata->hpd_lock);
+	} else if (in->ops.hdmi->disable_hpd) {
+		in->ops.hdmi->disable_hpd(in);
+	}
+}
+
 static int hdmic_set_hdmi_mode(struct omap_dss_device *dssdev, bool hdmi_mode)
 {
 	struct panel_drv_data *ddata = to_panel_data(dssdev);
@@ -197,10 +266,34 @@ static struct omap_dss_driver hdmic_driver = {
 
 	.read_edid		= hdmic_read_edid,
 	.detect			= hdmic_detect,
+	.register_hpd_cb	= hdmic_register_hpd_cb,
+	.unregister_hpd_cb	= hdmic_unregister_hpd_cb,
+	.enable_hpd		= hdmic_enable_hpd,
+	.disable_hpd		= hdmic_disable_hpd,
 	.set_hdmi_mode		= hdmic_set_hdmi_mode,
 	.set_hdmi_infoframe	= hdmic_set_infoframe,
 };
 
+static irqreturn_t hdmic_hpd_isr(int irq, void *data)
+{
+	struct panel_drv_data *ddata = data;
+
+	mutex_lock(&ddata->hpd_lock);
+	if (ddata->hpd_enabled && ddata->hpd_cb) {
+		enum drm_connector_status status;
+
+		if (hdmic_detect(&ddata->dssdev))
+			status = connector_status_connected;
+		else
+			status = connector_status_disconnected;
+
+		ddata->hpd_cb(ddata->hpd_cb_data, status);
+	}
+	mutex_unlock(&ddata->hpd_lock);
+
+	return IRQ_HANDLED;
+}
+
 static int hdmic_probe_of(struct platform_device *pdev)
 {
 	struct panel_drv_data *ddata = platform_get_drvdata(pdev);
@@ -246,11 +339,22 @@ static int hdmic_probe(struct platform_device *pdev)
 	if (r)
 		return r;
 
+	mutex_init(&ddata->hpd_lock);
+
 	if (gpio_is_valid(ddata->hpd_gpio)) {
 		r = devm_gpio_request_one(&pdev->dev, ddata->hpd_gpio,
 				GPIOF_DIR_IN, "hdmi_hpd");
 		if (r)
 			goto err_reg;
+
+		r = devm_request_threaded_irq(&pdev->dev,
+				gpio_to_irq(ddata->hpd_gpio),
+				NULL, hdmic_hpd_isr,
+				IRQF_TRIGGER_RISING | IRQF_TRIGGER_FALLING |
+				IRQF_ONESHOT,
+				"hdmic hpd", ddata);
+		if (r)
+			goto err_reg;
 	}
 
 	ddata->vm = hdmic_default_vm;
diff --git a/drivers/gpu/drm/omapdrm/displays/encoder-tpd12s015.c b/drivers/gpu/drm/omapdrm/displays/encoder-tpd12s015.c
index 58276a48112e..a9e9d667c55e 100644
--- a/drivers/gpu/drm/omapdrm/displays/encoder-tpd12s015.c
+++ b/drivers/gpu/drm/omapdrm/displays/encoder-tpd12s015.c
@@ -15,12 +15,17 @@
 #include <linux/slab.h>
 #include <linux/platform_device.h>
 #include <linux/gpio/consumer.h>
+#include <linux/mutex.h>
 
 #include "../dss/omapdss.h"
 
 struct panel_drv_data {
 	struct omap_dss_device dssdev;
 	struct omap_dss_device *in;
+	void (*hpd_cb)(void *cb_data, enum drm_connector_status status);
+	void *hpd_cb_data;
+	bool hpd_enabled;
+	struct mutex hpd_lock;
 
 	struct gpio_desc *ct_cp_hpd_gpio;
 	struct gpio_desc *ls_oe_gpio;
@@ -162,6 +167,49 @@ static bool tpd_detect(struct omap_dss_device *dssdev)
 	return gpiod_get_value_cansleep(ddata->hpd_gpio);
 }
 
+static int tpd_register_hpd_cb(struct omap_dss_device *dssdev,
+			       void (*cb)(void *cb_data,
+					  enum drm_connector_status status),
+			       void *cb_data)
+{
+	struct panel_drv_data *ddata = to_panel_data(dssdev);
+
+	mutex_lock(&ddata->hpd_lock);
+	ddata->hpd_cb = cb;
+	ddata->hpd_cb_data = cb_data;
+	mutex_unlock(&ddata->hpd_lock);
+
+	return 0;
+}
+
+static void tpd_unregister_hpd_cb(struct omap_dss_device *dssdev)
+{
+	struct panel_drv_data *ddata = to_panel_data(dssdev);
+
+	mutex_lock(&ddata->hpd_lock);
+	ddata->hpd_cb = NULL;
+	ddata->hpd_cb_data = NULL;
+	mutex_unlock(&ddata->hpd_lock);
+}
+
+static void tpd_enable_hpd(struct omap_dss_device *dssdev)
+{
+	struct panel_drv_data *ddata = to_panel_data(dssdev);
+
+	mutex_lock(&ddata->hpd_lock);
+	ddata->hpd_enabled = true;
+	mutex_unlock(&ddata->hpd_lock);
+}
+
+static void tpd_disable_hpd(struct omap_dss_device *dssdev)
+{
+	struct panel_drv_data *ddata = to_panel_data(dssdev);
+
+	mutex_lock(&ddata->hpd_lock);
+	ddata->hpd_enabled = false;
+	mutex_unlock(&ddata->hpd_lock);
+}
+
 static int tpd_set_infoframe(struct omap_dss_device *dssdev,
 		const struct hdmi_avi_infoframe *avi)
 {
@@ -193,10 +241,34 @@ static const struct omapdss_hdmi_ops tpd_hdmi_ops = {
 
 	.read_edid		= tpd_read_edid,
 	.detect			= tpd_detect,
+	.register_hpd_cb	= tpd_register_hpd_cb,
+	.unregister_hpd_cb	= tpd_unregister_hpd_cb,
+	.enable_hpd		= tpd_enable_hpd,
+	.disable_hpd		= tpd_disable_hpd,
 	.set_infoframe		= tpd_set_infoframe,
 	.set_hdmi_mode		= tpd_set_hdmi_mode,
 };
 
+static irqreturn_t tpd_hpd_isr(int irq, void *data)
+{
+	struct panel_drv_data *ddata = data;
+
+	mutex_lock(&ddata->hpd_lock);
+	if (ddata->hpd_enabled && ddata->hpd_cb) {
+		enum drm_connector_status status;
+
+		if (tpd_detect(&ddata->dssdev))
+			status = connector_status_connected;
+		else
+			status = connector_status_disconnected;
+
+		ddata->hpd_cb(ddata->hpd_cb_data, status);
+	}
+	mutex_unlock(&ddata->hpd_lock);
+
+	return IRQ_HANDLED;
+}
+
 static int tpd_probe_of(struct platform_device *pdev)
 {
 	struct panel_drv_data *ddata = platform_get_drvdata(pdev);
@@ -261,6 +333,15 @@ static int tpd_probe(struct platform_device *pdev)
 
 	ddata->hpd_gpio = gpio;
 
+	mutex_init(&ddata->hpd_lock);
+
+	r = devm_request_threaded_irq(&pdev->dev, gpiod_to_irq(ddata->hpd_gpio),
+		NULL, tpd_hpd_isr,
+		IRQF_TRIGGER_RISING | IRQF_TRIGGER_FALLING | IRQF_ONESHOT,
+		"tpd12s015 hpd", ddata);
+	if (r)
+		goto err_gpio;
+
 	dssdev = &ddata->dssdev;
 	dssdev->ops.hdmi = &tpd_hdmi_ops;
 	dssdev->dev = &pdev->dev;
diff --git a/drivers/gpu/drm/omapdrm/displays/panel-dpi.c b/drivers/gpu/drm/omapdrm/displays/panel-dpi.c
index 6468a765f3d1..e065f7e10cca 100644
--- a/drivers/gpu/drm/omapdrm/displays/panel-dpi.c
+++ b/drivers/gpu/drm/omapdrm/displays/panel-dpi.c
@@ -231,6 +231,9 @@ static int panel_dpi_probe(struct platform_device *pdev)
 	struct omap_dss_device *dssdev;
 	int r;
 
+	if (!pdev->dev.of_node)
+		return -ENODEV;
+
 	ddata = devm_kzalloc(&pdev->dev, sizeof(*ddata), GFP_KERNEL);
 	if (ddata == NULL)
 		return -ENOMEM;
diff --git a/drivers/gpu/drm/omapdrm/displays/panel-dsi-cm.c b/drivers/gpu/drm/omapdrm/displays/panel-dsi-cm.c
index 76787a75a4dc..92c556ac22c7 100644
--- a/drivers/gpu/drm/omapdrm/displays/panel-dsi-cm.c
+++ b/drivers/gpu/drm/omapdrm/displays/panel-dsi-cm.c
@@ -554,7 +554,7 @@ static struct attribute *dsicm_attrs[] = {
 	NULL,
 };
 
-static struct attribute_group dsicm_attr_group = {
+static const struct attribute_group dsicm_attr_group = {
 	.attrs = dsicm_attrs,
 };
 
diff --git a/drivers/gpu/drm/omapdrm/displays/panel-lgphilips-lb035q02.c b/drivers/gpu/drm/omapdrm/displays/panel-lgphilips-lb035q02.c
index c90474afaebd..74d13969b9ca 100644
--- a/drivers/gpu/drm/omapdrm/displays/panel-lgphilips-lb035q02.c
+++ b/drivers/gpu/drm/omapdrm/displays/panel-lgphilips-lb035q02.c
@@ -19,7 +19,7 @@
 
 #include "../dss/omapdss.h"
 
-static struct videomode lb035q02_vm = {
+static const struct videomode lb035q02_vm = {
 	.hactive = 320,
 	.vactive = 240,
 
diff --git a/drivers/gpu/drm/omapdrm/displays/panel-sony-acx565akm.c b/drivers/gpu/drm/omapdrm/displays/panel-sony-acx565akm.c
index 346aefdb015f..8e5bff4e5226 100644
--- a/drivers/gpu/drm/omapdrm/displays/panel-sony-acx565akm.c
+++ b/drivers/gpu/drm/omapdrm/displays/panel-sony-acx565akm.c
@@ -503,7 +503,7 @@ static struct attribute *bldev_attrs[] = {
 	NULL,
 };
 
-static struct attribute_group bldev_attr_group = {
+static const struct attribute_group bldev_attr_group = {
 	.attrs = bldev_attrs,
 };
 
@@ -720,6 +720,9 @@ static int acx565akm_probe(struct spi_device *spi)
 
 	dev_dbg(&spi->dev, "%s\n", __func__);
 
+	if (!spi->dev.of_node)
+		return -ENODEV;
+
 	spi->mode = SPI_MODE_3;
 
 	ddata = devm_kzalloc(&spi->dev, sizeof(*ddata), GFP_KERNEL);
diff --git a/drivers/gpu/drm/omapdrm/displays/panel-tpo-td028ttec1.c b/drivers/gpu/drm/omapdrm/displays/panel-tpo-td028ttec1.c
index cbf4c67c4933..0a38a0e8c925 100644
--- a/drivers/gpu/drm/omapdrm/displays/panel-tpo-td028ttec1.c
+++ b/drivers/gpu/drm/omapdrm/displays/panel-tpo-td028ttec1.c
@@ -40,7 +40,7 @@ struct panel_drv_data {
 	struct spi_device *spi_dev;
 };
 
-static struct videomode td028ttec1_panel_vm = {
+static const struct videomode td028ttec1_panel_vm = {
 	.hactive	= 480,
 	.vactive	= 640,
 	.pixelclock	= 22153000,
diff --git a/drivers/gpu/drm/omapdrm/displays/panel-tpo-td043mtea1.c b/drivers/gpu/drm/omapdrm/displays/panel-tpo-td043mtea1.c
index 20c6d8fe215a..ac4a6d4d134c 100644
--- a/drivers/gpu/drm/omapdrm/displays/panel-tpo-td043mtea1.c
+++ b/drivers/gpu/drm/omapdrm/displays/panel-tpo-td043mtea1.c
@@ -282,7 +282,7 @@ static struct attribute *tpo_td043_attrs[] = {
 	NULL,
 };
 
-static struct attribute_group tpo_td043_attr_group = {
+static const struct attribute_group tpo_td043_attr_group = {
 	.attrs = tpo_td043_attrs,
 };
 
diff --git a/drivers/gpu/drm/omapdrm/dss/Makefile b/drivers/gpu/drm/omapdrm/dss/Makefile
index 688195e448c5..142ce5a02542 100644
--- a/drivers/gpu/drm/omapdrm/dss/Makefile
+++ b/drivers/gpu/drm/omapdrm/dss/Makefile
@@ -5,7 +5,7 @@ omapdss-base-y := base.o display.o dss-of.o output.o
 
 obj-$(CONFIG_OMAP2_DSS) += omapdss.o
 # Core DSS files
-omapdss-y := core.o dss.o dss_features.o dispc.o dispc_coefs.o \
+omapdss-y := core.o dss.o dispc.o dispc_coefs.o \
 	pll.o video-pll.o
 omapdss-$(CONFIG_OMAP2_DSS_DPI) += dpi.o
 omapdss-$(CONFIG_OMAP2_DSS_VENC) += venc.o
diff --git a/drivers/gpu/drm/omapdrm/dss/core.c b/drivers/gpu/drm/omapdrm/dss/core.c
index bdce4bfdf6e0..197ddbc1512b 100644
--- a/drivers/gpu/drm/omapdrm/dss/core.c
+++ b/drivers/gpu/drm/omapdrm/dss/core.c
@@ -24,182 +24,10 @@
 
 #include <linux/kernel.h>
 #include <linux/module.h>
-#include <linux/clk.h>
-#include <linux/err.h>
 #include <linux/platform_device.h>
-#include <linux/seq_file.h>
-#include <linux/debugfs.h>
-#include <linux/io.h>
-#include <linux/device.h>
-#include <linux/regulator/consumer.h>
-#include <linux/suspend.h>
-#include <linux/slab.h>
 
 #include "omapdss.h"
 #include "dss.h"
-#include "dss_features.h"
-
-static struct {
-	struct platform_device *pdev;
-} core;
-
-enum omapdss_version omapdss_get_version(void)
-{
-	struct omap_dss_board_info *pdata = core.pdev->dev.platform_data;
-	return pdata->version;
-}
-EXPORT_SYMBOL(omapdss_get_version);
-
-int dss_dsi_enable_pads(int dsi_id, unsigned lane_mask)
-{
-	struct omap_dss_board_info *board_data = core.pdev->dev.platform_data;
-
-	if (!board_data->dsi_enable_pads)
-		return -ENOENT;
-
-	return board_data->dsi_enable_pads(dsi_id, lane_mask);
-}
-
-void dss_dsi_disable_pads(int dsi_id, unsigned lane_mask)
-{
-	struct omap_dss_board_info *board_data = core.pdev->dev.platform_data;
-
-	if (!board_data->dsi_disable_pads)
-		return;
-
-	return board_data->dsi_disable_pads(dsi_id, lane_mask);
-}
-
-int dss_set_min_bus_tput(struct device *dev, unsigned long tput)
-{
-	struct omap_dss_board_info *pdata = core.pdev->dev.platform_data;
-
-	if (pdata->set_min_bus_tput)
-		return pdata->set_min_bus_tput(dev, tput);
-	else
-		return 0;
-}
-
-#if defined(CONFIG_OMAP2_DSS_DEBUGFS)
-static int dss_debug_show(struct seq_file *s, void *unused)
-{
-	void (*func)(struct seq_file *) = s->private;
-	func(s);
-	return 0;
-}
-
-static int dss_debug_open(struct inode *inode, struct file *file)
-{
-	return single_open(file, dss_debug_show, inode->i_private);
-}
-
-static const struct file_operations dss_debug_fops = {
-	.open           = dss_debug_open,
-	.read           = seq_read,
-	.llseek         = seq_lseek,
-	.release        = single_release,
-};
-
-static struct dentry *dss_debugfs_dir;
-
-static int dss_initialize_debugfs(void)
-{
-	dss_debugfs_dir = debugfs_create_dir("omapdss", NULL);
-	if (IS_ERR(dss_debugfs_dir)) {
-		int err = PTR_ERR(dss_debugfs_dir);
-		dss_debugfs_dir = NULL;
-		return err;
-	}
-
-	debugfs_create_file("clk", S_IRUGO, dss_debugfs_dir,
-			&dss_debug_dump_clocks, &dss_debug_fops);
-
-	return 0;
-}
-
-static void dss_uninitialize_debugfs(void)
-{
-	if (dss_debugfs_dir)
-		debugfs_remove_recursive(dss_debugfs_dir);
-}
-
-int dss_debugfs_create_file(const char *name, void (*write)(struct seq_file *))
-{
-	struct dentry *d;
-
-	d = debugfs_create_file(name, S_IRUGO, dss_debugfs_dir,
-			write, &dss_debug_fops);
-
-	return PTR_ERR_OR_ZERO(d);
-}
-#else /* CONFIG_OMAP2_DSS_DEBUGFS */
-static inline int dss_initialize_debugfs(void)
-{
-	return 0;
-}
-static inline void dss_uninitialize_debugfs(void)
-{
-}
-int dss_debugfs_create_file(const char *name, void (*write)(struct seq_file *))
-{
-	return 0;
-}
-#endif /* CONFIG_OMAP2_DSS_DEBUGFS */
-
-/* PLATFORM DEVICE */
-
-static void dss_disable_all_devices(void)
-{
-	struct omap_dss_device *dssdev = NULL;
-
-	for_each_dss_dev(dssdev) {
-		if (!dssdev->driver)
-			continue;
-
-		if (dssdev->state == OMAP_DSS_DISPLAY_ACTIVE)
-			dssdev->driver->disable(dssdev);
-	}
-}
-
-static int __init omap_dss_probe(struct platform_device *pdev)
-{
-	int r;
-
-	core.pdev = pdev;
-
-	dss_features_init(omapdss_get_version());
-
-	r = dss_initialize_debugfs();
-	if (r)
-		goto err_debugfs;
-
-	return 0;
-
-err_debugfs:
-
-	return r;
-}
-
-static int omap_dss_remove(struct platform_device *pdev)
-{
-	dss_uninitialize_debugfs();
-
-	return 0;
-}
-
-static void omap_dss_shutdown(struct platform_device *pdev)
-{
-	DSSDBG("shutdown\n");
-	dss_disable_all_devices();
-}
-
-static struct platform_driver omap_dss_driver = {
-	.remove         = omap_dss_remove,
-	.shutdown	= omap_dss_shutdown,
-	.driver         = {
-		.name   = "omapdss",
-	},
-};
 
 /* INIT */
 static int (*dss_output_drv_reg_funcs[])(void) __initdata = {
@@ -236,21 +64,25 @@ static void (*dss_output_drv_unreg_funcs[])(void) = {
 	dss_uninit_platform_driver,
 };
 
+static struct platform_device *omap_drm_device;
+
 static int __init omap_dss_init(void)
 {
 	int r;
 	int i;
 
-	r = platform_driver_probe(&omap_dss_driver, omap_dss_probe);
-	if (r)
-		return r;
-
 	for (i = 0; i < ARRAY_SIZE(dss_output_drv_reg_funcs); ++i) {
 		r = dss_output_drv_reg_funcs[i]();
 		if (r)
 			goto err_reg;
 	}
 
+	omap_drm_device = platform_device_register_simple("omapdrm", 0, NULL, 0);
+	if (IS_ERR(omap_drm_device)) {
+		r = PTR_ERR(omap_drm_device);
+		goto err_reg;
+	}
+
 	return 0;
 
 err_reg:
@@ -259,8 +91,6 @@ err_reg:
 			++i)
 		dss_output_drv_unreg_funcs[i]();
 
-	platform_driver_unregister(&omap_dss_driver);
-
 	return r;
 }
 
@@ -268,10 +98,10 @@ static void __exit omap_dss_exit(void)
 {
 	int i;
 
+	platform_device_unregister(omap_drm_device);
+
 	for (i = 0; i < ARRAY_SIZE(dss_output_drv_unreg_funcs); ++i)
 		dss_output_drv_unreg_funcs[i]();
-
-	platform_driver_unregister(&omap_dss_driver);
 }
 
 module_init(omap_dss_init);
diff --git a/drivers/gpu/drm/omapdrm/dss/dispc.c b/drivers/gpu/drm/omapdrm/dss/dispc.c
index fd7504b37e3b..0f4fdb221498 100644
--- a/drivers/gpu/drm/omapdrm/dss/dispc.c
+++ b/drivers/gpu/drm/omapdrm/dss/dispc.c
@@ -39,13 +39,14 @@
 #include <linux/mfd/syscon.h>
 #include <linux/regmap.h>
 #include <linux/of.h>
+#include <linux/of_device.h>
 #include <linux/component.h>
+#include <linux/sys_soc.h>
 #include <drm/drm_fourcc.h>
 #include <drm/drm_blend.h>
 
 #include "omapdss.h"
 #include "dss.h"
-#include "dss_features.h"
 #include "dispc.h"
 
 /* DISPC */
@@ -63,6 +64,33 @@ enum omap_burst_size {
 #define REG_FLD_MOD(idx, val, start, end)				\
 	dispc_write_reg(idx, FLD_MOD(dispc_read_reg(idx), val, start, end))
 
+/* DISPC has feature id */
+enum dispc_feature_id {
+	FEAT_LCDENABLEPOL,
+	FEAT_LCDENABLESIGNAL,
+	FEAT_PCKFREEENABLE,
+	FEAT_FUNCGATED,
+	FEAT_MGR_LCD2,
+	FEAT_MGR_LCD3,
+	FEAT_LINEBUFFERSPLIT,
+	FEAT_ROWREPEATENABLE,
+	FEAT_RESIZECONF,
+	/* Independent core clk divider */
+	FEAT_CORE_CLK_DIV,
+	FEAT_HANDLE_UV_SEPARATE,
+	FEAT_ATTR2,
+	FEAT_CPR,
+	FEAT_PRELOAD,
+	FEAT_FIR_COEF_V,
+	FEAT_ALPHA_FIXED_ZORDER,
+	FEAT_ALPHA_FREE_ZORDER,
+	FEAT_FIFO_MERGE,
+	/* An unknown HW bug causing the normal FIFO thresholds not to work */
+	FEAT_OMAP3_DSI_FIFO_BUG,
+	FEAT_BURST_2D,
+	FEAT_MFLAG,
+};
+
 struct dispc_features {
 	u8 sw_start;
 	u8 fp_start;
@@ -76,6 +104,9 @@ struct dispc_features {
 	u16 mgr_height_max;
 	unsigned long max_lcd_pclk;
 	unsigned long max_tv_pclk;
+	unsigned int max_downscale;
+	unsigned int max_line_width;
+	unsigned int min_pcd;
 	int (*calc_scaling) (unsigned long pclk, unsigned long lclk,
 		const struct videomode *vm,
 		u16 width, u16 height, u16 out_width, u16 out_height,
@@ -86,6 +117,16 @@ struct dispc_features {
 		u16 width, u16 height, u16 out_width, u16 out_height,
 		bool mem_to_mem);
 	u8 num_fifos;
+	const enum dispc_feature_id *features;
+	unsigned int num_features;
+	const struct dss_reg_field *reg_fields;
+	const unsigned int num_reg_fields;
+	const enum omap_overlay_caps *overlay_caps;
+	const u32 **supported_color_modes;
+	unsigned int num_mgrs;
+	unsigned int num_ovls;
+	unsigned int buffer_size_unit;
+	unsigned int burst_size_unit;
 
 	/* swap GFX & WB fifos */
 	bool gfx_fifo_workaround:1;
@@ -180,6 +221,17 @@ enum mgr_reg_fields {
 	DISPC_MGR_FLD_NUM,
 };
 
+/* DISPC register field id */
+enum dispc_feat_reg_field {
+	FEAT_REG_FIRHINC,
+	FEAT_REG_FIRVINC,
+	FEAT_REG_FIFOHIGHTHRESHOLD,
+	FEAT_REG_FIFOLOWTHRESHOLD,
+	FEAT_REG_FIFOSIZE,
+	FEAT_REG_HORIZONTALACCU,
+	FEAT_REG_VERTICALACCU,
+};
+
 struct dispc_reg_field {
 	u16 reg;
 	u8 high;
@@ -343,6 +395,38 @@ static void mgr_fld_write(enum omap_channel channel,
 		spin_unlock_irqrestore(&dispc.control_lock, flags);
 }
 
+static int dispc_get_num_ovls(void)
+{
+	return dispc.feat->num_ovls;
+}
+
+static int dispc_get_num_mgrs(void)
+{
+	return dispc.feat->num_mgrs;
+}
+
+static void dispc_get_reg_field(enum dispc_feat_reg_field id,
+				u8 *start, u8 *end)
+{
+	if (id >= dispc.feat->num_reg_fields)
+		BUG();
+
+	*start = dispc.feat->reg_fields[id].start;
+	*end = dispc.feat->reg_fields[id].end;
+}
+
+static bool dispc_has_feature(enum dispc_feature_id id)
+{
+	unsigned int i;
+
+	for (i = 0; i < dispc.feat->num_features; i++) {
+		if (dispc.feat->features[i] == id)
+			return true;
+	}
+
+	return false;
+}
+
 #define SR(reg) \
 	dispc.ctx[DISPC_##reg / sizeof(u32)] = dispc_read_reg(DISPC_##reg)
 #define RR(reg) \
@@ -358,19 +442,19 @@ static void dispc_save_context(void)
 	SR(CONTROL);
 	SR(CONFIG);
 	SR(LINE_NUMBER);
-	if (dss_has_feature(FEAT_ALPHA_FIXED_ZORDER) ||
-			dss_has_feature(FEAT_ALPHA_FREE_ZORDER))
+	if (dispc_has_feature(FEAT_ALPHA_FIXED_ZORDER) ||
+			dispc_has_feature(FEAT_ALPHA_FREE_ZORDER))
 		SR(GLOBAL_ALPHA);
-	if (dss_has_feature(FEAT_MGR_LCD2)) {
+	if (dispc_has_feature(FEAT_MGR_LCD2)) {
 		SR(CONTROL2);
 		SR(CONFIG2);
 	}
-	if (dss_has_feature(FEAT_MGR_LCD3)) {
+	if (dispc_has_feature(FEAT_MGR_LCD3)) {
 		SR(CONTROL3);
 		SR(CONFIG3);
 	}
 
-	for (i = 0; i < dss_feat_get_num_mgrs(); i++) {
+	for (i = 0; i < dispc_get_num_mgrs(); i++) {
 		SR(DEFAULT_COLOR(i));
 		SR(TRANS_COLOR(i));
 		SR(SIZE_MGR(i));
@@ -385,14 +469,14 @@ static void dispc_save_context(void)
 		SR(DATA_CYCLE2(i));
 		SR(DATA_CYCLE3(i));
 
-		if (dss_has_feature(FEAT_CPR)) {
+		if (dispc_has_feature(FEAT_CPR)) {
 			SR(CPR_COEF_R(i));
 			SR(CPR_COEF_G(i));
 			SR(CPR_COEF_B(i));
 		}
 	}
 
-	for (i = 0; i < dss_feat_get_num_ovls(); i++) {
+	for (i = 0; i < dispc_get_num_ovls(); i++) {
 		SR(OVL_BA0(i));
 		SR(OVL_BA1(i));
 		SR(OVL_POSITION(i));
@@ -401,7 +485,7 @@ static void dispc_save_context(void)
 		SR(OVL_FIFO_THRESHOLD(i));
 		SR(OVL_ROW_INC(i));
 		SR(OVL_PIXEL_INC(i));
-		if (dss_has_feature(FEAT_PRELOAD))
+		if (dispc_has_feature(FEAT_PRELOAD))
 			SR(OVL_PRELOAD(i));
 		if (i == OMAP_DSS_GFX) {
 			SR(OVL_WINDOW_SKIP(i));
@@ -422,12 +506,12 @@ static void dispc_save_context(void)
 		for (j = 0; j < 5; j++)
 			SR(OVL_CONV_COEF(i, j));
 
-		if (dss_has_feature(FEAT_FIR_COEF_V)) {
+		if (dispc_has_feature(FEAT_FIR_COEF_V)) {
 			for (j = 0; j < 8; j++)
 				SR(OVL_FIR_COEF_V(i, j));
 		}
 
-		if (dss_has_feature(FEAT_HANDLE_UV_SEPARATE)) {
+		if (dispc_has_feature(FEAT_HANDLE_UV_SEPARATE)) {
 			SR(OVL_BA0_UV(i));
 			SR(OVL_BA1_UV(i));
 			SR(OVL_FIR2(i));
@@ -443,11 +527,11 @@ static void dispc_save_context(void)
 			for (j = 0; j < 8; j++)
 				SR(OVL_FIR_COEF_V2(i, j));
 		}
-		if (dss_has_feature(FEAT_ATTR2))
+		if (dispc_has_feature(FEAT_ATTR2))
 			SR(OVL_ATTRIBUTES2(i));
 	}
 
-	if (dss_has_feature(FEAT_CORE_CLK_DIV))
+	if (dispc_has_feature(FEAT_CORE_CLK_DIV))
 		SR(DIVISOR);
 
 	dispc.ctx_valid = true;
@@ -468,15 +552,15 @@ static void dispc_restore_context(void)
 	/*RR(CONTROL);*/
 	RR(CONFIG);
 	RR(LINE_NUMBER);
-	if (dss_has_feature(FEAT_ALPHA_FIXED_ZORDER) ||
-			dss_has_feature(FEAT_ALPHA_FREE_ZORDER))
+	if (dispc_has_feature(FEAT_ALPHA_FIXED_ZORDER) ||
+			dispc_has_feature(FEAT_ALPHA_FREE_ZORDER))
 		RR(GLOBAL_ALPHA);
-	if (dss_has_feature(FEAT_MGR_LCD2))
+	if (dispc_has_feature(FEAT_MGR_LCD2))
 		RR(CONFIG2);
-	if (dss_has_feature(FEAT_MGR_LCD3))
+	if (dispc_has_feature(FEAT_MGR_LCD3))
 		RR(CONFIG3);
 
-	for (i = 0; i < dss_feat_get_num_mgrs(); i++) {
+	for (i = 0; i < dispc_get_num_mgrs(); i++) {
 		RR(DEFAULT_COLOR(i));
 		RR(TRANS_COLOR(i));
 		RR(SIZE_MGR(i));
@@ -491,14 +575,14 @@ static void dispc_restore_context(void)
 		RR(DATA_CYCLE2(i));
 		RR(DATA_CYCLE3(i));
 
-		if (dss_has_feature(FEAT_CPR)) {
+		if (dispc_has_feature(FEAT_CPR)) {
 			RR(CPR_COEF_R(i));
 			RR(CPR_COEF_G(i));
 			RR(CPR_COEF_B(i));
 		}
 	}
 
-	for (i = 0; i < dss_feat_get_num_ovls(); i++) {
+	for (i = 0; i < dispc_get_num_ovls(); i++) {
 		RR(OVL_BA0(i));
 		RR(OVL_BA1(i));
 		RR(OVL_POSITION(i));
@@ -507,7 +591,7 @@ static void dispc_restore_context(void)
 		RR(OVL_FIFO_THRESHOLD(i));
 		RR(OVL_ROW_INC(i));
 		RR(OVL_PIXEL_INC(i));
-		if (dss_has_feature(FEAT_PRELOAD))
+		if (dispc_has_feature(FEAT_PRELOAD))
 			RR(OVL_PRELOAD(i));
 		if (i == OMAP_DSS_GFX) {
 			RR(OVL_WINDOW_SKIP(i));
@@ -528,12 +612,12 @@ static void dispc_restore_context(void)
 		for (j = 0; j < 5; j++)
 			RR(OVL_CONV_COEF(i, j));
 
-		if (dss_has_feature(FEAT_FIR_COEF_V)) {
+		if (dispc_has_feature(FEAT_FIR_COEF_V)) {
 			for (j = 0; j < 8; j++)
 				RR(OVL_FIR_COEF_V(i, j));
 		}
 
-		if (dss_has_feature(FEAT_HANDLE_UV_SEPARATE)) {
+		if (dispc_has_feature(FEAT_HANDLE_UV_SEPARATE)) {
 			RR(OVL_BA0_UV(i));
 			RR(OVL_BA1_UV(i));
 			RR(OVL_FIR2(i));
@@ -549,18 +633,18 @@ static void dispc_restore_context(void)
 			for (j = 0; j < 8; j++)
 				RR(OVL_FIR_COEF_V2(i, j));
 		}
-		if (dss_has_feature(FEAT_ATTR2))
+		if (dispc_has_feature(FEAT_ATTR2))
 			RR(OVL_ATTRIBUTES2(i));
 	}
 
-	if (dss_has_feature(FEAT_CORE_CLK_DIV))
+	if (dispc_has_feature(FEAT_CORE_CLK_DIV))
 		RR(DIVISOR);
 
 	/* enable last, because LCD & DIGIT enable are here */
 	RR(CONTROL);
-	if (dss_has_feature(FEAT_MGR_LCD2))
+	if (dispc_has_feature(FEAT_MGR_LCD2))
 		RR(CONTROL2);
-	if (dss_has_feature(FEAT_MGR_LCD3))
+	if (dispc_has_feature(FEAT_MGR_LCD3))
 		RR(CONTROL3);
 	/* clear spurious SYNC_LOST_DIGIT interrupts */
 	dispc_clear_irqstatus(DISPC_IRQ_SYNC_LOST_DIGIT);
@@ -779,7 +863,7 @@ static void dispc_ovl_write_color_conv_coef(enum omap_plane_id plane,
 static void dispc_setup_color_conv_coef(void)
 {
 	int i;
-	int num_ovl = dss_feat_get_num_ovls();
+	int num_ovl = dispc_get_num_ovls();
 	const struct color_conv_coef ctbl_bt601_5_ovl = {
 		/* YUV -> RGB */
 		298, 409, 0, 298, -208, -100, 298, 0, 517, 0,
@@ -868,10 +952,10 @@ static void dispc_ovl_enable_zorder_planes(void)
 {
 	int i;
 
-	if (!dss_has_feature(FEAT_ALPHA_FREE_ZORDER))
+	if (!dispc_has_feature(FEAT_ALPHA_FREE_ZORDER))
 		return;
 
-	for (i = 0; i < dss_feat_get_num_ovls(); i++)
+	for (i = 0; i < dispc_get_num_ovls(); i++)
 		REG_FLD_MOD(DISPC_OVL_ATTRIBUTES(i), 1, 25, 25);
 }
 
@@ -994,7 +1078,7 @@ static bool format_is_yuv(u32 fourcc)
 static void dispc_ovl_configure_burst_type(enum omap_plane_id plane,
 		enum omap_dss_rotation_type rotation_type)
 {
-	if (dss_has_feature(FEAT_BURST_2D) == 0)
+	if (dispc_has_feature(FEAT_BURST_2D) == 0)
 		return;
 
 	if (rotation_type == OMAP_DSS_ROT_TILER)
@@ -1025,7 +1109,7 @@ static void dispc_ovl_set_channel_out(enum omap_plane_id plane,
 	}
 
 	val = dispc_read_reg(DISPC_OVL_ATTRIBUTES(plane));
-	if (dss_has_feature(FEAT_MGR_LCD2)) {
+	if (dispc_has_feature(FEAT_MGR_LCD2)) {
 		switch (channel) {
 		case OMAP_DSS_CHANNEL_LCD:
 			chan = 0;
@@ -1040,7 +1124,7 @@ static void dispc_ovl_set_channel_out(enum omap_plane_id plane,
 			chan2 = 1;
 			break;
 		case OMAP_DSS_CHANNEL_LCD3:
-			if (dss_has_feature(FEAT_MGR_LCD3)) {
+			if (dispc_has_feature(FEAT_MGR_LCD3)) {
 				chan = 0;
 				chan2 = 2;
 			} else {
@@ -1089,7 +1173,7 @@ static enum omap_channel dispc_ovl_get_channel_out(enum omap_plane_id plane)
 	if (FLD_GET(val, shift, shift) == 1)
 		return OMAP_DSS_CHANNEL_DIGIT;
 
-	if (!dss_has_feature(FEAT_MGR_LCD2))
+	if (!dispc_has_feature(FEAT_MGR_LCD2))
 		return OMAP_DSS_CHANNEL_LCD;
 
 	switch (FLD_GET(val, 31, 30)) {
@@ -1128,7 +1212,7 @@ static void dispc_configure_burst_sizes(void)
 	const int burst_size = BURST_SIZE_X8;
 
 	/* Configure burst size always to maximum size */
-	for (i = 0; i < dss_feat_get_num_ovls(); ++i)
+	for (i = 0; i < dispc_get_num_ovls(); ++i)
 		dispc_ovl_set_burst_size(i, burst_size);
 	if (dispc.feat->has_writeback)
 		dispc_ovl_set_burst_size(OMAP_DSS_WB, burst_size);
@@ -1136,19 +1220,28 @@ static void dispc_configure_burst_sizes(void)
 
 static u32 dispc_ovl_get_burst_size(enum omap_plane_id plane)
 {
-	unsigned unit = dss_feat_get_burst_size_unit();
 	/* burst multiplier is always x8 (see dispc_configure_burst_sizes()) */
-	return unit * 8;
+	return dispc.feat->burst_size_unit * 8;
 }
 
-static const u32 *dispc_ovl_get_color_modes(enum omap_plane_id plane)
+static bool dispc_ovl_color_mode_supported(enum omap_plane_id plane, u32 fourcc)
 {
-	return dss_feat_get_supported_color_modes(plane);
+	const u32 *modes;
+	unsigned int i;
+
+	modes = dispc.feat->supported_color_modes[plane];
+
+	for (i = 0; modes[i]; ++i) {
+		if (modes[i] == fourcc)
+			return true;
+	}
+
+	return false;
 }
 
-static int dispc_get_num_ovls(void)
+static const u32 *dispc_ovl_get_color_modes(enum omap_plane_id plane)
 {
-	return dss_feat_get_num_ovls();
+	return dispc.feat->supported_color_modes[plane];
 }
 
 static void dispc_mgr_enable_cpr(enum omap_channel channel, bool enable)
@@ -1223,9 +1316,9 @@ static void dispc_init_fifos(void)
 	u32 unit;
 	int i;
 
-	unit = dss_feat_get_buffer_size_unit();
+	unit = dispc.feat->buffer_size_unit;
 
-	dss_feat_get_reg_field(FEAT_REG_FIFOSIZE, &start, &end);
+	dispc_get_reg_field(FEAT_REG_FIFOSIZE, &start, &end);
 
 	for (fifo = 0; fifo < dispc.feat->num_fifos; ++fifo) {
 		size = REG_GET(DISPC_OVL_FIFO_SIZE_STATUS(fifo), start, end);
@@ -1265,7 +1358,7 @@ static void dispc_init_fifos(void)
 	/*
 	 * Setup default fifo thresholds.
 	 */
-	for (i = 0; i < dss_feat_get_num_ovls(); ++i) {
+	for (i = 0; i < dispc_get_num_ovls(); ++i) {
 		u32 low, high;
 		const bool use_fifomerge = false;
 		const bool manual_update = false;
@@ -1307,7 +1400,7 @@ void dispc_ovl_set_fifo_threshold(enum omap_plane_id plane, u32 low,
 	u8 hi_start, hi_end, lo_start, lo_end;
 	u32 unit;
 
-	unit = dss_feat_get_buffer_size_unit();
+	unit = dispc.feat->buffer_size_unit;
 
 	WARN_ON(low % unit != 0);
 	WARN_ON(high % unit != 0);
@@ -1315,8 +1408,8 @@ void dispc_ovl_set_fifo_threshold(enum omap_plane_id plane, u32 low,
 	low /= unit;
 	high /= unit;
 
-	dss_feat_get_reg_field(FEAT_REG_FIFOHIGHTHRESHOLD, &hi_start, &hi_end);
-	dss_feat_get_reg_field(FEAT_REG_FIFOLOWTHRESHOLD, &lo_start, &lo_end);
+	dispc_get_reg_field(FEAT_REG_FIFOHIGHTHRESHOLD, &hi_start, &hi_end);
+	dispc_get_reg_field(FEAT_REG_FIFOLOWTHRESHOLD, &lo_start, &lo_end);
 
 	DSSDBG("fifo(%d) threshold (bytes), old %u/%u, new %u/%u\n",
 			plane,
@@ -1335,14 +1428,14 @@ void dispc_ovl_set_fifo_threshold(enum omap_plane_id plane, u32 low,
 	 * large for the preload field, set the threshold to the maximum value
 	 * that can be held by the preload register
 	 */
-	if (dss_has_feature(FEAT_PRELOAD) && dispc.feat->set_max_preload &&
+	if (dispc_has_feature(FEAT_PRELOAD) && dispc.feat->set_max_preload &&
 			plane != OMAP_DSS_WB)
 		dispc_write_reg(DISPC_OVL_PRELOAD(plane), min(high, 0xfffu));
 }
 
 void dispc_enable_fifomerge(bool enable)
 {
-	if (!dss_has_feature(FEAT_FIFO_MERGE)) {
+	if (!dispc_has_feature(FEAT_FIFO_MERGE)) {
 		WARN_ON(enable);
 		return;
 	}
@@ -1360,7 +1453,7 @@ void dispc_ovl_compute_fifo_thresholds(enum omap_plane_id plane,
 	 * buffer_units, and the fifo thresholds must be buffer_unit aligned.
 	 */
 
-	unsigned buf_unit = dss_feat_get_buffer_size_unit();
+	unsigned buf_unit = dispc.feat->buffer_size_unit;
 	unsigned ovl_fifo_size, total_fifo_size, burst_size;
 	int i;
 
@@ -1369,7 +1462,7 @@ void dispc_ovl_compute_fifo_thresholds(enum omap_plane_id plane,
 
 	if (use_fifomerge) {
 		total_fifo_size = 0;
-		for (i = 0; i < dss_feat_get_num_ovls(); ++i)
+		for (i = 0; i < dispc_get_num_ovls(); ++i)
 			total_fifo_size += dispc_ovl_get_fifo_size(i);
 	} else {
 		total_fifo_size = ovl_fifo_size;
@@ -1381,7 +1474,7 @@ void dispc_ovl_compute_fifo_thresholds(enum omap_plane_id plane,
 	 * combined fifo size
 	 */
 
-	if (manual_update && dss_has_feature(FEAT_OMAP3_DSI_FIFO_BUG)) {
+	if (manual_update && dispc_has_feature(FEAT_OMAP3_DSI_FIFO_BUG)) {
 		*fifo_low = ovl_fifo_size - burst_size * 2;
 		*fifo_high = total_fifo_size - burst_size;
 	} else if (plane == OMAP_DSS_WB) {
@@ -1435,9 +1528,9 @@ static void dispc_init_mflag(void)
 		(1 << 0) |	/* MFLAG_CTRL = force always on */
 		(0 << 2));	/* MFLAG_START = disable */
 
-	for (i = 0; i < dss_feat_get_num_ovls(); ++i) {
+	for (i = 0; i < dispc_get_num_ovls(); ++i) {
 		u32 size = dispc_ovl_get_fifo_size(i);
-		u32 unit = dss_feat_get_buffer_size_unit();
+		u32 unit = dispc.feat->buffer_size_unit;
 		u32 low, high;
 
 		dispc_ovl_set_mflag(i, true);
@@ -1456,7 +1549,7 @@ static void dispc_init_mflag(void)
 
 	if (dispc.feat->has_writeback) {
 		u32 size = dispc_ovl_get_fifo_size(OMAP_DSS_WB);
-		u32 unit = dss_feat_get_buffer_size_unit();
+		u32 unit = dispc.feat->buffer_size_unit;
 		u32 low, high;
 
 		dispc_ovl_set_mflag(OMAP_DSS_WB, true);
@@ -1483,10 +1576,8 @@ static void dispc_ovl_set_fir(enum omap_plane_id plane,
 	if (color_comp == DISPC_COLOR_COMPONENT_RGB_Y) {
 		u8 hinc_start, hinc_end, vinc_start, vinc_end;
 
-		dss_feat_get_reg_field(FEAT_REG_FIRHINC,
-					&hinc_start, &hinc_end);
-		dss_feat_get_reg_field(FEAT_REG_FIRVINC,
-					&vinc_start, &vinc_end);
+		dispc_get_reg_field(FEAT_REG_FIRHINC, &hinc_start, &hinc_end);
+		dispc_get_reg_field(FEAT_REG_FIRVINC, &vinc_start, &vinc_end);
 		val = FLD_VAL(vinc, vinc_start, vinc_end) |
 				FLD_VAL(hinc, hinc_start, hinc_end);
 
@@ -1503,8 +1594,8 @@ static void dispc_ovl_set_vid_accu0(enum omap_plane_id plane, int haccu,
 	u32 val;
 	u8 hor_start, hor_end, vert_start, vert_end;
 
-	dss_feat_get_reg_field(FEAT_REG_HORIZONTALACCU, &hor_start, &hor_end);
-	dss_feat_get_reg_field(FEAT_REG_VERTICALACCU, &vert_start, &vert_end);
+	dispc_get_reg_field(FEAT_REG_HORIZONTALACCU, &hor_start, &hor_end);
+	dispc_get_reg_field(FEAT_REG_VERTICALACCU, &vert_start, &vert_end);
 
 	val = FLD_VAL(vaccu, vert_start, vert_end) |
 			FLD_VAL(haccu, hor_start, hor_end);
@@ -1518,8 +1609,8 @@ static void dispc_ovl_set_vid_accu1(enum omap_plane_id plane, int haccu,
 	u32 val;
 	u8 hor_start, hor_end, vert_start, vert_end;
 
-	dss_feat_get_reg_field(FEAT_REG_HORIZONTALACCU, &hor_start, &hor_end);
-	dss_feat_get_reg_field(FEAT_REG_VERTICALACCU, &vert_start, &vert_end);
+	dispc_get_reg_field(FEAT_REG_HORIZONTALACCU, &hor_start, &hor_end);
+	dispc_get_reg_field(FEAT_REG_VERTICALACCU, &vert_start, &vert_end);
 
 	val = FLD_VAL(vaccu, vert_start, vert_end) |
 			FLD_VAL(haccu, hor_start, hor_end);
@@ -1671,14 +1762,14 @@ static void dispc_ovl_set_scaling_common(enum omap_plane_id plane,
 	l |= five_taps ? (1 << 21) : 0;
 
 	/* VRESIZECONF and HRESIZECONF */
-	if (dss_has_feature(FEAT_RESIZECONF)) {
+	if (dispc_has_feature(FEAT_RESIZECONF)) {
 		l &= ~(0x3 << 7);
 		l |= (orig_width <= out_width) ? 0 : (1 << 7);
 		l |= (orig_height <= out_height) ? 0 : (1 << 8);
 	}
 
 	/* LINEBUFFERSPLIT */
-	if (dss_has_feature(FEAT_LINEBUFFERSPLIT)) {
+	if (dispc_has_feature(FEAT_LINEBUFFERSPLIT)) {
 		l &= ~(0x1 << 22);
 		l |= five_taps ? (1 << 22) : 0;
 	}
@@ -1713,7 +1804,7 @@ static void dispc_ovl_set_scaling_uv(enum omap_plane_id plane,
 	int scale_y = out_height != orig_height;
 	bool chroma_upscale = plane != OMAP_DSS_WB;
 
-	if (!dss_has_feature(FEAT_HANDLE_UV_SEPARATE))
+	if (!dispc_has_feature(FEAT_HANDLE_UV_SEPARATE))
 		return;
 
 	if (!format_is_yuv(fourcc)) {
@@ -1860,11 +1951,11 @@ static void dispc_ovl_set_rotation_attrs(enum omap_plane_id plane, u8 rotation,
 		vidrot = 1;
 
 	REG_FLD_MOD(DISPC_OVL_ATTRIBUTES(plane), vidrot, 13, 12);
-	if (dss_has_feature(FEAT_ROWREPEATENABLE))
+	if (dispc_has_feature(FEAT_ROWREPEATENABLE))
 		REG_FLD_MOD(DISPC_OVL_ATTRIBUTES(plane),
 			row_repeat ? 1 : 0, 18, 18);
 
-	if (dss_feat_color_mode_supported(plane, DRM_FORMAT_NV12)) {
+	if (dispc_ovl_color_mode_supported(plane, DRM_FORMAT_NV12)) {
 		bool doublestride =
 			fourcc == DRM_FORMAT_NV12 &&
 			rotation_type == OMAP_DSS_ROT_TILER &&
@@ -2118,8 +2209,7 @@ static int dispc_ovl_calc_scaling_24xx(unsigned long pclk, unsigned long lclk,
 	int error;
 	u16 in_width, in_height;
 	int min_factor = min(*decim_x, *decim_y);
-	const int maxsinglelinewidth =
-			dss_feat_get_param_max(FEAT_PARAM_LINEWIDTH);
+	const int maxsinglelinewidth = dispc.feat->max_line_width;
 
 	*five_taps = false;
 
@@ -2163,8 +2253,7 @@ static int dispc_ovl_calc_scaling_34xx(unsigned long pclk, unsigned long lclk,
 {
 	int error;
 	u16 in_width, in_height;
-	const int maxsinglelinewidth =
-			dss_feat_get_param_max(FEAT_PARAM_LINEWIDTH);
+	const int maxsinglelinewidth = dispc.feat->max_line_width;
 
 	do {
 		in_height = height / *decim_y;
@@ -2249,9 +2338,8 @@ static int dispc_ovl_calc_scaling_44xx(unsigned long pclk, unsigned long lclk,
 	u16 in_width, in_width_max;
 	int decim_x_min = *decim_x;
 	u16 in_height = height / *decim_y;
-	const int maxsinglelinewidth =
-				dss_feat_get_param_max(FEAT_PARAM_LINEWIDTH);
-	const int maxdownscale = dss_feat_get_param_max(FEAT_PARAM_DOWNSCALE);
+	const int maxsinglelinewidth = dispc.feat->max_line_width;
+	const int maxdownscale = dispc.feat->max_downscale;
 
 	if (mem_to_mem) {
 		in_width_max = out_width * maxdownscale;
@@ -2311,7 +2399,7 @@ static int dispc_ovl_calc_scaling(unsigned long pclk, unsigned long lclk,
 		int *x_predecim, int *y_predecim, u16 pos_x,
 		enum omap_dss_rotation_type rotation_type, bool mem_to_mem)
 {
-	const int maxdownscale = dss_feat_get_param_max(FEAT_PARAM_DOWNSCALE);
+	const int maxdownscale = dispc.feat->max_downscale;
 	const int max_decim_limit = 16;
 	unsigned long core_clk = 0;
 	int decim_x, decim_y, ret;
@@ -2332,7 +2420,7 @@ static int dispc_ovl_calc_scaling(unsigned long pclk, unsigned long lclk,
 	} else {
 		*x_predecim = max_decim_limit;
 		*y_predecim = (rotation_type == OMAP_DSS_ROT_TILER &&
-				dss_has_feature(FEAT_BURST_2D)) ?
+				dispc_has_feature(FEAT_BURST_2D)) ?
 				2 : max_decim_limit;
 	}
 
@@ -2428,7 +2516,7 @@ static int dispc_ovl_setup_common(enum omap_plane_id plane,
 			out_height);
 	}
 
-	if (!dss_feat_color_mode_supported(plane, fourcc))
+	if (!dispc_ovl_color_mode_supported(plane, fourcc))
 		return -EINVAL;
 
 	r = dispc_ovl_calc_scaling(pclk, lclk, caps, vm, in_width,
@@ -2549,7 +2637,7 @@ static int dispc_ovl_setup(enum omap_plane_id plane,
 		enum omap_channel channel)
 {
 	int r;
-	enum omap_overlay_caps caps = dss_feat_get_overlay_caps(plane);
+	enum omap_overlay_caps caps = dispc.feat->overlay_caps[plane];
 	const bool replication = true;
 
 	DSSDBG("dispc_ovl_setup %d, pa %pad, pa_uv %pad, sw %d, %d,%d, %dx%d ->"
@@ -2647,12 +2735,12 @@ static int dispc_ovl_enable(enum omap_plane_id plane, bool enable)
 
 static enum omap_dss_output_id dispc_mgr_get_supported_outputs(enum omap_channel channel)
 {
-	return dss_feat_get_supported_outputs(channel);
+	return dss_get_supported_outputs(channel);
 }
 
 static void dispc_lcd_enable_signal_polarity(bool act_high)
 {
-	if (!dss_has_feature(FEAT_LCDENABLEPOL))
+	if (!dispc_has_feature(FEAT_LCDENABLEPOL))
 		return;
 
 	REG_FLD_MOD(DISPC_CONTROL, act_high ? 1 : 0, 29, 29);
@@ -2660,7 +2748,7 @@ static void dispc_lcd_enable_signal_polarity(bool act_high)
 
 void dispc_lcd_enable_signal(bool enable)
 {
-	if (!dss_has_feature(FEAT_LCDENABLESIGNAL))
+	if (!dispc_has_feature(FEAT_LCDENABLESIGNAL))
 		return;
 
 	REG_FLD_MOD(DISPC_CONTROL, enable ? 1 : 0, 28, 28);
@@ -2668,17 +2756,12 @@ void dispc_lcd_enable_signal(bool enable)
 
 void dispc_pck_free_enable(bool enable)
 {
-	if (!dss_has_feature(FEAT_PCKFREEENABLE))
+	if (!dispc_has_feature(FEAT_PCKFREEENABLE))
 		return;
 
 	REG_FLD_MOD(DISPC_CONTROL, enable ? 1 : 0, 27, 27);
 }
 
-static int dispc_get_num_mgrs(void)
-{
-	return dss_feat_get_num_mgrs();
-}
-
 static void dispc_mgr_enable_fifohandcheck(enum omap_channel channel, bool enable)
 {
 	mgr_fld_write(channel, DISPC_MGR_FLD_FIFOHANDCHECK, enable);
@@ -2718,7 +2801,7 @@ static void dispc_mgr_enable_trans_key(enum omap_channel ch, bool enable)
 static void dispc_mgr_enable_alpha_fixed_zorder(enum omap_channel ch,
 		bool enable)
 {
-	if (!dss_has_feature(FEAT_ALPHA_FIXED_ZORDER))
+	if (!dispc_has_feature(FEAT_ALPHA_FIXED_ZORDER))
 		return;
 
 	if (ch == OMAP_DSS_CHANNEL_LCD)
@@ -2735,7 +2818,7 @@ static void dispc_mgr_setup(enum omap_channel channel,
 	dispc_mgr_enable_trans_key(channel, info->trans_enabled);
 	dispc_mgr_enable_alpha_fixed_zorder(channel,
 			info->partial_alpha_enabled);
-	if (dss_has_feature(FEAT_CPR)) {
+	if (dispc_has_feature(FEAT_CPR)) {
 		dispc_mgr_enable_cpr(channel, info->cpr_enable);
 		dispc_mgr_set_cpr_coef(channel, &info->cpr_coefs);
 	}
@@ -3013,7 +3096,7 @@ static void dispc_mgr_set_lcd_divisor(enum omap_channel channel, u16 lck_div,
 	dispc_write_reg(DISPC_DIVISORo(channel),
 			FLD_VAL(lck_div, 23, 16) | FLD_VAL(pck_div, 7, 0));
 
-	if (!dss_has_feature(FEAT_CORE_CLK_DIV) &&
+	if (!dispc_has_feature(FEAT_CORE_CLK_DIV) &&
 			channel == OMAP_DSS_CHANNEL_LCD)
 		dispc.core_clk_rate = dispc_fclk_rate() / lck_div;
 }
@@ -3168,7 +3251,7 @@ void dispc_dump_clocks(struct seq_file *s)
 
 	seq_printf(s, "fck\t\t%-16lu\n", dispc_fclk_rate());
 
-	if (dss_has_feature(FEAT_CORE_CLK_DIV)) {
+	if (dispc_has_feature(FEAT_CORE_CLK_DIV)) {
 		seq_printf(s, "- DISPC-CORE-CLK -\n");
 		l = dispc_read_reg(DISPC_DIVISOR);
 		lcd = FLD_GET(l, 23, 16);
@@ -3179,9 +3262,9 @@ void dispc_dump_clocks(struct seq_file *s)
 
 	dispc_dump_clocks_channel(s, OMAP_DSS_CHANNEL_LCD);
 
-	if (dss_has_feature(FEAT_MGR_LCD2))
+	if (dispc_has_feature(FEAT_MGR_LCD2))
 		dispc_dump_clocks_channel(s, OMAP_DSS_CHANNEL_LCD2);
-	if (dss_has_feature(FEAT_MGR_LCD3))
+	if (dispc_has_feature(FEAT_MGR_LCD3))
 		dispc_dump_clocks_channel(s, OMAP_DSS_CHANNEL_LCD3);
 
 	dispc_runtime_put();
@@ -3221,18 +3304,18 @@ static void dispc_dump_regs(struct seq_file *s)
 	DUMPREG(DISPC_CAPABLE);
 	DUMPREG(DISPC_LINE_STATUS);
 	DUMPREG(DISPC_LINE_NUMBER);
-	if (dss_has_feature(FEAT_ALPHA_FIXED_ZORDER) ||
-			dss_has_feature(FEAT_ALPHA_FREE_ZORDER))
+	if (dispc_has_feature(FEAT_ALPHA_FIXED_ZORDER) ||
+			dispc_has_feature(FEAT_ALPHA_FREE_ZORDER))
 		DUMPREG(DISPC_GLOBAL_ALPHA);
-	if (dss_has_feature(FEAT_MGR_LCD2)) {
+	if (dispc_has_feature(FEAT_MGR_LCD2)) {
 		DUMPREG(DISPC_CONTROL2);
 		DUMPREG(DISPC_CONFIG2);
 	}
-	if (dss_has_feature(FEAT_MGR_LCD3)) {
+	if (dispc_has_feature(FEAT_MGR_LCD3)) {
 		DUMPREG(DISPC_CONTROL3);
 		DUMPREG(DISPC_CONFIG3);
 	}
-	if (dss_has_feature(FEAT_MFLAG))
+	if (dispc_has_feature(FEAT_MFLAG))
 		DUMPREG(DISPC_GLOBAL_MFLAG_ATTRIBUTE);
 
 #undef DUMPREG
@@ -3245,7 +3328,7 @@ static void dispc_dump_regs(struct seq_file *s)
 	p_names = mgr_names;
 
 	/* DISPC channel specific registers */
-	for (i = 0; i < dss_feat_get_num_mgrs(); i++) {
+	for (i = 0; i < dispc_get_num_mgrs(); i++) {
 		DUMPREG(i, DISPC_DEFAULT_COLOR);
 		DUMPREG(i, DISPC_TRANS_COLOR);
 		DUMPREG(i, DISPC_SIZE_MGR);
@@ -3262,7 +3345,7 @@ static void dispc_dump_regs(struct seq_file *s)
 		DUMPREG(i, DISPC_DATA_CYCLE2);
 		DUMPREG(i, DISPC_DATA_CYCLE3);
 
-		if (dss_has_feature(FEAT_CPR)) {
+		if (dispc_has_feature(FEAT_CPR)) {
 			DUMPREG(i, DISPC_CPR_COEF_R);
 			DUMPREG(i, DISPC_CPR_COEF_G);
 			DUMPREG(i, DISPC_CPR_COEF_B);
@@ -3271,7 +3354,7 @@ static void dispc_dump_regs(struct seq_file *s)
 
 	p_names = ovl_names;
 
-	for (i = 0; i < dss_feat_get_num_ovls(); i++) {
+	for (i = 0; i < dispc_get_num_ovls(); i++) {
 		DUMPREG(i, DISPC_OVL_BA0);
 		DUMPREG(i, DISPC_OVL_BA1);
 		DUMPREG(i, DISPC_OVL_POSITION);
@@ -3282,9 +3365,9 @@ static void dispc_dump_regs(struct seq_file *s)
 		DUMPREG(i, DISPC_OVL_ROW_INC);
 		DUMPREG(i, DISPC_OVL_PIXEL_INC);
 
-		if (dss_has_feature(FEAT_PRELOAD))
+		if (dispc_has_feature(FEAT_PRELOAD))
 			DUMPREG(i, DISPC_OVL_PRELOAD);
-		if (dss_has_feature(FEAT_MFLAG))
+		if (dispc_has_feature(FEAT_MFLAG))
 			DUMPREG(i, DISPC_OVL_MFLAG_THRESHOLD);
 
 		if (i == OMAP_DSS_GFX) {
@@ -3297,14 +3380,14 @@ static void dispc_dump_regs(struct seq_file *s)
 		DUMPREG(i, DISPC_OVL_PICTURE_SIZE);
 		DUMPREG(i, DISPC_OVL_ACCU0);
 		DUMPREG(i, DISPC_OVL_ACCU1);
-		if (dss_has_feature(FEAT_HANDLE_UV_SEPARATE)) {
+		if (dispc_has_feature(FEAT_HANDLE_UV_SEPARATE)) {
 			DUMPREG(i, DISPC_OVL_BA0_UV);
 			DUMPREG(i, DISPC_OVL_BA1_UV);
 			DUMPREG(i, DISPC_OVL_FIR2);
 			DUMPREG(i, DISPC_OVL_ACCU2_0);
 			DUMPREG(i, DISPC_OVL_ACCU2_1);
 		}
-		if (dss_has_feature(FEAT_ATTR2))
+		if (dispc_has_feature(FEAT_ATTR2))
 			DUMPREG(i, DISPC_OVL_ATTRIBUTES2);
 	}
 
@@ -3319,21 +3402,21 @@ static void dispc_dump_regs(struct seq_file *s)
 		DUMPREG(i, DISPC_OVL_ROW_INC);
 		DUMPREG(i, DISPC_OVL_PIXEL_INC);
 
-		if (dss_has_feature(FEAT_MFLAG))
+		if (dispc_has_feature(FEAT_MFLAG))
 			DUMPREG(i, DISPC_OVL_MFLAG_THRESHOLD);
 
 		DUMPREG(i, DISPC_OVL_FIR);
 		DUMPREG(i, DISPC_OVL_PICTURE_SIZE);
 		DUMPREG(i, DISPC_OVL_ACCU0);
 		DUMPREG(i, DISPC_OVL_ACCU1);
-		if (dss_has_feature(FEAT_HANDLE_UV_SEPARATE)) {
+		if (dispc_has_feature(FEAT_HANDLE_UV_SEPARATE)) {
 			DUMPREG(i, DISPC_OVL_BA0_UV);
 			DUMPREG(i, DISPC_OVL_BA1_UV);
 			DUMPREG(i, DISPC_OVL_FIR2);
 			DUMPREG(i, DISPC_OVL_ACCU2_0);
 			DUMPREG(i, DISPC_OVL_ACCU2_1);
 		}
-		if (dss_has_feature(FEAT_ATTR2))
+		if (dispc_has_feature(FEAT_ATTR2))
 			DUMPREG(i, DISPC_OVL_ATTRIBUTES2);
 	}
 
@@ -3349,7 +3432,7 @@ static void dispc_dump_regs(struct seq_file *s)
 	/* Video pipeline coefficient registers */
 
 	/* start from OMAP_DSS_VIDEO1 */
-	for (i = 1; i < dss_feat_get_num_ovls(); i++) {
+	for (i = 1; i < dispc_get_num_ovls(); i++) {
 		for (j = 0; j < 8; j++)
 			DUMPREG(i, DISPC_OVL_FIR_COEF_H, j);
 
@@ -3359,12 +3442,12 @@ static void dispc_dump_regs(struct seq_file *s)
 		for (j = 0; j < 5; j++)
 			DUMPREG(i, DISPC_OVL_CONV_COEF, j);
 
-		if (dss_has_feature(FEAT_FIR_COEF_V)) {
+		if (dispc_has_feature(FEAT_FIR_COEF_V)) {
 			for (j = 0; j < 8; j++)
 				DUMPREG(i, DISPC_OVL_FIR_COEF_V, j);
 		}
 
-		if (dss_has_feature(FEAT_HANDLE_UV_SEPARATE)) {
+		if (dispc_has_feature(FEAT_HANDLE_UV_SEPARATE)) {
 			for (j = 0; j < 8; j++)
 				DUMPREG(i, DISPC_OVL_FIR_COEF_H2, j);
 
@@ -3397,7 +3480,7 @@ int dispc_calc_clock_rates(unsigned long dispc_fclk_rate,
 	return 0;
 }
 
-bool dispc_div_calc(unsigned long dispc,
+bool dispc_div_calc(unsigned long dispc_freq,
 		unsigned long pck_min, unsigned long pck_max,
 		dispc_div_calc_func func, void *data)
 {
@@ -3415,19 +3498,19 @@ bool dispc_div_calc(unsigned long dispc,
 	min_fck_per_pck = 0;
 #endif
 
-	pckd_hw_min = dss_feat_get_param_min(FEAT_PARAM_DSS_PCD);
-	pckd_hw_max = dss_feat_get_param_max(FEAT_PARAM_DSS_PCD);
+	pckd_hw_min = dispc.feat->min_pcd;
+	pckd_hw_max = 255;
 
-	lck_max = dss_feat_get_param_max(FEAT_PARAM_DSS_FCK);
+	lck_max = dss_get_max_fck_rate();
 
 	pck_min = pck_min ? pck_min : 1;
 	pck_max = pck_max ? pck_max : ULONG_MAX;
 
-	lckd_start = max(DIV_ROUND_UP(dispc, lck_max), 1ul);
-	lckd_stop = min(dispc / pck_min, 255ul);
+	lckd_start = max(DIV_ROUND_UP(dispc_freq, lck_max), 1ul);
+	lckd_stop = min(dispc_freq / pck_min, 255ul);
 
 	for (lckd = lckd_start; lckd <= lckd_stop; ++lckd) {
-		lck = dispc / lckd;
+		lck = dispc_freq / lckd;
 
 		pckd_start = max(DIV_ROUND_UP(lck, pck_max), pckd_hw_min);
 		pckd_stop = min(lck / pck_min, pckd_hw_max);
@@ -3441,7 +3524,7 @@ bool dispc_div_calc(unsigned long dispc,
 			 * also. Thus we need to use the calculated lck. For
 			 * OMAP4+ the DISPC fclk is a separate clock.
 			 */
-			if (dss_has_feature(FEAT_CORE_CLK_DIV))
+			if (dispc_has_feature(FEAT_CORE_CLK_DIV))
 				fck = dispc_core_clk_rate();
 			else
 				fck = lck;
@@ -3556,10 +3639,10 @@ static void dispc_restore_gamma_tables(void)
 
 	dispc_mgr_write_gamma_table(OMAP_DSS_CHANNEL_DIGIT);
 
-	if (dss_has_feature(FEAT_MGR_LCD2))
+	if (dispc_has_feature(FEAT_MGR_LCD2))
 		dispc_mgr_write_gamma_table(OMAP_DSS_CHANNEL_LCD2);
 
-	if (dss_has_feature(FEAT_MGR_LCD3))
+	if (dispc_has_feature(FEAT_MGR_LCD3))
 		dispc_mgr_write_gamma_table(OMAP_DSS_CHANNEL_LCD3);
 }
 
@@ -3627,11 +3710,11 @@ static int dispc_init_gamma_tables(void)
 		u32 *gt;
 
 		if (channel == OMAP_DSS_CHANNEL_LCD2 &&
-		    !dss_has_feature(FEAT_MGR_LCD2))
+		    !dispc_has_feature(FEAT_MGR_LCD2))
 			continue;
 
 		if (channel == OMAP_DSS_CHANNEL_LCD3 &&
-		    !dss_has_feature(FEAT_MGR_LCD3))
+		    !dispc_has_feature(FEAT_MGR_LCD3))
 			continue;
 
 		gt = devm_kmalloc_array(&dispc.pdev->dev, gdesc->len,
@@ -3651,7 +3734,7 @@ static void _omap_dispc_initial_config(void)
 	u32 l;
 
 	/* Exclusively enable DISPC_CORE_CLK and set divider to 1 */
-	if (dss_has_feature(FEAT_CORE_CLK_DIV)) {
+	if (dispc_has_feature(FEAT_CORE_CLK_DIV)) {
 		l = dispc_read_reg(DISPC_DIVISOR);
 		/* Use DISPC_DIVISOR.LCD, instead of DISPC_DIVISOR1.LCD */
 		l = FLD_MOD(l, 1, 0, 0);
@@ -3669,7 +3752,7 @@ static void _omap_dispc_initial_config(void)
 	 * func-clock auto-gating. For newer versions
 	 * (dispc.feat->has_gamma_table) this enables tv-out gamma tables.
 	 */
-	if (dss_has_feature(FEAT_FUNCGATED) || dispc.feat->has_gamma_table)
+	if (dispc_has_feature(FEAT_FUNCGATED) || dispc.feat->has_gamma_table)
 		REG_FLD_MOD(DISPC_CONFIG, 1, 9, 9);
 
 	dispc_setup_color_conv_coef();
@@ -3685,10 +3768,272 @@ static void _omap_dispc_initial_config(void)
 	if (dispc.feat->mstandby_workaround)
 		REG_FLD_MOD(DISPC_MSTANDBY_CTRL, 1, 0, 0);
 
-	if (dss_has_feature(FEAT_MFLAG))
+	if (dispc_has_feature(FEAT_MFLAG))
 		dispc_init_mflag();
 }
 
+static const enum dispc_feature_id omap2_dispc_features_list[] = {
+	FEAT_LCDENABLEPOL,
+	FEAT_LCDENABLESIGNAL,
+	FEAT_PCKFREEENABLE,
+	FEAT_FUNCGATED,
+	FEAT_ROWREPEATENABLE,
+	FEAT_RESIZECONF,
+};
+
+static const enum dispc_feature_id omap3_dispc_features_list[] = {
+	FEAT_LCDENABLEPOL,
+	FEAT_LCDENABLESIGNAL,
+	FEAT_PCKFREEENABLE,
+	FEAT_FUNCGATED,
+	FEAT_LINEBUFFERSPLIT,
+	FEAT_ROWREPEATENABLE,
+	FEAT_RESIZECONF,
+	FEAT_CPR,
+	FEAT_PRELOAD,
+	FEAT_FIR_COEF_V,
+	FEAT_ALPHA_FIXED_ZORDER,
+	FEAT_FIFO_MERGE,
+	FEAT_OMAP3_DSI_FIFO_BUG,
+};
+
+static const enum dispc_feature_id am43xx_dispc_features_list[] = {
+	FEAT_LCDENABLEPOL,
+	FEAT_LCDENABLESIGNAL,
+	FEAT_PCKFREEENABLE,
+	FEAT_FUNCGATED,
+	FEAT_LINEBUFFERSPLIT,
+	FEAT_ROWREPEATENABLE,
+	FEAT_RESIZECONF,
+	FEAT_CPR,
+	FEAT_PRELOAD,
+	FEAT_FIR_COEF_V,
+	FEAT_ALPHA_FIXED_ZORDER,
+	FEAT_FIFO_MERGE,
+};
+
+static const enum dispc_feature_id omap4_dispc_features_list[] = {
+	FEAT_MGR_LCD2,
+	FEAT_CORE_CLK_DIV,
+	FEAT_HANDLE_UV_SEPARATE,
+	FEAT_ATTR2,
+	FEAT_CPR,
+	FEAT_PRELOAD,
+	FEAT_FIR_COEF_V,
+	FEAT_ALPHA_FREE_ZORDER,
+	FEAT_FIFO_MERGE,
+	FEAT_BURST_2D,
+};
+
+static const enum dispc_feature_id omap5_dispc_features_list[] = {
+	FEAT_MGR_LCD2,
+	FEAT_MGR_LCD3,
+	FEAT_CORE_CLK_DIV,
+	FEAT_HANDLE_UV_SEPARATE,
+	FEAT_ATTR2,
+	FEAT_CPR,
+	FEAT_PRELOAD,
+	FEAT_FIR_COEF_V,
+	FEAT_ALPHA_FREE_ZORDER,
+	FEAT_FIFO_MERGE,
+	FEAT_BURST_2D,
+	FEAT_MFLAG,
+};
+
+static const struct dss_reg_field omap2_dispc_reg_fields[] = {
+	[FEAT_REG_FIRHINC]			= { 11, 0 },
+	[FEAT_REG_FIRVINC]			= { 27, 16 },
+	[FEAT_REG_FIFOLOWTHRESHOLD]		= { 8, 0 },
+	[FEAT_REG_FIFOHIGHTHRESHOLD]		= { 24, 16 },
+	[FEAT_REG_FIFOSIZE]			= { 8, 0 },
+	[FEAT_REG_HORIZONTALACCU]		= { 9, 0 },
+	[FEAT_REG_VERTICALACCU]			= { 25, 16 },
+};
+
+static const struct dss_reg_field omap3_dispc_reg_fields[] = {
+	[FEAT_REG_FIRHINC]			= { 12, 0 },
+	[FEAT_REG_FIRVINC]			= { 28, 16 },
+	[FEAT_REG_FIFOLOWTHRESHOLD]		= { 11, 0 },
+	[FEAT_REG_FIFOHIGHTHRESHOLD]		= { 27, 16 },
+	[FEAT_REG_FIFOSIZE]			= { 10, 0 },
+	[FEAT_REG_HORIZONTALACCU]		= { 9, 0 },
+	[FEAT_REG_VERTICALACCU]			= { 25, 16 },
+};
+
+static const struct dss_reg_field omap4_dispc_reg_fields[] = {
+	[FEAT_REG_FIRHINC]			= { 12, 0 },
+	[FEAT_REG_FIRVINC]			= { 28, 16 },
+	[FEAT_REG_FIFOLOWTHRESHOLD]		= { 15, 0 },
+	[FEAT_REG_FIFOHIGHTHRESHOLD]		= { 31, 16 },
+	[FEAT_REG_FIFOSIZE]			= { 15, 0 },
+	[FEAT_REG_HORIZONTALACCU]		= { 10, 0 },
+	[FEAT_REG_VERTICALACCU]			= { 26, 16 },
+};
+
+static const enum omap_overlay_caps omap2_dispc_overlay_caps[] = {
+	/* OMAP_DSS_GFX */
+	OMAP_DSS_OVL_CAP_POS | OMAP_DSS_OVL_CAP_REPLICATION,
+
+	/* OMAP_DSS_VIDEO1 */
+	OMAP_DSS_OVL_CAP_SCALE | OMAP_DSS_OVL_CAP_POS |
+		OMAP_DSS_OVL_CAP_REPLICATION,
+
+	/* OMAP_DSS_VIDEO2 */
+	OMAP_DSS_OVL_CAP_SCALE | OMAP_DSS_OVL_CAP_POS |
+		OMAP_DSS_OVL_CAP_REPLICATION,
+};
+
+static const enum omap_overlay_caps omap3430_dispc_overlay_caps[] = {
+	/* OMAP_DSS_GFX */
+	OMAP_DSS_OVL_CAP_GLOBAL_ALPHA | OMAP_DSS_OVL_CAP_POS |
+		OMAP_DSS_OVL_CAP_REPLICATION,
+
+	/* OMAP_DSS_VIDEO1 */
+	OMAP_DSS_OVL_CAP_SCALE | OMAP_DSS_OVL_CAP_POS |
+		OMAP_DSS_OVL_CAP_REPLICATION,
+
+	/* OMAP_DSS_VIDEO2 */
+	OMAP_DSS_OVL_CAP_SCALE | OMAP_DSS_OVL_CAP_GLOBAL_ALPHA |
+		OMAP_DSS_OVL_CAP_POS | OMAP_DSS_OVL_CAP_REPLICATION,
+};
+
+static const enum omap_overlay_caps omap3630_dispc_overlay_caps[] = {
+	/* OMAP_DSS_GFX */
+	OMAP_DSS_OVL_CAP_GLOBAL_ALPHA | OMAP_DSS_OVL_CAP_PRE_MULT_ALPHA |
+		OMAP_DSS_OVL_CAP_POS | OMAP_DSS_OVL_CAP_REPLICATION,
+
+	/* OMAP_DSS_VIDEO1 */
+	OMAP_DSS_OVL_CAP_SCALE | OMAP_DSS_OVL_CAP_POS |
+		OMAP_DSS_OVL_CAP_REPLICATION,
+
+	/* OMAP_DSS_VIDEO2 */
+	OMAP_DSS_OVL_CAP_SCALE | OMAP_DSS_OVL_CAP_GLOBAL_ALPHA |
+		OMAP_DSS_OVL_CAP_PRE_MULT_ALPHA | OMAP_DSS_OVL_CAP_POS |
+		OMAP_DSS_OVL_CAP_REPLICATION,
+};
+
+static const enum omap_overlay_caps omap4_dispc_overlay_caps[] = {
+	/* OMAP_DSS_GFX */
+	OMAP_DSS_OVL_CAP_GLOBAL_ALPHA | OMAP_DSS_OVL_CAP_PRE_MULT_ALPHA |
+		OMAP_DSS_OVL_CAP_ZORDER | OMAP_DSS_OVL_CAP_POS |
+		OMAP_DSS_OVL_CAP_REPLICATION,
+
+	/* OMAP_DSS_VIDEO1 */
+	OMAP_DSS_OVL_CAP_SCALE | OMAP_DSS_OVL_CAP_GLOBAL_ALPHA |
+		OMAP_DSS_OVL_CAP_PRE_MULT_ALPHA | OMAP_DSS_OVL_CAP_ZORDER |
+		OMAP_DSS_OVL_CAP_POS | OMAP_DSS_OVL_CAP_REPLICATION,
+
+	/* OMAP_DSS_VIDEO2 */
+	OMAP_DSS_OVL_CAP_SCALE | OMAP_DSS_OVL_CAP_GLOBAL_ALPHA |
+		OMAP_DSS_OVL_CAP_PRE_MULT_ALPHA | OMAP_DSS_OVL_CAP_ZORDER |
+		OMAP_DSS_OVL_CAP_POS | OMAP_DSS_OVL_CAP_REPLICATION,
+
+	/* OMAP_DSS_VIDEO3 */
+	OMAP_DSS_OVL_CAP_SCALE | OMAP_DSS_OVL_CAP_GLOBAL_ALPHA |
+		OMAP_DSS_OVL_CAP_PRE_MULT_ALPHA | OMAP_DSS_OVL_CAP_ZORDER |
+		OMAP_DSS_OVL_CAP_POS | OMAP_DSS_OVL_CAP_REPLICATION,
+};
+
+#define COLOR_ARRAY(arr...) (const u32[]) { arr, 0 }
+
+static const u32 *omap2_dispc_supported_color_modes[] = {
+
+	/* OMAP_DSS_GFX */
+	COLOR_ARRAY(
+	DRM_FORMAT_RGBX4444, DRM_FORMAT_RGB565,
+	DRM_FORMAT_XRGB8888, DRM_FORMAT_RGB888),
+
+	/* OMAP_DSS_VIDEO1 */
+	COLOR_ARRAY(
+	DRM_FORMAT_RGB565, DRM_FORMAT_XRGB8888,
+	DRM_FORMAT_RGB888, DRM_FORMAT_YUYV,
+	DRM_FORMAT_UYVY),
+
+	/* OMAP_DSS_VIDEO2 */
+	COLOR_ARRAY(
+	DRM_FORMAT_RGB565, DRM_FORMAT_XRGB8888,
+	DRM_FORMAT_RGB888, DRM_FORMAT_YUYV,
+	DRM_FORMAT_UYVY),
+};
+
+static const u32 *omap3_dispc_supported_color_modes[] = {
+	/* OMAP_DSS_GFX */
+	COLOR_ARRAY(
+	DRM_FORMAT_RGBX4444, DRM_FORMAT_ARGB4444,
+	DRM_FORMAT_RGB565, DRM_FORMAT_XRGB8888,
+	DRM_FORMAT_RGB888, DRM_FORMAT_ARGB8888,
+	DRM_FORMAT_RGBA8888, DRM_FORMAT_RGBX8888),
+
+	/* OMAP_DSS_VIDEO1 */
+	COLOR_ARRAY(
+	DRM_FORMAT_XRGB8888, DRM_FORMAT_RGB888,
+	DRM_FORMAT_RGBX4444, DRM_FORMAT_RGB565,
+	DRM_FORMAT_YUYV, DRM_FORMAT_UYVY),
+
+	/* OMAP_DSS_VIDEO2 */
+	COLOR_ARRAY(
+	DRM_FORMAT_RGBX4444, DRM_FORMAT_ARGB4444,
+	DRM_FORMAT_RGB565, DRM_FORMAT_XRGB8888,
+	DRM_FORMAT_RGB888, DRM_FORMAT_YUYV,
+	DRM_FORMAT_UYVY, DRM_FORMAT_ARGB8888,
+	DRM_FORMAT_RGBA8888, DRM_FORMAT_RGBX8888),
+};
+
+static const u32 *omap4_dispc_supported_color_modes[] = {
+	/* OMAP_DSS_GFX */
+	COLOR_ARRAY(
+	DRM_FORMAT_RGBX4444, DRM_FORMAT_ARGB4444,
+	DRM_FORMAT_RGB565, DRM_FORMAT_XRGB8888,
+	DRM_FORMAT_RGB888, DRM_FORMAT_ARGB8888,
+	DRM_FORMAT_RGBA8888, DRM_FORMAT_RGBX8888,
+	DRM_FORMAT_ARGB1555, DRM_FORMAT_XRGB4444,
+	DRM_FORMAT_RGBA4444, DRM_FORMAT_XRGB1555),
+
+	/* OMAP_DSS_VIDEO1 */
+	COLOR_ARRAY(
+	DRM_FORMAT_RGB565, DRM_FORMAT_RGBX4444,
+	DRM_FORMAT_YUYV, DRM_FORMAT_ARGB1555,
+	DRM_FORMAT_RGBA8888, DRM_FORMAT_NV12,
+	DRM_FORMAT_RGBA4444, DRM_FORMAT_XRGB8888,
+	DRM_FORMAT_RGB888, DRM_FORMAT_UYVY,
+	DRM_FORMAT_ARGB4444, DRM_FORMAT_XRGB1555,
+	DRM_FORMAT_ARGB8888, DRM_FORMAT_XRGB4444,
+	DRM_FORMAT_RGBX8888),
+
+       /* OMAP_DSS_VIDEO2 */
+	COLOR_ARRAY(
+	DRM_FORMAT_RGB565, DRM_FORMAT_RGBX4444,
+	DRM_FORMAT_YUYV, DRM_FORMAT_ARGB1555,
+	DRM_FORMAT_RGBA8888, DRM_FORMAT_NV12,
+	DRM_FORMAT_RGBA4444, DRM_FORMAT_XRGB8888,
+	DRM_FORMAT_RGB888, DRM_FORMAT_UYVY,
+	DRM_FORMAT_ARGB4444, DRM_FORMAT_XRGB1555,
+	DRM_FORMAT_ARGB8888, DRM_FORMAT_XRGB4444,
+	DRM_FORMAT_RGBX8888),
+
+	/* OMAP_DSS_VIDEO3 */
+	COLOR_ARRAY(
+	DRM_FORMAT_RGB565, DRM_FORMAT_RGBX4444,
+	DRM_FORMAT_YUYV, DRM_FORMAT_ARGB1555,
+	DRM_FORMAT_RGBA8888, DRM_FORMAT_NV12,
+	DRM_FORMAT_RGBA4444, DRM_FORMAT_XRGB8888,
+	DRM_FORMAT_RGB888, DRM_FORMAT_UYVY,
+	DRM_FORMAT_ARGB4444, DRM_FORMAT_XRGB1555,
+	DRM_FORMAT_ARGB8888, DRM_FORMAT_XRGB4444,
+	DRM_FORMAT_RGBX8888),
+
+	/* OMAP_DSS_WB */
+	COLOR_ARRAY(
+	DRM_FORMAT_RGB565, DRM_FORMAT_RGBX4444,
+	DRM_FORMAT_YUYV, DRM_FORMAT_ARGB1555,
+	DRM_FORMAT_RGBA8888, DRM_FORMAT_NV12,
+	DRM_FORMAT_RGBA4444, DRM_FORMAT_XRGB8888,
+	DRM_FORMAT_RGB888, DRM_FORMAT_UYVY,
+	DRM_FORMAT_ARGB4444, DRM_FORMAT_XRGB1555,
+	DRM_FORMAT_ARGB8888, DRM_FORMAT_XRGB4444,
+	DRM_FORMAT_RGBX8888),
+};
+
 static const struct dispc_features omap24xx_dispc_feats = {
 	.sw_start		=	5,
 	.fp_start		=	15,
@@ -3701,9 +4046,26 @@ static const struct dispc_features omap24xx_dispc_feats = {
 	.mgr_width_max		=	2048,
 	.mgr_height_max		=	2048,
 	.max_lcd_pclk		=	66500000,
+	.max_downscale		=	2,
+	/*
+	 * Assume the line width buffer to be 768 pixels as OMAP2 DISPC scaler
+	 * cannot scale an image width larger than 768.
+	 */
+	.max_line_width		=	768,
+	.min_pcd		=	2,
 	.calc_scaling		=	dispc_ovl_calc_scaling_24xx,
 	.calc_core_clk		=	calc_core_clk_24xx,
 	.num_fifos		=	3,
+	.features		=	omap2_dispc_features_list,
+	.num_features		=	ARRAY_SIZE(omap2_dispc_features_list),
+	.reg_fields		=	omap2_dispc_reg_fields,
+	.num_reg_fields		=	ARRAY_SIZE(omap2_dispc_reg_fields),
+	.overlay_caps		=	omap2_dispc_overlay_caps,
+	.supported_color_modes	=	omap2_dispc_supported_color_modes,
+	.num_mgrs		=	2,
+	.num_ovls		=	3,
+	.buffer_size_unit	=	1,
+	.burst_size_unit	=	8,
 	.no_framedone_tv	=	true,
 	.set_max_preload	=	false,
 	.last_pixel_inc_missing	=	true,
@@ -3722,9 +4084,22 @@ static const struct dispc_features omap34xx_rev1_0_dispc_feats = {
 	.mgr_height_max		=	2048,
 	.max_lcd_pclk		=	173000000,
 	.max_tv_pclk		=	59000000,
+	.max_downscale		=	4,
+	.max_line_width		=	1024,
+	.min_pcd		=	1,
 	.calc_scaling		=	dispc_ovl_calc_scaling_34xx,
 	.calc_core_clk		=	calc_core_clk_34xx,
 	.num_fifos		=	3,
+	.features		=	omap3_dispc_features_list,
+	.num_features		=	ARRAY_SIZE(omap3_dispc_features_list),
+	.reg_fields		=	omap3_dispc_reg_fields,
+	.num_reg_fields		=	ARRAY_SIZE(omap3_dispc_reg_fields),
+	.overlay_caps		=	omap3430_dispc_overlay_caps,
+	.supported_color_modes	=	omap3_dispc_supported_color_modes,
+	.num_mgrs		=	2,
+	.num_ovls		=	3,
+	.buffer_size_unit	=	1,
+	.burst_size_unit	=	8,
 	.no_framedone_tv	=	true,
 	.set_max_preload	=	false,
 	.last_pixel_inc_missing	=	true,
@@ -3743,9 +4118,90 @@ static const struct dispc_features omap34xx_rev3_0_dispc_feats = {
 	.mgr_height_max		=	2048,
 	.max_lcd_pclk		=	173000000,
 	.max_tv_pclk		=	59000000,
+	.max_downscale		=	4,
+	.max_line_width		=	1024,
+	.min_pcd		=	1,
+	.calc_scaling		=	dispc_ovl_calc_scaling_34xx,
+	.calc_core_clk		=	calc_core_clk_34xx,
+	.num_fifos		=	3,
+	.features		=	omap3_dispc_features_list,
+	.num_features		=	ARRAY_SIZE(omap3_dispc_features_list),
+	.reg_fields		=	omap3_dispc_reg_fields,
+	.num_reg_fields		=	ARRAY_SIZE(omap3_dispc_reg_fields),
+	.overlay_caps		=	omap3430_dispc_overlay_caps,
+	.supported_color_modes	=	omap3_dispc_supported_color_modes,
+	.num_mgrs		=	2,
+	.num_ovls		=	3,
+	.buffer_size_unit	=	1,
+	.burst_size_unit	=	8,
+	.no_framedone_tv	=	true,
+	.set_max_preload	=	false,
+	.last_pixel_inc_missing	=	true,
+};
+
+static const struct dispc_features omap36xx_dispc_feats = {
+	.sw_start		=	7,
+	.fp_start		=	19,
+	.bp_start		=	31,
+	.sw_max			=	256,
+	.vp_max			=	4095,
+	.hp_max			=	4096,
+	.mgr_width_start	=	10,
+	.mgr_height_start	=	26,
+	.mgr_width_max		=	2048,
+	.mgr_height_max		=	2048,
+	.max_lcd_pclk		=	173000000,
+	.max_tv_pclk		=	59000000,
+	.max_downscale		=	4,
+	.max_line_width		=	1024,
+	.min_pcd		=	1,
 	.calc_scaling		=	dispc_ovl_calc_scaling_34xx,
 	.calc_core_clk		=	calc_core_clk_34xx,
 	.num_fifos		=	3,
+	.features		=	omap3_dispc_features_list,
+	.num_features		=	ARRAY_SIZE(omap3_dispc_features_list),
+	.reg_fields		=	omap3_dispc_reg_fields,
+	.num_reg_fields		=	ARRAY_SIZE(omap3_dispc_reg_fields),
+	.overlay_caps		=	omap3630_dispc_overlay_caps,
+	.supported_color_modes	=	omap3_dispc_supported_color_modes,
+	.num_mgrs		=	2,
+	.num_ovls		=	3,
+	.buffer_size_unit	=	1,
+	.burst_size_unit	=	8,
+	.no_framedone_tv	=	true,
+	.set_max_preload	=	false,
+	.last_pixel_inc_missing	=	true,
+};
+
+static const struct dispc_features am43xx_dispc_feats = {
+	.sw_start		=	7,
+	.fp_start		=	19,
+	.bp_start		=	31,
+	.sw_max			=	256,
+	.vp_max			=	4095,
+	.hp_max			=	4096,
+	.mgr_width_start	=	10,
+	.mgr_height_start	=	26,
+	.mgr_width_max		=	2048,
+	.mgr_height_max		=	2048,
+	.max_lcd_pclk		=	173000000,
+	.max_tv_pclk		=	59000000,
+	.max_downscale		=	4,
+	.max_line_width		=	1024,
+	.min_pcd		=	1,
+	.calc_scaling		=	dispc_ovl_calc_scaling_34xx,
+	.calc_core_clk		=	calc_core_clk_34xx,
+	.num_fifos		=	3,
+	.features		=	am43xx_dispc_features_list,
+	.num_features		=	ARRAY_SIZE(am43xx_dispc_features_list),
+	.reg_fields		=	omap3_dispc_reg_fields,
+	.num_reg_fields		=	ARRAY_SIZE(omap3_dispc_reg_fields),
+	.overlay_caps		=	omap3430_dispc_overlay_caps,
+	.supported_color_modes	=	omap3_dispc_supported_color_modes,
+	.num_mgrs		=	1,
+	.num_ovls		=	3,
+	.buffer_size_unit	=	1,
+	.burst_size_unit	=	8,
 	.no_framedone_tv	=	true,
 	.set_max_preload	=	false,
 	.last_pixel_inc_missing	=	true,
@@ -3764,9 +4220,22 @@ static const struct dispc_features omap44xx_dispc_feats = {
 	.mgr_height_max		=	2048,
 	.max_lcd_pclk		=	170000000,
 	.max_tv_pclk		=	185625000,
+	.max_downscale		=	4,
+	.max_line_width		=	2048,
+	.min_pcd		=	1,
 	.calc_scaling		=	dispc_ovl_calc_scaling_44xx,
 	.calc_core_clk		=	calc_core_clk_44xx,
 	.num_fifos		=	5,
+	.features		=	omap4_dispc_features_list,
+	.num_features		=	ARRAY_SIZE(omap4_dispc_features_list),
+	.reg_fields		=	omap4_dispc_reg_fields,
+	.num_reg_fields		=	ARRAY_SIZE(omap4_dispc_reg_fields),
+	.overlay_caps		=	omap4_dispc_overlay_caps,
+	.supported_color_modes	=	omap4_dispc_supported_color_modes,
+	.num_mgrs		=	3,
+	.num_ovls		=	4,
+	.buffer_size_unit	=	16,
+	.burst_size_unit	=	16,
 	.gfx_fifo_workaround	=	true,
 	.set_max_preload	=	true,
 	.supports_sync_align	=	true,
@@ -3790,9 +4259,22 @@ static const struct dispc_features omap54xx_dispc_feats = {
 	.mgr_height_max		=	4096,
 	.max_lcd_pclk		=	170000000,
 	.max_tv_pclk		=	186000000,
+	.max_downscale		=	4,
+	.max_line_width		=	2048,
+	.min_pcd		=	1,
 	.calc_scaling		=	dispc_ovl_calc_scaling_44xx,
 	.calc_core_clk		=	calc_core_clk_44xx,
 	.num_fifos		=	5,
+	.features		=	omap5_dispc_features_list,
+	.num_features		=	ARRAY_SIZE(omap5_dispc_features_list),
+	.reg_fields		=	omap4_dispc_reg_fields,
+	.num_reg_fields		=	ARRAY_SIZE(omap4_dispc_reg_fields),
+	.overlay_caps		=	omap4_dispc_overlay_caps,
+	.supported_color_modes	=	omap4_dispc_supported_color_modes,
+	.num_mgrs		=	4,
+	.num_ovls		=	4,
+	.buffer_size_unit	=	16,
+	.burst_size_unit	=	16,
 	.gfx_fifo_workaround	=	true,
 	.mstandby_workaround	=	true,
 	.set_max_preload	=	true,
@@ -3804,54 +4286,6 @@ static const struct dispc_features omap54xx_dispc_feats = {
 	.has_gamma_i734_bug	=	true,
 };
 
-static int dispc_init_features(struct platform_device *pdev)
-{
-	const struct dispc_features *src;
-	struct dispc_features *dst;
-
-	dst = devm_kzalloc(&pdev->dev, sizeof(*dst), GFP_KERNEL);
-	if (!dst) {
-		dev_err(&pdev->dev, "Failed to allocate DISPC Features\n");
-		return -ENOMEM;
-	}
-
-	switch (omapdss_get_version()) {
-	case OMAPDSS_VER_OMAP24xx:
-		src = &omap24xx_dispc_feats;
-		break;
-
-	case OMAPDSS_VER_OMAP34xx_ES1:
-		src = &omap34xx_rev1_0_dispc_feats;
-		break;
-
-	case OMAPDSS_VER_OMAP34xx_ES3:
-	case OMAPDSS_VER_OMAP3630:
-	case OMAPDSS_VER_AM35xx:
-	case OMAPDSS_VER_AM43xx:
-		src = &omap34xx_rev3_0_dispc_feats;
-		break;
-
-	case OMAPDSS_VER_OMAP4430_ES1:
-	case OMAPDSS_VER_OMAP4430_ES2:
-	case OMAPDSS_VER_OMAP4:
-		src = &omap44xx_dispc_feats;
-		break;
-
-	case OMAPDSS_VER_OMAP5:
-	case OMAPDSS_VER_DRA7xx:
-		src = &omap54xx_dispc_feats;
-		break;
-
-	default:
-		return -ENODEV;
-	}
-
-	memcpy(dst, src, sizeof(*dst));
-	dispc.feat = dst;
-
-	return 0;
-}
-
 static irqreturn_t dispc_irq_handler(int irq, void *arg)
 {
 	if (!dispc.is_enabled)
@@ -4083,9 +4517,28 @@ static const struct dispc_ops dispc_ops = {
 };
 
 /* DISPC HW IP initialisation */
+static const struct of_device_id dispc_of_match[] = {
+	{ .compatible = "ti,omap2-dispc", .data = &omap24xx_dispc_feats },
+	{ .compatible = "ti,omap3-dispc", .data = &omap36xx_dispc_feats },
+	{ .compatible = "ti,omap4-dispc", .data = &omap44xx_dispc_feats },
+	{ .compatible = "ti,omap5-dispc", .data = &omap54xx_dispc_feats },
+	{ .compatible = "ti,dra7-dispc",  .data = &omap54xx_dispc_feats },
+	{},
+};
+
+static const struct soc_device_attribute dispc_soc_devices[] = {
+	{ .machine = "OMAP3[45]*",
+	  .revision = "ES[12].?",	.data = &omap34xx_rev1_0_dispc_feats },
+	{ .machine = "OMAP3[45]*",	.data = &omap34xx_rev3_0_dispc_feats },
+	{ .machine = "AM35*",		.data = &omap34xx_rev3_0_dispc_feats },
+	{ .machine = "AM43*",		.data = &am43xx_dispc_feats },
+	{ /* sentinel */ }
+};
+
 static int dispc_bind(struct device *dev, struct device *master, void *data)
 {
 	struct platform_device *pdev = to_platform_device(dev);
+	const struct soc_device_attribute *soc;
 	u32 rev;
 	int r = 0;
 	struct resource *dispc_mem;
@@ -4095,9 +4548,15 @@ static int dispc_bind(struct device *dev, struct device *master, void *data)
 
 	spin_lock_init(&dispc.control_lock);
 
-	r = dispc_init_features(dispc.pdev);
-	if (r)
-		return r;
+	/*
+	 * The OMAP3-based models can't be told apart using the compatible
+	 * string, use SoC device matching.
+	 */
+	soc = soc_device_match(dispc_soc_devices);
+	if (soc)
+		dispc.feat = soc->data;
+	else
+		dispc.feat = of_match_device(dispc_of_match, &pdev->dev)->data;
 
 	r = dispc_errata_i734_wa_init();
 	if (r)
@@ -4226,15 +4685,6 @@ static const struct dev_pm_ops dispc_pm_ops = {
 	.runtime_resume = dispc_runtime_resume,
 };
 
-static const struct of_device_id dispc_of_match[] = {
-	{ .compatible = "ti,omap2-dispc", },
-	{ .compatible = "ti,omap3-dispc", },
-	{ .compatible = "ti,omap4-dispc", },
-	{ .compatible = "ti,omap5-dispc", },
-	{ .compatible = "ti,dra7-dispc", },
-	{},
-};
-
 static struct platform_driver omap_dispchw_driver = {
 	.probe		= dispc_probe,
 	.remove         = dispc_remove,
diff --git a/drivers/gpu/drm/omapdrm/dss/dpi.c b/drivers/gpu/drm/omapdrm/dss/dpi.c
index 86dbb65a6c28..daf286fc8a40 100644
--- a/drivers/gpu/drm/omapdrm/dss/dpi.c
+++ b/drivers/gpu/drm/omapdrm/dss/dpi.c
@@ -32,13 +32,14 @@
 #include <linux/string.h>
 #include <linux/of.h>
 #include <linux/clk.h>
+#include <linux/sys_soc.h>
 
 #include "omapdss.h"
 #include "dss.h"
-#include "dss_features.h"
 
 struct dpi_data {
 	struct platform_device *pdev;
+	enum dss_model dss_model;
 
 	struct regulator *vdds_dsi_reg;
 	enum dss_clk_source clk_src;
@@ -99,25 +100,21 @@ static enum dss_clk_source dpi_get_clk_src_dra7xx(enum omap_channel channel)
 	return DSS_CLK_SRC_FCK;
 }
 
-static enum dss_clk_source dpi_get_clk_src(enum omap_channel channel)
+static enum dss_clk_source dpi_get_clk_src(struct dpi_data *dpi)
 {
+	enum omap_channel channel = dpi->output.dispc_channel;
+
 	/*
 	 * XXX we can't currently use DSI PLL for DPI with OMAP3, as the DSI PLL
 	 * would also be used for DISPC fclk. Meaning, when the DPI output is
 	 * disabled, DISPC clock will be disabled, and TV out will stop.
 	 */
-	switch (omapdss_get_version()) {
-	case OMAPDSS_VER_OMAP24xx:
-	case OMAPDSS_VER_OMAP34xx_ES1:
-	case OMAPDSS_VER_OMAP34xx_ES3:
-	case OMAPDSS_VER_OMAP3630:
-	case OMAPDSS_VER_AM35xx:
-	case OMAPDSS_VER_AM43xx:
+	switch (dpi->dss_model) {
+	case DSS_MODEL_OMAP2:
+	case DSS_MODEL_OMAP3:
 		return DSS_CLK_SRC_FCK;
 
-	case OMAPDSS_VER_OMAP4430_ES1:
-	case OMAPDSS_VER_OMAP4430_ES2:
-	case OMAPDSS_VER_OMAP4:
+	case DSS_MODEL_OMAP4:
 		switch (channel) {
 		case OMAP_DSS_CHANNEL_LCD:
 			return DSS_CLK_SRC_PLL1_1;
@@ -127,7 +124,7 @@ static enum dss_clk_source dpi_get_clk_src(enum omap_channel channel)
 			return DSS_CLK_SRC_FCK;
 		}
 
-	case OMAPDSS_VER_OMAP5:
+	case DSS_MODEL_OMAP5:
 		switch (channel) {
 		case OMAP_DSS_CHANNEL_LCD:
 			return DSS_CLK_SRC_PLL1_1;
@@ -138,7 +135,7 @@ static enum dss_clk_source dpi_get_clk_src(enum omap_channel channel)
 			return DSS_CLK_SRC_FCK;
 		}
 
-	case OMAPDSS_VER_DRA7xx:
+	case DSS_MODEL_DRA7:
 		return dpi_get_clk_src_dra7xx(channel);
 
 	default:
@@ -213,7 +210,7 @@ static bool dpi_calc_pll_cb(int n, int m, unsigned long fint,
 	ctx->pll_cinfo.clkdco = clkdco;
 
 	return dss_pll_hsdiv_calc_a(ctx->pll, clkdco,
-		ctx->pck_min, dss_feat_get_param_max(FEAT_PARAM_DSS_FCK),
+		ctx->pck_min, dss_get_max_fck_rate(),
 		dpi_calc_hsdiv_cb, ctx);
 }
 
@@ -403,19 +400,13 @@ static int dpi_display_enable(struct omap_dss_device *dssdev)
 
 	mutex_lock(&dpi->lock);
 
-	if (dss_has_feature(FEAT_DPI_USES_VDDS_DSI) && !dpi->vdds_dsi_reg) {
-		DSSERR("no VDSS_DSI regulator\n");
-		r = -ENODEV;
-		goto err_no_reg;
-	}
-
 	if (!out->dispc_channel_connected) {
 		DSSERR("failed to enable display: no output/manager\n");
 		r = -ENODEV;
 		goto err_no_out_mgr;
 	}
 
-	if (dss_has_feature(FEAT_DPI_USES_VDDS_DSI)) {
+	if (dpi->vdds_dsi_reg) {
 		r = regulator_enable(dpi->vdds_dsi_reg);
 		if (r)
 			goto err_reg_enable;
@@ -459,11 +450,10 @@ err_pll_init:
 err_src_sel:
 	dispc_runtime_put();
 err_get_dispc:
-	if (dss_has_feature(FEAT_DPI_USES_VDDS_DSI))
+	if (dpi->vdds_dsi_reg)
 		regulator_disable(dpi->vdds_dsi_reg);
 err_reg_enable:
 err_no_out_mgr:
-err_no_reg:
 	mutex_unlock(&dpi->lock);
 	return r;
 }
@@ -484,7 +474,7 @@ static void dpi_display_disable(struct omap_dss_device *dssdev)
 
 	dispc_runtime_put();
 
-	if (dss_has_feature(FEAT_DPI_USES_VDDS_DSI))
+	if (dpi->vdds_dsi_reg)
 		regulator_disable(dpi->vdds_dsi_reg);
 
 	mutex_unlock(&dpi->lock);
@@ -575,11 +565,21 @@ static int dpi_verify_pll(struct dss_pll *pll)
 	return 0;
 }
 
+static const struct soc_device_attribute dpi_soc_devices[] = {
+	{ .family = "OMAP3[456]*" },
+	{ .family = "[AD]M37*" },
+	{ /* sentinel */ }
+};
+
 static int dpi_init_regulator(struct dpi_data *dpi)
 {
 	struct regulator *vdds_dsi;
 
-	if (!dss_has_feature(FEAT_DPI_USES_VDDS_DSI))
+	/*
+	 * The DPI uses the DSI VDDS on OMAP34xx, OMAP35xx, OMAP36xx, AM37xx and
+	 * DM37xx only.
+	 */
+	if (!soc_device_match(dpi_soc_devices))
 		return 0;
 
 	if (dpi->vdds_dsi_reg)
@@ -604,7 +604,7 @@ static void dpi_init_pll(struct dpi_data *dpi)
 	if (dpi->pll)
 		return;
 
-	dpi->clk_src = dpi_get_clk_src(dpi->output.dispc_channel);
+	dpi->clk_src = dpi_get_clk_src(dpi);
 
 	pll = dss_pll_find_by_src(dpi->clk_src);
 	if (!pll)
@@ -624,18 +624,14 @@ static void dpi_init_pll(struct dpi_data *dpi)
  * the channel in some more dynamic manner, or get the channel as a user
  * parameter.
  */
-static enum omap_channel dpi_get_channel(int port_num)
+static enum omap_channel dpi_get_channel(struct dpi_data *dpi, int port_num)
 {
-	switch (omapdss_get_version()) {
-	case OMAPDSS_VER_OMAP24xx:
-	case OMAPDSS_VER_OMAP34xx_ES1:
-	case OMAPDSS_VER_OMAP34xx_ES3:
-	case OMAPDSS_VER_OMAP3630:
-	case OMAPDSS_VER_AM35xx:
-	case OMAPDSS_VER_AM43xx:
+	switch (dpi->dss_model) {
+	case DSS_MODEL_OMAP2:
+	case DSS_MODEL_OMAP3:
 		return OMAP_DSS_CHANNEL_LCD;
 
-	case OMAPDSS_VER_DRA7xx:
+	case DSS_MODEL_DRA7:
 		switch (port_num) {
 		case 2:
 			return OMAP_DSS_CHANNEL_LCD3;
@@ -646,12 +642,10 @@ static enum omap_channel dpi_get_channel(int port_num)
 			return OMAP_DSS_CHANNEL_LCD;
 		}
 
-	case OMAPDSS_VER_OMAP4430_ES1:
-	case OMAPDSS_VER_OMAP4430_ES2:
-	case OMAPDSS_VER_OMAP4:
+	case DSS_MODEL_OMAP4:
 		return OMAP_DSS_CHANNEL_LCD2;
 
-	case OMAPDSS_VER_OMAP5:
+	case DSS_MODEL_OMAP5:
 		return OMAP_DSS_CHANNEL_LCD3;
 
 	default:
@@ -716,10 +710,8 @@ static const struct omapdss_dpi_ops dpi_ops = {
 	.get_timings = dpi_get_timings,
 };
 
-static void dpi_init_output_port(struct platform_device *pdev,
-	struct device_node *port)
+static void dpi_init_output_port(struct dpi_data *dpi, struct device_node *port)
 {
-	struct dpi_data *dpi = port->data;
 	struct omap_dss_device *out = &dpi->output;
 	int r;
 	u32 port_num;
@@ -741,10 +733,10 @@ static void dpi_init_output_port(struct platform_device *pdev,
 		break;
 	}
 
-	out->dev = &pdev->dev;
+	out->dev = &dpi->pdev->dev;
 	out->id = OMAP_DSS_OUTPUT_DPI;
 	out->output_type = OMAP_DISPLAY_TYPE_DPI;
-	out->dispc_channel = dpi_get_channel(port_num);
+	out->dispc_channel = dpi_get_channel(dpi, port_num);
 	out->port_num = port_num;
 	out->ops.dpi = &dpi_ops;
 	out->owner = THIS_MODULE;
@@ -760,7 +752,8 @@ static void dpi_uninit_output_port(struct device_node *port)
 	omapdss_unregister_output(out);
 }
 
-int dpi_init_port(struct platform_device *pdev, struct device_node *port)
+int dpi_init_port(struct platform_device *pdev, struct device_node *port,
+		  enum dss_model dss_model)
 {
 	struct dpi_data *dpi;
 	struct device_node *ep;
@@ -786,11 +779,12 @@ int dpi_init_port(struct platform_device *pdev, struct device_node *port)
 	of_node_put(ep);
 
 	dpi->pdev = pdev;
+	dpi->dss_model = dss_model;
 	port->data = dpi;
 
 	mutex_init(&dpi->lock);
 
-	dpi_init_output_port(pdev, port);
+	dpi_init_output_port(dpi, port);
 
 	dpi->port_initialized = true;
 
diff --git a/drivers/gpu/drm/omapdrm/dss/dsi.c b/drivers/gpu/drm/omapdrm/dss/dsi.c
index 835f49004bc3..b56a05730314 100644
--- a/drivers/gpu/drm/omapdrm/dss/dsi.c
+++ b/drivers/gpu/drm/omapdrm/dss/dsi.c
@@ -20,6 +20,8 @@
 #define DSS_SUBSYS_NAME "DSI"
 
 #include <linux/kernel.h>
+#include <linux/mfd/syscon.h>
+#include <linux/regmap.h>
 #include <linux/io.h>
 #include <linux/clk.h>
 #include <linux/device.h>
@@ -42,12 +44,12 @@
 #include <linux/of_graph.h>
 #include <linux/of_platform.h>
 #include <linux/component.h>
+#include <linux/sys_soc.h>
 
 #include <video/mipi_display.h>
 
 #include "omapdss.h"
 #include "dss.h"
-#include "dss_features.h"
 
 #define DSI_CATCH_MISSING_TE
 
@@ -228,6 +230,12 @@ static int dsi_vc_send_null(struct omap_dss_device *dssdev, int channel);
 #define DSI_MAX_NR_ISRS                2
 #define DSI_MAX_NR_LANES	5
 
+enum dsi_model {
+	DSI_MODEL_OMAP3,
+	DSI_MODEL_OMAP4,
+	DSI_MODEL_OMAP5,
+};
+
 enum dsi_lane_function {
 	DSI_LANE_UNUSED	= 0,
 	DSI_LANE_CLK,
@@ -299,12 +307,36 @@ struct dsi_lp_clock_info {
 	u16 lp_clk_div;
 };
 
+struct dsi_module_id_data {
+	u32 address;
+	int id;
+};
+
+enum dsi_quirks {
+	DSI_QUIRK_PLL_PWR_BUG = (1 << 0),	/* DSI-PLL power command 0x3 is not working */
+	DSI_QUIRK_DCS_CMD_CONFIG_VC = (1 << 1),
+	DSI_QUIRK_VC_OCP_WIDTH = (1 << 2),
+	DSI_QUIRK_REVERSE_TXCLKESC = (1 << 3),
+	DSI_QUIRK_GNQ = (1 << 4),
+	DSI_QUIRK_PHY_DCC = (1 << 5),
+};
+
+struct dsi_of_data {
+	enum dsi_model model;
+	const struct dss_pll_hw *pll_hw;
+	const struct dsi_module_id_data *modules;
+	unsigned int max_fck_freq;
+	unsigned int max_pll_lpdiv;
+	enum dsi_quirks quirks;
+};
+
 struct dsi_data {
 	struct platform_device *pdev;
 	void __iomem *proto_base;
 	void __iomem *phy_base;
 	void __iomem *pll_base;
 
+	const struct dsi_of_data *data;
 	int module_id;
 
 	int irq;
@@ -312,6 +344,7 @@ struct dsi_data {
 	bool is_enabled;
 
 	struct clk *dss_clk;
+	struct regmap *syscon;
 
 	struct dispc_clock_info user_dispc_cinfo;
 	struct dss_pll_clock_info user_dsi_cinfo;
@@ -397,13 +430,6 @@ struct dsi_packet_sent_handler_data {
 	struct completion *completion;
 };
 
-struct dsi_module_id_data {
-	u32 address;
-	int id;
-};
-
-static const struct of_device_id dsi_of_match[];
-
 #ifdef DSI_PERF_MEASURE
 static bool dsi_perf;
 module_param(dsi_perf, bool, 0644);
@@ -1186,6 +1212,7 @@ static int dsi_regulator_init(struct platform_device *dsidev)
 
 static void _dsi_print_reset_status(struct platform_device *dsidev)
 {
+	struct dsi_data *dsi = dsi_get_dsidrv_data(dsidev);
 	u32 l;
 	int b0, b1, b2;
 
@@ -1194,7 +1221,7 @@ static void _dsi_print_reset_status(struct platform_device *dsidev)
 	 * I/O. */
 	l = dsi_read_reg(dsidev, DSI_DSIPHY_CFG5);
 
-	if (dss_has_feature(FEAT_DSI_REVERSE_TXCLKESC)) {
+	if (dsi->data->quirks & DSI_QUIRK_REVERSE_TXCLKESC) {
 		b0 = 28;
 		b1 = 27;
 		b2 = 26;
@@ -1297,7 +1324,7 @@ static int dsi_set_lp_clk_divisor(struct platform_device *dsidev)
 	unsigned long dsi_fclk;
 	unsigned lp_clk_div;
 	unsigned long lp_clk;
-	unsigned lpdiv_max = dss_feat_get_param_max(FEAT_PARAM_DSIPLL_LPDIV);
+	unsigned lpdiv_max = dsi->data->max_pll_lpdiv;
 
 
 	lp_clk_div = dsi->user_lp_cinfo.lp_clk_div;
@@ -1349,11 +1376,12 @@ enum dsi_pll_power_state {
 static int dsi_pll_power(struct platform_device *dsidev,
 		enum dsi_pll_power_state state)
 {
+	struct dsi_data *dsi = dsi_get_dsidrv_data(dsidev);
 	int t = 0;
 
 	/* DSI-PLL power command 0x3 is not working */
-	if (dss_has_feature(FEAT_DSI_PLL_PWR_BUG) &&
-			state == DSI_PLL_POWER_ON_DIV)
+	if ((dsi->data->quirks & DSI_QUIRK_PLL_PWR_BUG) &&
+	    state == DSI_PLL_POWER_ON_DIV)
 		state = DSI_PLL_POWER_ON_ALL;
 
 	/* PLL_PWR_CMD */
@@ -1373,11 +1401,12 @@ static int dsi_pll_power(struct platform_device *dsidev,
 }
 
 
-static void dsi_pll_calc_dsi_fck(struct dss_pll_clock_info *cinfo)
+static void dsi_pll_calc_dsi_fck(struct dsi_data *dsi,
+				 struct dss_pll_clock_info *cinfo)
 {
 	unsigned long max_dsi_fck;
 
-	max_dsi_fck = dss_feat_get_param_max(FEAT_PARAM_DSI_FCK);
+	max_dsi_fck = dsi->data->max_fck_freq;
 
 	cinfo->mX[HSDIV_DSI] = DIV_ROUND_UP(cinfo->clkdco, max_dsi_fck);
 	cinfo->clkout[HSDIV_DSI] = cinfo->clkdco / cinfo->mX[HSDIV_DSI];
@@ -1773,13 +1802,14 @@ static int dsi_cio_power(struct platform_device *dsidev,
 
 static unsigned dsi_get_line_buf_size(struct platform_device *dsidev)
 {
+	struct dsi_data *dsi = dsi_get_dsidrv_data(dsidev);
 	int val;
 
 	/* line buffer on OMAP3 is 1024 x 24bits */
 	/* XXX: for some reason using full buffer size causes
 	 * considerable TX slowdown with update sizes that fill the
 	 * whole buffer */
-	if (!dss_has_feature(FEAT_DSI_GNQ))
+	if (!(dsi->data->quirks & DSI_QUIRK_GNQ))
 		return 1023 * 3;
 
 	val = REG_GET(dsidev, DSI_GNQ, 14, 12); /* VP1_LINE_BUFFER_SIZE */
@@ -1872,6 +1902,7 @@ static inline unsigned ddr2ns(struct platform_device *dsidev, unsigned ddr)
 
 static void dsi_cio_timings(struct platform_device *dsidev)
 {
+	struct dsi_data *dsi = dsi_get_dsidrv_data(dsidev);
 	u32 r;
 	u32 ths_prepare, ths_prepare_ths_zero, ths_trail, ths_exit;
 	u32 tlpx_half, tclk_trail, tclk_zero;
@@ -1934,7 +1965,7 @@ static void dsi_cio_timings(struct platform_device *dsidev)
 	r = FLD_MOD(r, tclk_trail, 15, 8);
 	r = FLD_MOD(r, tclk_zero, 7, 0);
 
-	if (dss_has_feature(FEAT_DSI_PHY_DCC)) {
+	if (dsi->data->quirks & DSI_QUIRK_PHY_DCC) {
 		r = FLD_MOD(r, 0, 21, 21);	/* DCCEN = disable */
 		r = FLD_MOD(r, 1, 22, 22);	/* CLKINP_DIVBY2EN = enable */
 		r = FLD_MOD(r, 1, 23, 23);	/* CLKINP_SEL = enable */
@@ -2006,7 +2037,7 @@ static int dsi_cio_wait_tx_clk_esc_reset(struct platform_device *dsidev)
 	static const u8 offsets_new[] = { 24, 25, 26, 27, 28 };
 	const u8 *offsets;
 
-	if (dss_has_feature(FEAT_DSI_REVERSE_TXCLKESC))
+	if (dsi->data->quirks & DSI_QUIRK_REVERSE_TXCLKESC)
 		offsets = offsets_old;
 	else
 		offsets = offsets_new;
@@ -2060,6 +2091,83 @@ static unsigned dsi_get_lane_mask(struct platform_device *dsidev)
 	return mask;
 }
 
+/* OMAP4 CONTROL_DSIPHY */
+#define OMAP4_DSIPHY_SYSCON_OFFSET			0x78
+
+#define OMAP4_DSI2_LANEENABLE_SHIFT			29
+#define OMAP4_DSI2_LANEENABLE_MASK			(0x7 << 29)
+#define OMAP4_DSI1_LANEENABLE_SHIFT			24
+#define OMAP4_DSI1_LANEENABLE_MASK			(0x1f << 24)
+#define OMAP4_DSI1_PIPD_SHIFT				19
+#define OMAP4_DSI1_PIPD_MASK				(0x1f << 19)
+#define OMAP4_DSI2_PIPD_SHIFT				14
+#define OMAP4_DSI2_PIPD_MASK				(0x1f << 14)
+
+static int dsi_omap4_mux_pads(struct dsi_data *dsi, unsigned int lanes)
+{
+	u32 enable_mask, enable_shift;
+	u32 pipd_mask, pipd_shift;
+
+	if (dsi->module_id == 0) {
+		enable_mask = OMAP4_DSI1_LANEENABLE_MASK;
+		enable_shift = OMAP4_DSI1_LANEENABLE_SHIFT;
+		pipd_mask = OMAP4_DSI1_PIPD_MASK;
+		pipd_shift = OMAP4_DSI1_PIPD_SHIFT;
+	} else if (dsi->module_id == 1) {
+		enable_mask = OMAP4_DSI2_LANEENABLE_MASK;
+		enable_shift = OMAP4_DSI2_LANEENABLE_SHIFT;
+		pipd_mask = OMAP4_DSI2_PIPD_MASK;
+		pipd_shift = OMAP4_DSI2_PIPD_SHIFT;
+	} else {
+		return -ENODEV;
+	}
+
+	return regmap_update_bits(dsi->syscon, OMAP4_DSIPHY_SYSCON_OFFSET,
+		enable_mask | pipd_mask,
+		(lanes << enable_shift) | (lanes << pipd_shift));
+}
+
+/* OMAP5 CONTROL_DSIPHY */
+
+#define OMAP5_DSIPHY_SYSCON_OFFSET	0x74
+
+#define OMAP5_DSI1_LANEENABLE_SHIFT	24
+#define OMAP5_DSI2_LANEENABLE_SHIFT	19
+#define OMAP5_DSI_LANEENABLE_MASK	0x1f
+
+static int dsi_omap5_mux_pads(struct dsi_data *dsi, unsigned int lanes)
+{
+	u32 enable_shift;
+
+	if (dsi->module_id == 0)
+		enable_shift = OMAP5_DSI1_LANEENABLE_SHIFT;
+	else if (dsi->module_id == 1)
+		enable_shift = OMAP5_DSI2_LANEENABLE_SHIFT;
+	else
+		return -ENODEV;
+
+	return regmap_update_bits(dsi->syscon, OMAP5_DSIPHY_SYSCON_OFFSET,
+		OMAP5_DSI_LANEENABLE_MASK << enable_shift,
+		lanes << enable_shift);
+}
+
+static int dsi_enable_pads(struct dsi_data *dsi, unsigned int lane_mask)
+{
+	if (dsi->data->model == DSI_MODEL_OMAP4)
+		return dsi_omap4_mux_pads(dsi, lane_mask);
+	if (dsi->data->model == DSI_MODEL_OMAP5)
+		return dsi_omap5_mux_pads(dsi, lane_mask);
+	return 0;
+}
+
+static void dsi_disable_pads(struct dsi_data *dsi)
+{
+	if (dsi->data->model == DSI_MODEL_OMAP4)
+		dsi_omap4_mux_pads(dsi, 0);
+	else if (dsi->data->model == DSI_MODEL_OMAP5)
+		dsi_omap5_mux_pads(dsi, 0);
+}
+
 static int dsi_cio_init(struct platform_device *dsidev)
 {
 	struct dsi_data *dsi = dsi_get_dsidrv_data(dsidev);
@@ -2068,7 +2176,7 @@ static int dsi_cio_init(struct platform_device *dsidev)
 
 	DSSDBG("DSI CIO init starts");
 
-	r = dss_dsi_enable_pads(dsi->module_id, dsi_get_lane_mask(dsidev));
+	r = dsi_enable_pads(dsi, dsi_get_lane_mask(dsidev));
 	if (r)
 		return r;
 
@@ -2178,7 +2286,7 @@ err_cio_pwr:
 		dsi_cio_disable_lane_override(dsidev);
 err_scp_clk_dom:
 	dsi_disable_scp_clk(dsidev);
-	dss_dsi_disable_pads(dsi->module_id, dsi_get_lane_mask(dsidev));
+	dsi_disable_pads(dsi);
 	return r;
 }
 
@@ -2191,7 +2299,7 @@ static void dsi_cio_uninit(struct platform_device *dsidev)
 
 	dsi_cio_power(dsidev, DSI_COMPLEXIO_POWER_OFF);
 	dsi_disable_scp_clk(dsidev);
-	dss_dsi_disable_pads(dsi->module_id, dsi_get_lane_mask(dsidev));
+	dsi_disable_pads(dsi);
 }
 
 static void dsi_config_tx_fifo(struct platform_device *dsidev,
@@ -2439,7 +2547,7 @@ static void dsi_vc_initial_config(struct platform_device *dsidev, int channel)
 	r = FLD_MOD(r, 1, 7, 7); /* CS_TX_EN */
 	r = FLD_MOD(r, 1, 8, 8); /* ECC_TX_EN */
 	r = FLD_MOD(r, 0, 9, 9); /* MODE_SPEED, high speed on/off */
-	if (dss_has_feature(FEAT_DSI_VC_OCP_WIDTH))
+	if (dsi->data->quirks & DSI_QUIRK_VC_OCP_WIDTH)
 		r = FLD_MOD(r, 3, 11, 10);	/* OCP_WIDTH = 32 bit */
 
 	r = FLD_MOD(r, 4, 29, 27); /* DMA_RX_REQ_NB = no dma */
@@ -2474,7 +2582,7 @@ static int dsi_vc_config_source(struct platform_device *dsidev, int channel,
 	REG_FLD_MOD(dsidev, DSI_VC_CTRL(channel), source, 1, 1);
 
 	/* DCS_CMD_ENABLE */
-	if (dss_has_feature(FEAT_DSI_DCS_CMD_CONFIG_VC)) {
+	if (dsi->data->quirks & DSI_QUIRK_DCS_CMD_CONFIG_VC) {
 		bool enable = source == DSI_VC_SOURCE_VP;
 		REG_FLD_MOD(dsidev, DSI_VC_CTRL(channel), enable, 30, 30);
 	}
@@ -3607,7 +3715,7 @@ static int dsi_proto_config(struct platform_device *dsidev)
 	r = FLD_MOD(r, 0, 8, 8);	/* VP_CLK_POL */
 	r = FLD_MOD(r, 1, 14, 14);	/* TRIGGER_RESET_MODE */
 	r = FLD_MOD(r, 1, 19, 19);	/* EOT_ENABLE */
-	if (!dss_has_feature(FEAT_DSI_DCS_CMD_CONFIG_VC)) {
+	if (!(dsi->data->quirks & DSI_QUIRK_DCS_CMD_CONFIG_VC)) {
 		r = FLD_MOD(r, 1, 24, 24);	/* DCS_CMD_ENABLE */
 		/* DCS_CMD_CODE, 1=start, 0=continue */
 		r = FLD_MOD(r, 0, 25, 25);
@@ -4450,6 +4558,7 @@ static bool dsi_cm_calc_pll_cb(int n, int m, unsigned long fint,
 		unsigned long clkdco, void *data)
 {
 	struct dsi_clk_calc_ctx *ctx = data;
+	struct dsi_data *dsi = dsi_get_dsidrv_data(ctx->dsidev);
 
 	ctx->dsi_cinfo.n = n;
 	ctx->dsi_cinfo.m = m;
@@ -4457,7 +4566,7 @@ static bool dsi_cm_calc_pll_cb(int n, int m, unsigned long fint,
 	ctx->dsi_cinfo.clkdco = clkdco;
 
 	return dss_pll_hsdiv_calc_a(ctx->pll, clkdco, ctx->req_pck_min,
-			dss_feat_get_param_max(FEAT_PARAM_DSS_FCK),
+			dsi->data->max_fck_freq,
 			dsi_cm_calc_hsdiv_cb, ctx);
 }
 
@@ -4749,6 +4858,7 @@ static bool dsi_vm_calc_pll_cb(int n, int m, unsigned long fint,
 		unsigned long clkdco, void *data)
 {
 	struct dsi_clk_calc_ctx *ctx = data;
+	struct dsi_data *dsi = dsi_get_dsidrv_data(ctx->dsidev);
 
 	ctx->dsi_cinfo.n = n;
 	ctx->dsi_cinfo.m = m;
@@ -4756,7 +4866,7 @@ static bool dsi_vm_calc_pll_cb(int n, int m, unsigned long fint,
 	ctx->dsi_cinfo.clkdco = clkdco;
 
 	return dss_pll_hsdiv_calc_a(ctx->pll, clkdco, ctx->req_pck_min,
-			dss_feat_get_param_max(FEAT_PARAM_DSS_FCK),
+			dsi->data->max_fck_freq,
 			dsi_vm_calc_hsdiv_cb, ctx);
 }
 
@@ -4827,7 +4937,7 @@ static int dsi_set_config(struct omap_dss_device *dssdev,
 		goto err;
 	}
 
-	dsi_pll_calc_dsi_fck(&ctx.dsi_cinfo);
+	dsi_pll_calc_dsi_fck(dsi, &ctx.dsi_cinfo);
 
 	r = dsi_lp_clock_calc(ctx.dsi_cinfo.clkout[HSDIV_DSI],
 		config->lp_clk_min, config->lp_clk_max, &dsi->user_lp_cinfo);
@@ -4857,24 +4967,14 @@ err:
  * the channel in some more dynamic manner, or get the channel as a user
  * parameter.
  */
-static enum omap_channel dsi_get_channel(int module_id)
+static enum omap_channel dsi_get_channel(struct dsi_data *dsi)
 {
-	switch (omapdss_get_version()) {
-	case OMAPDSS_VER_OMAP24xx:
-	case OMAPDSS_VER_AM43xx:
-		DSSWARN("DSI not supported\n");
+	switch (dsi->data->model) {
+	case DSI_MODEL_OMAP3:
 		return OMAP_DSS_CHANNEL_LCD;
 
-	case OMAPDSS_VER_OMAP34xx_ES1:
-	case OMAPDSS_VER_OMAP34xx_ES3:
-	case OMAPDSS_VER_OMAP3630:
-	case OMAPDSS_VER_AM35xx:
-		return OMAP_DSS_CHANNEL_LCD;
-
-	case OMAPDSS_VER_OMAP4430_ES1:
-	case OMAPDSS_VER_OMAP4430_ES2:
-	case OMAPDSS_VER_OMAP4:
-		switch (module_id) {
+	case DSI_MODEL_OMAP4:
+		switch (dsi->module_id) {
 		case 0:
 			return OMAP_DSS_CHANNEL_LCD;
 		case 1:
@@ -4884,8 +4984,8 @@ static enum omap_channel dsi_get_channel(int module_id)
 			return OMAP_DSS_CHANNEL_LCD;
 		}
 
-	case OMAPDSS_VER_OMAP5:
-		switch (module_id) {
+	case DSI_MODEL_OMAP5:
+		switch (dsi->module_id) {
 		case 0:
 			return OMAP_DSS_CHANNEL_LCD;
 		case 1:
@@ -5065,7 +5165,7 @@ static void dsi_init_output(struct platform_device *dsidev)
 
 	out->output_type = OMAP_DISPLAY_TYPE_DSI;
 	out->name = dsi->module_id == 0 ? "dsi.0" : "dsi.1";
-	out->dispc_channel = dsi_get_channel(dsi->module_id);
+	out->dispc_channel = dsi_get_channel(dsi);
 	out->ops.dsi = &dsi_ops;
 	out->owner = THIS_MODULE;
 
@@ -5240,29 +5340,7 @@ static int dsi_init_pll_data(struct platform_device *dsidev)
 	pll->id = dsi->module_id == 0 ? DSS_PLL_DSI1 : DSS_PLL_DSI2;
 	pll->clkin = clk;
 	pll->base = dsi->pll_base;
-
-	switch (omapdss_get_version()) {
-	case OMAPDSS_VER_OMAP34xx_ES1:
-	case OMAPDSS_VER_OMAP34xx_ES3:
-	case OMAPDSS_VER_OMAP3630:
-	case OMAPDSS_VER_AM35xx:
-		pll->hw = &dss_omap3_dsi_pll_hw;
-		break;
-
-	case OMAPDSS_VER_OMAP4430_ES1:
-	case OMAPDSS_VER_OMAP4430_ES2:
-	case OMAPDSS_VER_OMAP4:
-		pll->hw = &dss_omap4_dsi_pll_hw;
-		break;
-
-	case OMAPDSS_VER_OMAP5:
-		pll->hw = &dss_omap5_dsi_pll_hw;
-		break;
-
-	default:
-		return -ENODEV;
-	}
-
+	pll->hw = dsi->data->pll_hw;
 	pll->ops = &dsi_pll_ops;
 
 	r = dss_pll_register(pll);
@@ -5273,9 +5351,74 @@ static int dsi_init_pll_data(struct platform_device *dsidev)
 }
 
 /* DSI1 HW IP initialisation */
+static const struct dsi_of_data dsi_of_data_omap34xx = {
+	.model = DSI_MODEL_OMAP3,
+	.pll_hw = &dss_omap3_dsi_pll_hw,
+	.modules = (const struct dsi_module_id_data[]) {
+		{ .address = 0x4804fc00, .id = 0, },
+		{ },
+	},
+	.max_fck_freq = 173000000,
+	.max_pll_lpdiv = (1 << 13) - 1,
+	.quirks = DSI_QUIRK_REVERSE_TXCLKESC,
+};
+
+static const struct dsi_of_data dsi_of_data_omap36xx = {
+	.model = DSI_MODEL_OMAP3,
+	.pll_hw = &dss_omap3_dsi_pll_hw,
+	.modules = (const struct dsi_module_id_data[]) {
+		{ .address = 0x4804fc00, .id = 0, },
+		{ },
+	},
+	.max_fck_freq = 173000000,
+	.max_pll_lpdiv = (1 << 13) - 1,
+	.quirks = DSI_QUIRK_PLL_PWR_BUG,
+};
+
+static const struct dsi_of_data dsi_of_data_omap4 = {
+	.model = DSI_MODEL_OMAP4,
+	.pll_hw = &dss_omap4_dsi_pll_hw,
+	.modules = (const struct dsi_module_id_data[]) {
+		{ .address = 0x58004000, .id = 0, },
+		{ .address = 0x58005000, .id = 1, },
+		{ },
+	},
+	.max_fck_freq = 170000000,
+	.max_pll_lpdiv = (1 << 13) - 1,
+	.quirks = DSI_QUIRK_DCS_CMD_CONFIG_VC | DSI_QUIRK_VC_OCP_WIDTH
+		| DSI_QUIRK_GNQ,
+};
+
+static const struct dsi_of_data dsi_of_data_omap5 = {
+	.model = DSI_MODEL_OMAP5,
+	.pll_hw = &dss_omap5_dsi_pll_hw,
+	.modules = (const struct dsi_module_id_data[]) {
+		{ .address = 0x58004000, .id = 0, },
+		{ .address = 0x58009000, .id = 1, },
+		{ },
+	},
+	.max_fck_freq = 209250000,
+	.max_pll_lpdiv = (1 << 13) - 1,
+	.quirks = DSI_QUIRK_DCS_CMD_CONFIG_VC | DSI_QUIRK_VC_OCP_WIDTH
+		| DSI_QUIRK_GNQ | DSI_QUIRK_PHY_DCC,
+};
+
+static const struct of_device_id dsi_of_match[] = {
+	{ .compatible = "ti,omap3-dsi", .data = &dsi_of_data_omap36xx, },
+	{ .compatible = "ti,omap4-dsi", .data = &dsi_of_data_omap4, },
+	{ .compatible = "ti,omap5-dsi", .data = &dsi_of_data_omap5, },
+	{},
+};
+
+static const struct soc_device_attribute dsi_soc_devices[] = {
+	{ .machine = "OMAP3[45]*",	.data = &dsi_of_data_omap34xx },
+	{ .machine = "AM35*",		.data = &dsi_of_data_omap34xx },
+	{ /* sentinel */ }
+};
 static int dsi_bind(struct device *dev, struct device *master, void *data)
 {
 	struct platform_device *dsidev = to_platform_device(dev);
+	const struct soc_device_attribute *soc;
 	const struct dsi_module_id_data *d;
 	u32 rev;
 	int r, i;
@@ -5339,7 +5482,13 @@ static int dsi_bind(struct device *dev, struct device *master, void *data)
 		return r;
 	}
 
-	d = of_match_node(dsi_of_match, dsidev->dev.of_node)->data;
+	soc = soc_device_match(dsi_soc_devices);
+	if (soc)
+		dsi->data = soc->data;
+	else
+		dsi->data = of_match_node(dsi_of_match, dev->of_node)->data;
+
+	d = dsi->data->modules;
 	while (d->address != 0 && d->address != dsi_mem->start)
 		d++;
 
@@ -5350,6 +5499,24 @@ static int dsi_bind(struct device *dev, struct device *master, void *data)
 
 	dsi->module_id = d->id;
 
+	if (dsi->data->model == DSI_MODEL_OMAP4 ||
+	    dsi->data->model == DSI_MODEL_OMAP5) {
+		struct device_node *np;
+
+		/*
+		 * The OMAP4/5 display DT bindings don't reference the padconf
+		 * syscon. Our only option to retrieve it is to find it by name.
+		 */
+		np = of_find_node_by_name(NULL,
+			dsi->data->model == DSI_MODEL_OMAP4 ?
+			"omap4_padconf_global" : "omap5_padconf_global");
+		if (!np)
+			return -ENODEV;
+
+		dsi->syscon = syscon_node_to_regmap(np);
+		of_node_put(np);
+	}
+
 	/* DSI VCs initialization */
 	for (i = 0; i < ARRAY_SIZE(dsi->vc); i++) {
 		dsi->vc[i].source = DSI_VC_SOURCE_L4;
@@ -5375,7 +5542,7 @@ static int dsi_bind(struct device *dev, struct device *master, void *data)
 
 	/* DSI on OMAP3 doesn't have register DSI_GNQ, set number
 	 * of data to 3 by default */
-	if (dss_has_feature(FEAT_DSI_GNQ))
+	if (dsi->data->quirks & DSI_QUIRK_GNQ)
 		/* NB_DATA_LANES */
 		dsi->num_lanes_supported = 1 + REG_GET(dsidev, DSI_GNQ, 11, 9);
 	else
@@ -5495,30 +5662,6 @@ static const struct dev_pm_ops dsi_pm_ops = {
 	.runtime_resume = dsi_runtime_resume,
 };
 
-static const struct dsi_module_id_data dsi_of_data_omap3[] = {
-	{ .address = 0x4804fc00, .id = 0, },
-	{ },
-};
-
-static const struct dsi_module_id_data dsi_of_data_omap4[] = {
-	{ .address = 0x58004000, .id = 0, },
-	{ .address = 0x58005000, .id = 1, },
-	{ },
-};
-
-static const struct dsi_module_id_data dsi_of_data_omap5[] = {
-	{ .address = 0x58004000, .id = 0, },
-	{ .address = 0x58009000, .id = 1, },
-	{ },
-};
-
-static const struct of_device_id dsi_of_match[] = {
-	{ .compatible = "ti,omap3-dsi", .data = dsi_of_data_omap3, },
-	{ .compatible = "ti,omap4-dsi", .data = dsi_of_data_omap4, },
-	{ .compatible = "ti,omap5-dsi", .data = dsi_of_data_omap5, },
-	{},
-};
-
 static struct platform_driver omap_dsihw_driver = {
 	.probe		= dsi_probe,
 	.remove		= dsi_remove,
diff --git a/drivers/gpu/drm/omapdrm/dss/dss.c b/drivers/gpu/drm/omapdrm/dss/dss.c
index 99e22ca972c7..d1755f12236b 100644
--- a/drivers/gpu/drm/omapdrm/dss/dss.c
+++ b/drivers/gpu/drm/omapdrm/dss/dss.c
@@ -22,6 +22,7 @@
 
 #define DSS_SUBSYS_NAME "DSS"
 
+#include <linux/debugfs.h>
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/io.h>
@@ -38,14 +39,15 @@
 #include <linux/mfd/syscon.h>
 #include <linux/regmap.h>
 #include <linux/of.h>
+#include <linux/of_device.h>
 #include <linux/of_graph.h>
 #include <linux/regulator/consumer.h>
 #include <linux/suspend.h>
 #include <linux/component.h>
+#include <linux/sys_soc.h>
 
 #include "omapdss.h"
 #include "dss.h"
-#include "dss_features.h"
 
 #define DSS_SZ_REGS			SZ_512
 
@@ -69,15 +71,24 @@ struct dss_reg {
 #define REG_FLD_MOD(idx, val, start, end) \
 	dss_write_reg(idx, FLD_MOD(dss_read_reg(idx), val, start, end))
 
+struct dss_ops {
+	int (*dpi_select_source)(int port, enum omap_channel channel);
+	int (*select_lcd_source)(enum omap_channel channel,
+		enum dss_clk_source clk_src);
+};
+
 struct dss_features {
+	enum dss_model model;
 	u8 fck_div_max;
+	unsigned int fck_freq_max;
 	u8 dss_fck_multiplier;
 	const char *parent_clk_name;
 	const enum omap_display_type *ports;
 	int num_ports;
-	int (*dpi_select_source)(int port, enum omap_channel channel);
-	int (*select_lcd_source)(enum omap_channel channel,
-		enum dss_clk_source clk_src);
+	const enum omap_dss_output_id *outputs;
+	const struct dss_ops *ops;
+	struct dss_reg_field dispc_clk_switch;
+	bool has_lcd_clk_src;
 };
 
 static struct {
@@ -139,8 +150,7 @@ static void dss_save_context(void)
 
 	SR(CONTROL);
 
-	if (dss_feat_get_supported_displays(OMAP_DSS_CHANNEL_LCD) &
-			OMAP_DISPLAY_TYPE_SDI) {
+	if (dss.feat->outputs[OMAP_DSS_CHANNEL_LCD] & OMAP_DSS_OUTPUT_SDI) {
 		SR(SDI_CONTROL);
 		SR(PLL_CONTROL);
 	}
@@ -159,8 +169,7 @@ static void dss_restore_context(void)
 
 	RR(CONTROL);
 
-	if (dss_feat_get_supported_displays(OMAP_DSS_CHANNEL_LCD) &
-			OMAP_DISPLAY_TYPE_SDI) {
+	if (dss.feat->outputs[OMAP_DSS_CHANNEL_LCD] & OMAP_DSS_OUTPUT_SDI) {
 		RR(SDI_CONTROL);
 		RR(PLL_CONTROL);
 	}
@@ -390,8 +399,7 @@ static void dss_dump_regs(struct seq_file *s)
 	DUMPREG(DSS_SYSSTATUS);
 	DUMPREG(DSS_CONTROL);
 
-	if (dss_feat_get_supported_displays(OMAP_DSS_CHANNEL_LCD) &
-			OMAP_DISPLAY_TYPE_SDI) {
+	if (dss.feat->outputs[OMAP_DSS_CHANNEL_LCD] & OMAP_DSS_OUTPUT_SDI) {
 		DUMPREG(DSS_SDI_CONTROL);
 		DUMPREG(DSS_PLL_CONTROL);
 		DUMPREG(DSS_SDI_STATUS);
@@ -419,14 +427,12 @@ static int dss_get_channel_index(enum omap_channel channel)
 static void dss_select_dispc_clk_source(enum dss_clk_source clk_src)
 {
 	int b;
-	u8 start, end;
 
 	/*
 	 * We always use PRCM clock as the DISPC func clock, except on DSS3,
 	 * where we don't have separate DISPC and LCD clock sources.
 	 */
-	if (WARN_ON(dss_has_feature(FEAT_LCD_CLK_SRC) &&
-		clk_src != DSS_CLK_SRC_FCK))
+	if (WARN_ON(dss.feat->has_lcd_clk_src && clk_src != DSS_CLK_SRC_FCK))
 		return;
 
 	switch (clk_src) {
@@ -444,9 +450,9 @@ static void dss_select_dispc_clk_source(enum dss_clk_source clk_src)
 		return;
 	}
 
-	dss_feat_get_reg_field(FEAT_REG_DISPC_CLK_SWITCH, &start, &end);
-
-	REG_FLD_MOD(DSS_CONTROL, b, start, end);	/* DISPC_CLK_SWITCH */
+	REG_FLD_MOD(DSS_CONTROL, b,			/* DISPC_CLK_SWITCH */
+		    dss.feat->dispc_clk_switch.start,
+		    dss.feat->dispc_clk_switch.end);
 
 	dss.dispc_clk_source = clk_src;
 }
@@ -570,13 +576,13 @@ void dss_select_lcd_clk_source(enum omap_channel channel,
 	int idx = dss_get_channel_index(channel);
 	int r;
 
-	if (!dss_has_feature(FEAT_LCD_CLK_SRC)) {
+	if (!dss.feat->has_lcd_clk_src) {
 		dss_select_dispc_clk_source(clk_src);
 		dss.lcd_clk_source[idx] = clk_src;
 		return;
 	}
 
-	r = dss.feat->select_lcd_source(channel, clk_src);
+	r = dss.feat->ops->select_lcd_source(channel, clk_src);
 	if (r)
 		return;
 
@@ -595,7 +601,7 @@ enum dss_clk_source dss_get_dsi_clk_source(int dsi_module)
 
 enum dss_clk_source dss_get_lcd_clk_source(enum omap_channel channel)
 {
-	if (dss_has_feature(FEAT_LCD_CLK_SRC)) {
+	if (dss.feat->has_lcd_clk_src) {
 		int idx = dss_get_channel_index(channel);
 		return dss.lcd_clk_source[idx];
 	} else {
@@ -615,7 +621,7 @@ bool dss_div_calc(unsigned long pck, unsigned long fck_min,
 	unsigned long prate;
 	unsigned m;
 
-	fck_hw_max = dss_feat_get_param_max(FEAT_PARAM_DSS_FCK);
+	fck_hw_max = dss.feat->fck_freq_max;
 
 	if (dss.parent_clk == NULL) {
 		unsigned pckd;
@@ -673,6 +679,16 @@ unsigned long dss_get_dispc_clk_rate(void)
 	return dss.dss_clk_rate;
 }
 
+unsigned long dss_get_max_fck_rate(void)
+{
+	return dss.feat->fck_freq_max;
+}
+
+enum omap_dss_output_id dss_get_supported_outputs(enum omap_channel channel)
+{
+	return dss.feat->outputs[channel];
+}
+
 static int dss_setup_default_clock(void)
 {
 	unsigned long max_dss_fck, prate;
@@ -680,7 +696,7 @@ static int dss_setup_default_clock(void)
 	unsigned fck_div;
 	int r;
 
-	max_dss_fck = dss_feat_get_param_max(FEAT_PARAM_DSS_FCK);
+	max_dss_fck = dss.feat->fck_freq_max;
 
 	if (dss.parent_clk == NULL) {
 		fck = clk_round_rate(dss.dss_clk, max_dss_fck);
@@ -721,27 +737,29 @@ void dss_set_dac_pwrdn_bgz(bool enable)
 
 void dss_select_hdmi_venc_clk_source(enum dss_hdmi_venc_clk_source_select src)
 {
-	enum omap_display_type dp;
-	dp = dss_feat_get_supported_displays(OMAP_DSS_CHANNEL_DIGIT);
+	enum omap_dss_output_id outputs;
+
+	outputs = dss.feat->outputs[OMAP_DSS_CHANNEL_DIGIT];
 
 	/* Complain about invalid selections */
-	WARN_ON((src == DSS_VENC_TV_CLK) && !(dp & OMAP_DISPLAY_TYPE_VENC));
-	WARN_ON((src == DSS_HDMI_M_PCLK) && !(dp & OMAP_DISPLAY_TYPE_HDMI));
+	WARN_ON((src == DSS_VENC_TV_CLK) && !(outputs & OMAP_DSS_OUTPUT_VENC));
+	WARN_ON((src == DSS_HDMI_M_PCLK) && !(outputs & OMAP_DSS_OUTPUT_HDMI));
 
 	/* Select only if we have options */
-	if ((dp & OMAP_DISPLAY_TYPE_VENC) && (dp & OMAP_DISPLAY_TYPE_HDMI))
+	if ((outputs & OMAP_DSS_OUTPUT_VENC) &&
+	    (outputs & OMAP_DSS_OUTPUT_HDMI))
 		REG_FLD_MOD(DSS_CONTROL, src, 15, 15);	/* VENC_HDMI_SWITCH */
 }
 
 enum dss_hdmi_venc_clk_source_select dss_get_hdmi_venc_clk_source(void)
 {
-	enum omap_display_type displays;
+	enum omap_dss_output_id outputs;
 
-	displays = dss_feat_get_supported_displays(OMAP_DSS_CHANNEL_DIGIT);
-	if ((displays & OMAP_DISPLAY_TYPE_HDMI) == 0)
+	outputs = dss.feat->outputs[OMAP_DSS_CHANNEL_DIGIT];
+	if ((outputs & OMAP_DSS_OUTPUT_HDMI) == 0)
 		return DSS_VENC_TV_CLK;
 
-	if ((displays & OMAP_DISPLAY_TYPE_VENC) == 0)
+	if ((outputs & OMAP_DSS_OUTPUT_VENC) == 0)
 		return DSS_HDMI_M_PCLK;
 
 	return REG_GET(DSS_CONTROL, 15, 15);
@@ -823,7 +841,7 @@ static int dss_dpi_select_source_dra7xx(int port, enum omap_channel channel)
 
 int dss_dpi_select_source(int port, enum omap_channel channel)
 {
-	return dss.feat->dpi_select_source(port, channel);
+	return dss.feat->ops->dpi_select_source(port, channel);
 }
 
 static int dss_get_clocks(void)
@@ -882,7 +900,7 @@ void dss_runtime_put(void)
 
 /* DEBUGFS */
 #if defined(CONFIG_OMAP2_DSS_DEBUGFS)
-void dss_debug_dump_clocks(struct seq_file *s)
+static void dss_debug_dump_clocks(struct seq_file *s)
 {
 	dss_dump_clocks(s);
 	dispc_dump_clocks(s);
@@ -890,8 +908,88 @@ void dss_debug_dump_clocks(struct seq_file *s)
 	dsi_dump_clocks(s);
 #endif
 }
-#endif
 
+static int dss_debug_show(struct seq_file *s, void *unused)
+{
+	void (*func)(struct seq_file *) = s->private;
+
+	func(s);
+	return 0;
+}
+
+static int dss_debug_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, dss_debug_show, inode->i_private);
+}
+
+static const struct file_operations dss_debug_fops = {
+	.open           = dss_debug_open,
+	.read           = seq_read,
+	.llseek         = seq_lseek,
+	.release        = single_release,
+};
+
+static struct dentry *dss_debugfs_dir;
+
+static int dss_initialize_debugfs(void)
+{
+	dss_debugfs_dir = debugfs_create_dir("omapdss", NULL);
+	if (IS_ERR(dss_debugfs_dir)) {
+		int err = PTR_ERR(dss_debugfs_dir);
+
+		dss_debugfs_dir = NULL;
+		return err;
+	}
+
+	debugfs_create_file("clk", S_IRUGO, dss_debugfs_dir,
+			&dss_debug_dump_clocks, &dss_debug_fops);
+
+	return 0;
+}
+
+static void dss_uninitialize_debugfs(void)
+{
+	if (dss_debugfs_dir)
+		debugfs_remove_recursive(dss_debugfs_dir);
+}
+
+int dss_debugfs_create_file(const char *name, void (*write)(struct seq_file *))
+{
+	struct dentry *d;
+
+	d = debugfs_create_file(name, S_IRUGO, dss_debugfs_dir,
+			write, &dss_debug_fops);
+
+	return PTR_ERR_OR_ZERO(d);
+}
+#else /* CONFIG_OMAP2_DSS_DEBUGFS */
+static inline int dss_initialize_debugfs(void)
+{
+	return 0;
+}
+static inline void dss_uninitialize_debugfs(void)
+{
+}
+#endif /* CONFIG_OMAP2_DSS_DEBUGFS */
+
+static const struct dss_ops dss_ops_omap2_omap3 = {
+	.dpi_select_source = &dss_dpi_select_source_omap2_omap3,
+};
+
+static const struct dss_ops dss_ops_omap4 = {
+	.dpi_select_source = &dss_dpi_select_source_omap4,
+	.select_lcd_source = &dss_lcd_clk_mux_omap4,
+};
+
+static const struct dss_ops dss_ops_omap5 = {
+	.dpi_select_source = &dss_dpi_select_source_omap5,
+	.select_lcd_source = &dss_lcd_clk_mux_omap5,
+};
+
+static const struct dss_ops dss_ops_dra7 = {
+	.dpi_select_source = &dss_dpi_select_source_dra7xx,
+	.select_lcd_source = &dss_lcd_clk_mux_dra7,
+};
 
 static const enum omap_display_type omap2plus_ports[] = {
 	OMAP_DISPLAY_TYPE_DPI,
@@ -908,130 +1006,168 @@ static const enum omap_display_type dra7xx_ports[] = {
 	OMAP_DISPLAY_TYPE_DPI,
 };
 
+static const enum omap_dss_output_id omap2_dss_supported_outputs[] = {
+	/* OMAP_DSS_CHANNEL_LCD */
+	OMAP_DSS_OUTPUT_DPI | OMAP_DSS_OUTPUT_DBI,
+
+	/* OMAP_DSS_CHANNEL_DIGIT */
+	OMAP_DSS_OUTPUT_VENC,
+};
+
+static const enum omap_dss_output_id omap3430_dss_supported_outputs[] = {
+	/* OMAP_DSS_CHANNEL_LCD */
+	OMAP_DSS_OUTPUT_DPI | OMAP_DSS_OUTPUT_DBI |
+	OMAP_DSS_OUTPUT_SDI | OMAP_DSS_OUTPUT_DSI1,
+
+	/* OMAP_DSS_CHANNEL_DIGIT */
+	OMAP_DSS_OUTPUT_VENC,
+};
+
+static const enum omap_dss_output_id omap3630_dss_supported_outputs[] = {
+	/* OMAP_DSS_CHANNEL_LCD */
+	OMAP_DSS_OUTPUT_DPI | OMAP_DSS_OUTPUT_DBI |
+	OMAP_DSS_OUTPUT_DSI1,
+
+	/* OMAP_DSS_CHANNEL_DIGIT */
+	OMAP_DSS_OUTPUT_VENC,
+};
+
+static const enum omap_dss_output_id am43xx_dss_supported_outputs[] = {
+	/* OMAP_DSS_CHANNEL_LCD */
+	OMAP_DSS_OUTPUT_DPI | OMAP_DSS_OUTPUT_DBI,
+};
+
+static const enum omap_dss_output_id omap4_dss_supported_outputs[] = {
+	/* OMAP_DSS_CHANNEL_LCD */
+	OMAP_DSS_OUTPUT_DBI | OMAP_DSS_OUTPUT_DSI1,
+
+	/* OMAP_DSS_CHANNEL_DIGIT */
+	OMAP_DSS_OUTPUT_VENC | OMAP_DSS_OUTPUT_HDMI,
+
+	/* OMAP_DSS_CHANNEL_LCD2 */
+	OMAP_DSS_OUTPUT_DPI | OMAP_DSS_OUTPUT_DBI |
+	OMAP_DSS_OUTPUT_DSI2,
+};
+
+static const enum omap_dss_output_id omap5_dss_supported_outputs[] = {
+	/* OMAP_DSS_CHANNEL_LCD */
+	OMAP_DSS_OUTPUT_DPI | OMAP_DSS_OUTPUT_DBI |
+	OMAP_DSS_OUTPUT_DSI1 | OMAP_DSS_OUTPUT_DSI2,
+
+	/* OMAP_DSS_CHANNEL_DIGIT */
+	OMAP_DSS_OUTPUT_HDMI,
+
+	/* OMAP_DSS_CHANNEL_LCD2 */
+	OMAP_DSS_OUTPUT_DPI | OMAP_DSS_OUTPUT_DBI |
+	OMAP_DSS_OUTPUT_DSI1,
+
+	/* OMAP_DSS_CHANNEL_LCD3 */
+	OMAP_DSS_OUTPUT_DPI | OMAP_DSS_OUTPUT_DBI |
+	OMAP_DSS_OUTPUT_DSI2,
+};
+
 static const struct dss_features omap24xx_dss_feats = {
+	.model			=	DSS_MODEL_OMAP2,
 	/*
 	 * fck div max is really 16, but the divider range has gaps. The range
 	 * from 1 to 6 has no gaps, so let's use that as a max.
 	 */
 	.fck_div_max		=	6,
+	.fck_freq_max		=	133000000,
 	.dss_fck_multiplier	=	2,
 	.parent_clk_name	=	"core_ck",
-	.dpi_select_source	=	&dss_dpi_select_source_omap2_omap3,
 	.ports			=	omap2plus_ports,
 	.num_ports		=	ARRAY_SIZE(omap2plus_ports),
+	.outputs		=	omap2_dss_supported_outputs,
+	.ops			=	&dss_ops_omap2_omap3,
+	.dispc_clk_switch	=	{ 0, 0 },
+	.has_lcd_clk_src	=	false,
 };
 
 static const struct dss_features omap34xx_dss_feats = {
+	.model			=	DSS_MODEL_OMAP3,
 	.fck_div_max		=	16,
+	.fck_freq_max		=	173000000,
 	.dss_fck_multiplier	=	2,
 	.parent_clk_name	=	"dpll4_ck",
-	.dpi_select_source	=	&dss_dpi_select_source_omap2_omap3,
 	.ports			=	omap34xx_ports,
+	.outputs		=	omap3430_dss_supported_outputs,
 	.num_ports		=	ARRAY_SIZE(omap34xx_ports),
+	.ops			=	&dss_ops_omap2_omap3,
+	.dispc_clk_switch	=	{ 0, 0 },
+	.has_lcd_clk_src	=	false,
 };
 
 static const struct dss_features omap3630_dss_feats = {
+	.model			=	DSS_MODEL_OMAP3,
 	.fck_div_max		=	32,
+	.fck_freq_max		=	173000000,
 	.dss_fck_multiplier	=	1,
 	.parent_clk_name	=	"dpll4_ck",
-	.dpi_select_source	=	&dss_dpi_select_source_omap2_omap3,
 	.ports			=	omap2plus_ports,
 	.num_ports		=	ARRAY_SIZE(omap2plus_ports),
+	.outputs		=	omap3630_dss_supported_outputs,
+	.ops			=	&dss_ops_omap2_omap3,
+	.dispc_clk_switch	=	{ 0, 0 },
+	.has_lcd_clk_src	=	false,
 };
 
 static const struct dss_features omap44xx_dss_feats = {
+	.model			=	DSS_MODEL_OMAP4,
 	.fck_div_max		=	32,
+	.fck_freq_max		=	186000000,
 	.dss_fck_multiplier	=	1,
 	.parent_clk_name	=	"dpll_per_x2_ck",
-	.dpi_select_source	=	&dss_dpi_select_source_omap4,
 	.ports			=	omap2plus_ports,
 	.num_ports		=	ARRAY_SIZE(omap2plus_ports),
-	.select_lcd_source	=	&dss_lcd_clk_mux_omap4,
+	.outputs		=	omap4_dss_supported_outputs,
+	.ops			=	&dss_ops_omap4,
+	.dispc_clk_switch	=	{ 9, 8 },
+	.has_lcd_clk_src	=	true,
 };
 
 static const struct dss_features omap54xx_dss_feats = {
+	.model			=	DSS_MODEL_OMAP5,
 	.fck_div_max		=	64,
+	.fck_freq_max		=	209250000,
 	.dss_fck_multiplier	=	1,
 	.parent_clk_name	=	"dpll_per_x2_ck",
-	.dpi_select_source	=	&dss_dpi_select_source_omap5,
 	.ports			=	omap2plus_ports,
 	.num_ports		=	ARRAY_SIZE(omap2plus_ports),
-	.select_lcd_source	=	&dss_lcd_clk_mux_omap5,
+	.outputs		=	omap5_dss_supported_outputs,
+	.ops			=	&dss_ops_omap5,
+	.dispc_clk_switch	=	{ 9, 7 },
+	.has_lcd_clk_src	=	true,
 };
 
 static const struct dss_features am43xx_dss_feats = {
+	.model			=	DSS_MODEL_OMAP3,
 	.fck_div_max		=	0,
+	.fck_freq_max		=	200000000,
 	.dss_fck_multiplier	=	0,
 	.parent_clk_name	=	NULL,
-	.dpi_select_source	=	&dss_dpi_select_source_omap2_omap3,
 	.ports			=	omap2plus_ports,
 	.num_ports		=	ARRAY_SIZE(omap2plus_ports),
+	.outputs		=	am43xx_dss_supported_outputs,
+	.ops			=	&dss_ops_omap2_omap3,
+	.dispc_clk_switch	=	{ 0, 0 },
+	.has_lcd_clk_src	=	true,
 };
 
 static const struct dss_features dra7xx_dss_feats = {
+	.model			=	DSS_MODEL_DRA7,
 	.fck_div_max		=	64,
+	.fck_freq_max		=	209250000,
 	.dss_fck_multiplier	=	1,
 	.parent_clk_name	=	"dpll_per_x2_ck",
-	.dpi_select_source	=	&dss_dpi_select_source_dra7xx,
 	.ports			=	dra7xx_ports,
 	.num_ports		=	ARRAY_SIZE(dra7xx_ports),
-	.select_lcd_source	=	&dss_lcd_clk_mux_dra7,
+	.outputs		=	omap5_dss_supported_outputs,
+	.ops			=	&dss_ops_dra7,
+	.dispc_clk_switch	=	{ 9, 7 },
+	.has_lcd_clk_src	=	true,
 };
 
-static int dss_init_features(struct platform_device *pdev)
-{
-	const struct dss_features *src;
-	struct dss_features *dst;
-
-	dst = devm_kzalloc(&pdev->dev, sizeof(*dst), GFP_KERNEL);
-	if (!dst) {
-		dev_err(&pdev->dev, "Failed to allocate local DSS Features\n");
-		return -ENOMEM;
-	}
-
-	switch (omapdss_get_version()) {
-	case OMAPDSS_VER_OMAP24xx:
-		src = &omap24xx_dss_feats;
-		break;
-
-	case OMAPDSS_VER_OMAP34xx_ES1:
-	case OMAPDSS_VER_OMAP34xx_ES3:
-	case OMAPDSS_VER_AM35xx:
-		src = &omap34xx_dss_feats;
-		break;
-
-	case OMAPDSS_VER_OMAP3630:
-		src = &omap3630_dss_feats;
-		break;
-
-	case OMAPDSS_VER_OMAP4430_ES1:
-	case OMAPDSS_VER_OMAP4430_ES2:
-	case OMAPDSS_VER_OMAP4:
-		src = &omap44xx_dss_feats;
-		break;
-
-	case OMAPDSS_VER_OMAP5:
-		src = &omap54xx_dss_feats;
-		break;
-
-	case OMAPDSS_VER_AM43xx:
-		src = &am43xx_dss_feats;
-		break;
-
-	case OMAPDSS_VER_DRA7xx:
-		src = &dra7xx_dss_feats;
-		break;
-
-	default:
-		return -ENODEV;
-	}
-
-	memcpy(dst, src, sizeof(*dst));
-	dss.feat = dst;
-
-	return 0;
-}
-
 static int dss_init_ports(struct platform_device *pdev)
 {
 	struct device_node *parent = pdev->dev.of_node;
@@ -1045,7 +1181,7 @@ static int dss_init_ports(struct platform_device *pdev)
 
 		switch (dss.feat->ports[i]) {
 		case OMAP_DISPLAY_TYPE_DPI:
-			dpi_init_port(pdev, port);
+			dpi_init_port(pdev, port, dss.feat->model);
 			break;
 		case OMAP_DISPLAY_TYPE_SDI:
 			sdi_init_port(pdev, port);
@@ -1144,6 +1280,23 @@ static int dss_video_pll_probe(struct platform_device *pdev)
 }
 
 /* DSS HW IP initialisation */
+static const struct of_device_id dss_of_match[] = {
+	{ .compatible = "ti,omap2-dss", .data = &omap24xx_dss_feats },
+	{ .compatible = "ti,omap3-dss", .data = &omap3630_dss_feats },
+	{ .compatible = "ti,omap4-dss", .data = &omap44xx_dss_feats },
+	{ .compatible = "ti,omap5-dss", .data = &omap54xx_dss_feats },
+	{ .compatible = "ti,dra7-dss",  .data = &dra7xx_dss_feats },
+	{},
+};
+MODULE_DEVICE_TABLE(of, dss_of_match);
+
+static const struct soc_device_attribute dss_soc_devices[] = {
+	{ .machine = "OMAP3430/3530", .data = &omap34xx_dss_feats },
+	{ .machine = "AM35??",        .data = &omap34xx_dss_feats },
+	{ .family  = "AM43xx",        .data = &am43xx_dss_feats },
+	{ /* sentinel */ }
+};
+
 static int dss_bind(struct device *dev)
 {
 	struct platform_device *pdev = to_platform_device(dev);
@@ -1151,12 +1304,6 @@ static int dss_bind(struct device *dev)
 	u32 rev;
 	int r;
 
-	dss.pdev = pdev;
-
-	r = dss_init_features(dss.pdev);
-	if (r)
-		return r;
-
 	dss_mem = platform_get_resource(dss.pdev, IORESOURCE_MEM, 0);
 	dss.base = devm_ioremap_resource(&pdev->dev, dss_mem);
 	if (IS_ERR(dss.base))
@@ -1288,15 +1435,34 @@ static int dss_add_child_component(struct device *dev, void *data)
 
 static int dss_probe(struct platform_device *pdev)
 {
+	const struct soc_device_attribute *soc;
 	struct component_match *match = NULL;
 	int r;
 
+	dss.pdev = pdev;
+
+	/*
+	 * The various OMAP3-based SoCs can't be told apart using the compatible
+	 * string, use SoC device matching.
+	 */
+	soc = soc_device_match(dss_soc_devices);
+	if (soc)
+		dss.feat = soc->data;
+	else
+		dss.feat = of_match_device(dss_of_match, &pdev->dev)->data;
+
+	r = dss_initialize_debugfs();
+	if (r)
+		return r;
+
 	/* add all the child devices as components */
 	device_for_each_child(&pdev->dev, &match, dss_add_child_component);
 
 	r = component_master_add_with_match(&pdev->dev, &dss_component_ops, match);
-	if (r)
+	if (r) {
+		dss_uninitialize_debugfs();
 		return r;
+	}
 
 	return 0;
 }
@@ -1304,9 +1470,27 @@ static int dss_probe(struct platform_device *pdev)
 static int dss_remove(struct platform_device *pdev)
 {
 	component_master_del(&pdev->dev, &dss_component_ops);
+
+	dss_uninitialize_debugfs();
+
 	return 0;
 }
 
+static void dss_shutdown(struct platform_device *pdev)
+{
+	struct omap_dss_device *dssdev = NULL;
+
+	DSSDBG("shutdown\n");
+
+	for_each_dss_dev(dssdev) {
+		if (!dssdev->driver)
+			continue;
+
+		if (dssdev->state == OMAP_DSS_DISPLAY_ACTIVE)
+			dssdev->driver->disable(dssdev);
+	}
+}
+
 static int dss_runtime_suspend(struct device *dev)
 {
 	dss_save_context();
@@ -1343,20 +1527,10 @@ static const struct dev_pm_ops dss_pm_ops = {
 	.runtime_resume = dss_runtime_resume,
 };
 
-static const struct of_device_id dss_of_match[] = {
-	{ .compatible = "ti,omap2-dss", },
-	{ .compatible = "ti,omap3-dss", },
-	{ .compatible = "ti,omap4-dss", },
-	{ .compatible = "ti,omap5-dss", },
-	{ .compatible = "ti,dra7-dss", },
-	{},
-};
-
-MODULE_DEVICE_TABLE(of, dss_of_match);
-
 static struct platform_driver omap_dsshw_driver = {
 	.probe		= dss_probe,
 	.remove		= dss_remove,
+	.shutdown	= dss_shutdown,
 	.driver         = {
 		.name   = "omapdss_dss",
 		.pm	= &dss_pm_ops,
diff --git a/drivers/gpu/drm/omapdrm/dss/dss.h b/drivers/gpu/drm/omapdrm/dss/dss.h
index 8dbf35f3ab23..ed465572491e 100644
--- a/drivers/gpu/drm/omapdrm/dss/dss.h
+++ b/drivers/gpu/drm/omapdrm/dss/dss.h
@@ -27,6 +27,9 @@
 
 #include "omapdss.h"
 
+#define MAX_DSS_LCD_MANAGERS	3
+#define MAX_NUM_DSI		2
+
 #ifdef pr_fmt
 #undef pr_fmt
 #endif
@@ -72,6 +75,14 @@
 #define FLD_MOD(orig, val, start, end) \
 	(((orig) & ~FLD_MASK(start, end)) | FLD_VAL(val, start, end))
 
+enum dss_model {
+	DSS_MODEL_OMAP2,
+	DSS_MODEL_OMAP3,
+	DSS_MODEL_OMAP4,
+	DSS_MODEL_OMAP5,
+	DSS_MODEL_DRA7,
+};
+
 enum dss_io_pad_mode {
 	DSS_IO_PAD_MODE_RESET,
 	DSS_IO_PAD_MODE_RFBI,
@@ -174,6 +185,9 @@ struct dss_pll_hw {
 	bool has_freqsel;
 	bool has_selfreqdco;
 	bool has_refsel;
+
+	/* DRA7 errata i886: use high N & M to avoid jitter */
+	bool errata_i886;
 };
 
 struct dss_pll {
@@ -192,6 +206,11 @@ struct dss_pll {
 	struct dss_pll_clock_info cinfo;
 };
 
+/* Defines a generic omap register field */
+struct dss_reg_field {
+	u8 start, end;
+};
+
 struct dispc_clock_info {
 	/* rates that we get with dividers below */
 	unsigned long lck;
@@ -219,10 +238,11 @@ struct seq_file;
 struct platform_device;
 
 /* core */
-int dss_dsi_enable_pads(int dsi_id, unsigned lane_mask);
-void dss_dsi_disable_pads(int dsi_id, unsigned lane_mask);
-int dss_set_min_bus_tput(struct device *dev, unsigned long tput);
-int dss_debugfs_create_file(const char *name, void (*write)(struct seq_file *));
+static inline int dss_set_min_bus_tput(struct device *dev, unsigned long tput)
+{
+	/* To be implemented when the OMAP platform will provide this feature */
+	return 0;
+}
 
 static inline bool dss_mgr_is_lcd(enum omap_channel id)
 {
@@ -234,6 +254,16 @@ static inline bool dss_mgr_is_lcd(enum omap_channel id)
 }
 
 /* DSS */
+#if defined(CONFIG_OMAP2_DSS_DEBUGFS)
+int dss_debugfs_create_file(const char *name, void (*write)(struct seq_file *));
+#else
+static inline int dss_debugfs_create_file(const char *name,
+					  void (*write)(struct seq_file *))
+{
+	return 0;
+}
+#endif /* CONFIG_OMAP2_DSS_DEBUGFS */
+
 int dss_init_platform_driver(void) __init;
 void dss_uninit_platform_driver(void);
 
@@ -241,6 +271,8 @@ int dss_runtime_get(void);
 void dss_runtime_put(void);
 
 unsigned long dss_get_dispc_clk_rate(void);
+unsigned long dss_get_max_fck_rate(void);
+enum omap_dss_output_id dss_get_supported_outputs(enum omap_channel channel);
 int dss_dpi_select_source(int port, enum omap_channel channel);
 void dss_select_hdmi_venc_clk_source(enum dss_hdmi_venc_clk_source_select);
 enum dss_hdmi_venc_clk_source_select dss_get_hdmi_venc_clk_source(void);
@@ -252,10 +284,6 @@ struct dss_pll *dss_video_pll_init(struct platform_device *pdev, int id,
 	struct regulator *regulator);
 void dss_video_pll_uninit(struct dss_pll *pll);
 
-#if defined(CONFIG_OMAP2_DSS_DEBUGFS)
-void dss_debug_dump_clocks(struct seq_file *s);
-#endif
-
 void dss_ctrl_pll_enable(enum dss_pll_id pll_id, bool enable);
 
 void dss_sdi_init(int datapairs);
@@ -312,11 +340,12 @@ void dsi_irq_handler(void);
 
 /* DPI */
 #ifdef CONFIG_OMAP2_DSS_DPI
-int dpi_init_port(struct platform_device *pdev, struct device_node *port);
+int dpi_init_port(struct platform_device *pdev, struct device_node *port,
+		  enum dss_model dss_model);
 void dpi_uninit_port(struct device_node *port);
 #else
 static inline int dpi_init_port(struct platform_device *pdev,
-		struct device_node *port)
+		struct device_node *port, enum dss_model dss_model)
 {
 	return 0;
 }
diff --git a/drivers/gpu/drm/omapdrm/dss/dss_features.c b/drivers/gpu/drm/omapdrm/dss/dss_features.c
deleted file mode 100644
index 0e599710dd95..000000000000
--- a/drivers/gpu/drm/omapdrm/dss/dss_features.c
+++ /dev/null
@@ -1,905 +0,0 @@
-/*
- * linux/drivers/video/omap2/dss/dss_features.c
- *
- * Copyright (C) 2010 Texas Instruments
- * Author: Archit Taneja <archit@ti.com>
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License version 2 as published by
- * the Free Software Foundation.
- *
- * This program is distributed in the hope that it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along with
- * this program.  If not, see <http://www.gnu.org/licenses/>.
- */
-
-#include <linux/kernel.h>
-#include <linux/module.h>
-#include <linux/types.h>
-#include <linux/err.h>
-#include <linux/slab.h>
-#include <drm/drm_fourcc.h>
-
-#include "omapdss.h"
-#include "dss.h"
-#include "dss_features.h"
-
-/* Defines a generic omap register field */
-struct dss_reg_field {
-	u8 start, end;
-};
-
-struct dss_param_range {
-	int min, max;
-};
-
-struct omap_dss_features {
-	const struct dss_reg_field *reg_fields;
-	const int num_reg_fields;
-
-	const enum dss_feat_id *features;
-	const int num_features;
-
-	const int num_mgrs;
-	const int num_ovls;
-	const enum omap_display_type *supported_displays;
-	const enum omap_dss_output_id *supported_outputs;
-	const u32 **supported_color_modes;
-	const enum omap_overlay_caps *overlay_caps;
-	const struct dss_param_range *dss_params;
-
-	const u32 buffer_size_unit;
-	const u32 burst_size_unit;
-};
-
-/* This struct is assigned to one of the below during initialization */
-static const struct omap_dss_features *omap_current_dss_features;
-
-static const struct dss_reg_field omap2_dss_reg_fields[] = {
-	[FEAT_REG_FIRHINC]			= { 11, 0 },
-	[FEAT_REG_FIRVINC]			= { 27, 16 },
-	[FEAT_REG_FIFOLOWTHRESHOLD]		= { 8, 0 },
-	[FEAT_REG_FIFOHIGHTHRESHOLD]		= { 24, 16 },
-	[FEAT_REG_FIFOSIZE]			= { 8, 0 },
-	[FEAT_REG_HORIZONTALACCU]		= { 9, 0 },
-	[FEAT_REG_VERTICALACCU]			= { 25, 16 },
-	[FEAT_REG_DISPC_CLK_SWITCH]		= { 0, 0 },
-};
-
-static const struct dss_reg_field omap3_dss_reg_fields[] = {
-	[FEAT_REG_FIRHINC]			= { 12, 0 },
-	[FEAT_REG_FIRVINC]			= { 28, 16 },
-	[FEAT_REG_FIFOLOWTHRESHOLD]		= { 11, 0 },
-	[FEAT_REG_FIFOHIGHTHRESHOLD]		= { 27, 16 },
-	[FEAT_REG_FIFOSIZE]			= { 10, 0 },
-	[FEAT_REG_HORIZONTALACCU]		= { 9, 0 },
-	[FEAT_REG_VERTICALACCU]			= { 25, 16 },
-	[FEAT_REG_DISPC_CLK_SWITCH]		= { 0, 0 },
-};
-
-static const struct dss_reg_field am43xx_dss_reg_fields[] = {
-	[FEAT_REG_FIRHINC]			= { 12, 0 },
-	[FEAT_REG_FIRVINC]			= { 28, 16 },
-	[FEAT_REG_FIFOLOWTHRESHOLD]	= { 11, 0 },
-	[FEAT_REG_FIFOHIGHTHRESHOLD]		= { 27, 16 },
-	[FEAT_REG_FIFOSIZE]		= { 10, 0 },
-	[FEAT_REG_HORIZONTALACCU]		= { 9, 0 },
-	[FEAT_REG_VERTICALACCU]			= { 25, 16 },
-	[FEAT_REG_DISPC_CLK_SWITCH]		= { 0, 0 },
-};
-
-static const struct dss_reg_field omap4_dss_reg_fields[] = {
-	[FEAT_REG_FIRHINC]			= { 12, 0 },
-	[FEAT_REG_FIRVINC]			= { 28, 16 },
-	[FEAT_REG_FIFOLOWTHRESHOLD]		= { 15, 0 },
-	[FEAT_REG_FIFOHIGHTHRESHOLD]		= { 31, 16 },
-	[FEAT_REG_FIFOSIZE]			= { 15, 0 },
-	[FEAT_REG_HORIZONTALACCU]		= { 10, 0 },
-	[FEAT_REG_VERTICALACCU]			= { 26, 16 },
-	[FEAT_REG_DISPC_CLK_SWITCH]		= { 9, 8 },
-};
-
-static const struct dss_reg_field omap5_dss_reg_fields[] = {
-	[FEAT_REG_FIRHINC]			= { 12, 0 },
-	[FEAT_REG_FIRVINC]			= { 28, 16 },
-	[FEAT_REG_FIFOLOWTHRESHOLD]		= { 15, 0 },
-	[FEAT_REG_FIFOHIGHTHRESHOLD]		= { 31, 16 },
-	[FEAT_REG_FIFOSIZE]			= { 15, 0 },
-	[FEAT_REG_HORIZONTALACCU]		= { 10, 0 },
-	[FEAT_REG_VERTICALACCU]			= { 26, 16 },
-	[FEAT_REG_DISPC_CLK_SWITCH]		= { 9, 7 },
-};
-
-static const enum omap_display_type omap2_dss_supported_displays[] = {
-	/* OMAP_DSS_CHANNEL_LCD */
-	OMAP_DISPLAY_TYPE_DPI | OMAP_DISPLAY_TYPE_DBI,
-
-	/* OMAP_DSS_CHANNEL_DIGIT */
-	OMAP_DISPLAY_TYPE_VENC,
-};
-
-static const enum omap_display_type omap3430_dss_supported_displays[] = {
-	/* OMAP_DSS_CHANNEL_LCD */
-	OMAP_DISPLAY_TYPE_DPI | OMAP_DISPLAY_TYPE_DBI |
-	OMAP_DISPLAY_TYPE_SDI | OMAP_DISPLAY_TYPE_DSI,
-
-	/* OMAP_DSS_CHANNEL_DIGIT */
-	OMAP_DISPLAY_TYPE_VENC,
-};
-
-static const enum omap_display_type omap3630_dss_supported_displays[] = {
-	/* OMAP_DSS_CHANNEL_LCD */
-	OMAP_DISPLAY_TYPE_DPI | OMAP_DISPLAY_TYPE_DBI |
-	OMAP_DISPLAY_TYPE_DSI,
-
-	/* OMAP_DSS_CHANNEL_DIGIT */
-	OMAP_DISPLAY_TYPE_VENC,
-};
-
-static const enum omap_display_type am43xx_dss_supported_displays[] = {
-	/* OMAP_DSS_CHANNEL_LCD */
-	OMAP_DISPLAY_TYPE_DPI | OMAP_DISPLAY_TYPE_DBI,
-};
-
-static const enum omap_display_type omap4_dss_supported_displays[] = {
-	/* OMAP_DSS_CHANNEL_LCD */
-	OMAP_DISPLAY_TYPE_DBI | OMAP_DISPLAY_TYPE_DSI,
-
-	/* OMAP_DSS_CHANNEL_DIGIT */
-	OMAP_DISPLAY_TYPE_VENC | OMAP_DISPLAY_TYPE_HDMI,
-
-	/* OMAP_DSS_CHANNEL_LCD2 */
-	OMAP_DISPLAY_TYPE_DPI | OMAP_DISPLAY_TYPE_DBI |
-	OMAP_DISPLAY_TYPE_DSI,
-};
-
-static const enum omap_display_type omap5_dss_supported_displays[] = {
-	/* OMAP_DSS_CHANNEL_LCD */
-	OMAP_DISPLAY_TYPE_DPI | OMAP_DISPLAY_TYPE_DBI |
-	OMAP_DISPLAY_TYPE_DSI,
-
-	/* OMAP_DSS_CHANNEL_DIGIT */
-	OMAP_DISPLAY_TYPE_HDMI | OMAP_DISPLAY_TYPE_DPI,
-
-	/* OMAP_DSS_CHANNEL_LCD2 */
-	OMAP_DISPLAY_TYPE_DPI | OMAP_DISPLAY_TYPE_DBI |
-	OMAP_DISPLAY_TYPE_DSI,
-};
-
-static const enum omap_dss_output_id omap2_dss_supported_outputs[] = {
-	/* OMAP_DSS_CHANNEL_LCD */
-	OMAP_DSS_OUTPUT_DPI | OMAP_DSS_OUTPUT_DBI,
-
-	/* OMAP_DSS_CHANNEL_DIGIT */
-	OMAP_DSS_OUTPUT_VENC,
-};
-
-static const enum omap_dss_output_id omap3430_dss_supported_outputs[] = {
-	/* OMAP_DSS_CHANNEL_LCD */
-	OMAP_DSS_OUTPUT_DPI | OMAP_DSS_OUTPUT_DBI |
-	OMAP_DSS_OUTPUT_SDI | OMAP_DSS_OUTPUT_DSI1,
-
-	/* OMAP_DSS_CHANNEL_DIGIT */
-	OMAP_DSS_OUTPUT_VENC,
-};
-
-static const enum omap_dss_output_id omap3630_dss_supported_outputs[] = {
-	/* OMAP_DSS_CHANNEL_LCD */
-	OMAP_DSS_OUTPUT_DPI | OMAP_DSS_OUTPUT_DBI |
-	OMAP_DSS_OUTPUT_DSI1,
-
-	/* OMAP_DSS_CHANNEL_DIGIT */
-	OMAP_DSS_OUTPUT_VENC,
-};
-
-static const enum omap_dss_output_id am43xx_dss_supported_outputs[] = {
-	/* OMAP_DSS_CHANNEL_LCD */
-	OMAP_DSS_OUTPUT_DPI | OMAP_DSS_OUTPUT_DBI,
-};
-
-static const enum omap_dss_output_id omap4_dss_supported_outputs[] = {
-	/* OMAP_DSS_CHANNEL_LCD */
-	OMAP_DSS_OUTPUT_DBI | OMAP_DSS_OUTPUT_DSI1,
-
-	/* OMAP_DSS_CHANNEL_DIGIT */
-	OMAP_DSS_OUTPUT_VENC | OMAP_DSS_OUTPUT_HDMI,
-
-	/* OMAP_DSS_CHANNEL_LCD2 */
-	OMAP_DSS_OUTPUT_DPI | OMAP_DSS_OUTPUT_DBI |
-	OMAP_DSS_OUTPUT_DSI2,
-};
-
-static const enum omap_dss_output_id omap5_dss_supported_outputs[] = {
-	/* OMAP_DSS_CHANNEL_LCD */
-	OMAP_DSS_OUTPUT_DPI | OMAP_DSS_OUTPUT_DBI |
-	OMAP_DSS_OUTPUT_DSI1 | OMAP_DSS_OUTPUT_DSI2,
-
-	/* OMAP_DSS_CHANNEL_DIGIT */
-	OMAP_DSS_OUTPUT_HDMI,
-
-	/* OMAP_DSS_CHANNEL_LCD2 */
-	OMAP_DSS_OUTPUT_DPI | OMAP_DSS_OUTPUT_DBI |
-	OMAP_DSS_OUTPUT_DSI1,
-
-	/* OMAP_DSS_CHANNEL_LCD3 */
-	OMAP_DSS_OUTPUT_DPI | OMAP_DSS_OUTPUT_DBI |
-	OMAP_DSS_OUTPUT_DSI2,
-};
-
-#define COLOR_ARRAY(arr...) (const u32[]) { arr, 0 }
-
-static const u32 *omap2_dss_supported_color_modes[] = {
-
-	/* OMAP_DSS_GFX */
-	COLOR_ARRAY(
-	DRM_FORMAT_RGBX4444, DRM_FORMAT_RGB565,
-	DRM_FORMAT_XRGB8888, DRM_FORMAT_RGB888),
-
-	/* OMAP_DSS_VIDEO1 */
-	COLOR_ARRAY(
-	DRM_FORMAT_RGB565, DRM_FORMAT_XRGB8888,
-	DRM_FORMAT_RGB888, DRM_FORMAT_YUYV,
-	DRM_FORMAT_UYVY),
-
-	/* OMAP_DSS_VIDEO2 */
-	COLOR_ARRAY(
-	DRM_FORMAT_RGB565, DRM_FORMAT_XRGB8888,
-	DRM_FORMAT_RGB888, DRM_FORMAT_YUYV,
-	DRM_FORMAT_UYVY),
-};
-
-static const u32 *omap3_dss_supported_color_modes[] = {
-	/* OMAP_DSS_GFX */
-	COLOR_ARRAY(
-	DRM_FORMAT_RGBX4444, DRM_FORMAT_ARGB4444,
-	DRM_FORMAT_RGB565, DRM_FORMAT_XRGB8888,
-	DRM_FORMAT_RGB888, DRM_FORMAT_ARGB8888,
-	DRM_FORMAT_RGBA8888, DRM_FORMAT_RGBX8888),
-
-	/* OMAP_DSS_VIDEO1 */
-	COLOR_ARRAY(
-	DRM_FORMAT_XRGB8888, DRM_FORMAT_RGB888,
-	DRM_FORMAT_RGBX4444, DRM_FORMAT_RGB565,
-	DRM_FORMAT_YUYV, DRM_FORMAT_UYVY),
-
-	/* OMAP_DSS_VIDEO2 */
-	COLOR_ARRAY(
-	DRM_FORMAT_RGBX4444, DRM_FORMAT_ARGB4444,
-	DRM_FORMAT_RGB565, DRM_FORMAT_XRGB8888,
-	DRM_FORMAT_RGB888, DRM_FORMAT_YUYV,
-	DRM_FORMAT_UYVY, DRM_FORMAT_ARGB8888,
-	DRM_FORMAT_RGBA8888, DRM_FORMAT_RGBX8888),
-};
-
-static const u32 *omap4_dss_supported_color_modes[] = {
-	/* OMAP_DSS_GFX */
-	COLOR_ARRAY(
-	DRM_FORMAT_RGBX4444, DRM_FORMAT_ARGB4444,
-	DRM_FORMAT_RGB565, DRM_FORMAT_XRGB8888,
-	DRM_FORMAT_RGB888, DRM_FORMAT_ARGB8888,
-	DRM_FORMAT_RGBA8888, DRM_FORMAT_RGBX8888,
-	DRM_FORMAT_ARGB1555, DRM_FORMAT_XRGB4444,
-	DRM_FORMAT_RGBA4444, DRM_FORMAT_XRGB1555),
-
-	/* OMAP_DSS_VIDEO1 */
-	COLOR_ARRAY(
-	DRM_FORMAT_RGB565, DRM_FORMAT_RGBX4444,
-	DRM_FORMAT_YUYV, DRM_FORMAT_ARGB1555,
-	DRM_FORMAT_RGBA8888, DRM_FORMAT_NV12,
-	DRM_FORMAT_RGBA4444, DRM_FORMAT_XRGB8888,
-	DRM_FORMAT_RGB888, DRM_FORMAT_UYVY,
-	DRM_FORMAT_ARGB4444, DRM_FORMAT_XRGB1555,
-	DRM_FORMAT_ARGB8888, DRM_FORMAT_XRGB4444,
-	DRM_FORMAT_RGBX8888),
-
-       /* OMAP_DSS_VIDEO2 */
-	COLOR_ARRAY(
-	DRM_FORMAT_RGB565, DRM_FORMAT_RGBX4444,
-	DRM_FORMAT_YUYV, DRM_FORMAT_ARGB1555,
-	DRM_FORMAT_RGBA8888, DRM_FORMAT_NV12,
-	DRM_FORMAT_RGBA4444, DRM_FORMAT_XRGB8888,
-	DRM_FORMAT_RGB888, DRM_FORMAT_UYVY,
-	DRM_FORMAT_ARGB4444, DRM_FORMAT_XRGB1555,
-	DRM_FORMAT_ARGB8888, DRM_FORMAT_XRGB4444,
-	DRM_FORMAT_RGBX8888),
-
-	/* OMAP_DSS_VIDEO3 */
-	COLOR_ARRAY(
-	DRM_FORMAT_RGB565, DRM_FORMAT_RGBX4444,
-	DRM_FORMAT_YUYV, DRM_FORMAT_ARGB1555,
-	DRM_FORMAT_RGBA8888, DRM_FORMAT_NV12,
-	DRM_FORMAT_RGBA4444, DRM_FORMAT_XRGB8888,
-	DRM_FORMAT_RGB888, DRM_FORMAT_UYVY,
-	DRM_FORMAT_ARGB4444, DRM_FORMAT_XRGB1555,
-	DRM_FORMAT_ARGB8888, DRM_FORMAT_XRGB4444,
-	DRM_FORMAT_RGBX8888),
-
-	/* OMAP_DSS_WB */
-	COLOR_ARRAY(
-	DRM_FORMAT_RGB565, DRM_FORMAT_RGBX4444,
-	DRM_FORMAT_YUYV, DRM_FORMAT_ARGB1555,
-	DRM_FORMAT_RGBA8888, DRM_FORMAT_NV12,
-	DRM_FORMAT_RGBA4444, DRM_FORMAT_XRGB8888,
-	DRM_FORMAT_RGB888, DRM_FORMAT_UYVY,
-	DRM_FORMAT_ARGB4444, DRM_FORMAT_XRGB1555,
-	DRM_FORMAT_ARGB8888, DRM_FORMAT_XRGB4444,
-	DRM_FORMAT_RGBX8888),
-};
-
-static const enum omap_overlay_caps omap2_dss_overlay_caps[] = {
-	/* OMAP_DSS_GFX */
-	OMAP_DSS_OVL_CAP_POS | OMAP_DSS_OVL_CAP_REPLICATION,
-
-	/* OMAP_DSS_VIDEO1 */
-	OMAP_DSS_OVL_CAP_SCALE | OMAP_DSS_OVL_CAP_POS |
-		OMAP_DSS_OVL_CAP_REPLICATION,
-
-	/* OMAP_DSS_VIDEO2 */
-	OMAP_DSS_OVL_CAP_SCALE | OMAP_DSS_OVL_CAP_POS |
-		OMAP_DSS_OVL_CAP_REPLICATION,
-};
-
-static const enum omap_overlay_caps omap3430_dss_overlay_caps[] = {
-	/* OMAP_DSS_GFX */
-	OMAP_DSS_OVL_CAP_GLOBAL_ALPHA | OMAP_DSS_OVL_CAP_POS |
-		OMAP_DSS_OVL_CAP_REPLICATION,
-
-	/* OMAP_DSS_VIDEO1 */
-	OMAP_DSS_OVL_CAP_SCALE | OMAP_DSS_OVL_CAP_POS |
-		OMAP_DSS_OVL_CAP_REPLICATION,
-
-	/* OMAP_DSS_VIDEO2 */
-	OMAP_DSS_OVL_CAP_SCALE | OMAP_DSS_OVL_CAP_GLOBAL_ALPHA |
-		OMAP_DSS_OVL_CAP_POS | OMAP_DSS_OVL_CAP_REPLICATION,
-};
-
-static const enum omap_overlay_caps omap3630_dss_overlay_caps[] = {
-	/* OMAP_DSS_GFX */
-	OMAP_DSS_OVL_CAP_GLOBAL_ALPHA | OMAP_DSS_OVL_CAP_PRE_MULT_ALPHA |
-		OMAP_DSS_OVL_CAP_POS | OMAP_DSS_OVL_CAP_REPLICATION,
-
-	/* OMAP_DSS_VIDEO1 */
-	OMAP_DSS_OVL_CAP_SCALE | OMAP_DSS_OVL_CAP_POS |
-		OMAP_DSS_OVL_CAP_REPLICATION,
-
-	/* OMAP_DSS_VIDEO2 */
-	OMAP_DSS_OVL_CAP_SCALE | OMAP_DSS_OVL_CAP_GLOBAL_ALPHA |
-		OMAP_DSS_OVL_CAP_PRE_MULT_ALPHA | OMAP_DSS_OVL_CAP_POS |
-		OMAP_DSS_OVL_CAP_REPLICATION,
-};
-
-static const enum omap_overlay_caps omap4_dss_overlay_caps[] = {
-	/* OMAP_DSS_GFX */
-	OMAP_DSS_OVL_CAP_GLOBAL_ALPHA | OMAP_DSS_OVL_CAP_PRE_MULT_ALPHA |
-		OMAP_DSS_OVL_CAP_ZORDER | OMAP_DSS_OVL_CAP_POS |
-		OMAP_DSS_OVL_CAP_REPLICATION,
-
-	/* OMAP_DSS_VIDEO1 */
-	OMAP_DSS_OVL_CAP_SCALE | OMAP_DSS_OVL_CAP_GLOBAL_ALPHA |
-		OMAP_DSS_OVL_CAP_PRE_MULT_ALPHA | OMAP_DSS_OVL_CAP_ZORDER |
-		OMAP_DSS_OVL_CAP_POS | OMAP_DSS_OVL_CAP_REPLICATION,
-
-	/* OMAP_DSS_VIDEO2 */
-	OMAP_DSS_OVL_CAP_SCALE | OMAP_DSS_OVL_CAP_GLOBAL_ALPHA |
-		OMAP_DSS_OVL_CAP_PRE_MULT_ALPHA | OMAP_DSS_OVL_CAP_ZORDER |
-		OMAP_DSS_OVL_CAP_POS | OMAP_DSS_OVL_CAP_REPLICATION,
-
-	/* OMAP_DSS_VIDEO3 */
-	OMAP_DSS_OVL_CAP_SCALE | OMAP_DSS_OVL_CAP_GLOBAL_ALPHA |
-		OMAP_DSS_OVL_CAP_PRE_MULT_ALPHA | OMAP_DSS_OVL_CAP_ZORDER |
-		OMAP_DSS_OVL_CAP_POS | OMAP_DSS_OVL_CAP_REPLICATION,
-};
-
-static const struct dss_param_range omap2_dss_param_range[] = {
-	[FEAT_PARAM_DSS_FCK]			= { 0, 133000000 },
-	[FEAT_PARAM_DSS_PCD]			= { 2, 255 },
-	[FEAT_PARAM_DOWNSCALE]			= { 1, 2 },
-	/*
-	 * Assuming the line width buffer to be 768 pixels as OMAP2 DISPC
-	 * scaler cannot scale a image with width more than 768.
-	 */
-	[FEAT_PARAM_LINEWIDTH]			= { 1, 768 },
-};
-
-static const struct dss_param_range omap3_dss_param_range[] = {
-	[FEAT_PARAM_DSS_FCK]			= { 0, 173000000 },
-	[FEAT_PARAM_DSS_PCD]			= { 1, 255 },
-	[FEAT_PARAM_DSIPLL_LPDIV]		= { 1, (1 << 13) - 1},
-	[FEAT_PARAM_DSI_FCK]			= { 0, 173000000 },
-	[FEAT_PARAM_DOWNSCALE]			= { 1, 4 },
-	[FEAT_PARAM_LINEWIDTH]			= { 1, 1024 },
-};
-
-static const struct dss_param_range am43xx_dss_param_range[] = {
-	[FEAT_PARAM_DSS_FCK]			= { 0, 200000000 },
-	[FEAT_PARAM_DSS_PCD]			= { 1, 255 },
-	[FEAT_PARAM_DOWNSCALE]			= { 1, 4 },
-	[FEAT_PARAM_LINEWIDTH]			= { 1, 1024 },
-};
-
-static const struct dss_param_range omap4_dss_param_range[] = {
-	[FEAT_PARAM_DSS_FCK]			= { 0, 186000000 },
-	[FEAT_PARAM_DSS_PCD]			= { 1, 255 },
-	[FEAT_PARAM_DSIPLL_LPDIV]		= { 0, (1 << 13) - 1 },
-	[FEAT_PARAM_DSI_FCK]			= { 0, 170000000 },
-	[FEAT_PARAM_DOWNSCALE]			= { 1, 4 },
-	[FEAT_PARAM_LINEWIDTH]			= { 1, 2048 },
-};
-
-static const struct dss_param_range omap5_dss_param_range[] = {
-	[FEAT_PARAM_DSS_FCK]			= { 0, 209250000 },
-	[FEAT_PARAM_DSS_PCD]			= { 1, 255 },
-	[FEAT_PARAM_DSIPLL_LPDIV]		= { 0, (1 << 13) - 1 },
-	[FEAT_PARAM_DSI_FCK]			= { 0, 209250000 },
-	[FEAT_PARAM_DOWNSCALE]			= { 1, 4 },
-	[FEAT_PARAM_LINEWIDTH]			= { 1, 2048 },
-};
-
-static const enum dss_feat_id omap2_dss_feat_list[] = {
-	FEAT_LCDENABLEPOL,
-	FEAT_LCDENABLESIGNAL,
-	FEAT_PCKFREEENABLE,
-	FEAT_FUNCGATED,
-	FEAT_ROWREPEATENABLE,
-	FEAT_RESIZECONF,
-};
-
-static const enum dss_feat_id omap3430_dss_feat_list[] = {
-	FEAT_LCDENABLEPOL,
-	FEAT_LCDENABLESIGNAL,
-	FEAT_PCKFREEENABLE,
-	FEAT_FUNCGATED,
-	FEAT_LINEBUFFERSPLIT,
-	FEAT_ROWREPEATENABLE,
-	FEAT_RESIZECONF,
-	FEAT_DSI_REVERSE_TXCLKESC,
-	FEAT_VENC_REQUIRES_TV_DAC_CLK,
-	FEAT_CPR,
-	FEAT_PRELOAD,
-	FEAT_FIR_COEF_V,
-	FEAT_ALPHA_FIXED_ZORDER,
-	FEAT_FIFO_MERGE,
-	FEAT_OMAP3_DSI_FIFO_BUG,
-	FEAT_DPI_USES_VDDS_DSI,
-};
-
-static const enum dss_feat_id am35xx_dss_feat_list[] = {
-	FEAT_LCDENABLEPOL,
-	FEAT_LCDENABLESIGNAL,
-	FEAT_PCKFREEENABLE,
-	FEAT_FUNCGATED,
-	FEAT_LINEBUFFERSPLIT,
-	FEAT_ROWREPEATENABLE,
-	FEAT_RESIZECONF,
-	FEAT_DSI_REVERSE_TXCLKESC,
-	FEAT_VENC_REQUIRES_TV_DAC_CLK,
-	FEAT_CPR,
-	FEAT_PRELOAD,
-	FEAT_FIR_COEF_V,
-	FEAT_ALPHA_FIXED_ZORDER,
-	FEAT_FIFO_MERGE,
-	FEAT_OMAP3_DSI_FIFO_BUG,
-};
-
-static const enum dss_feat_id am43xx_dss_feat_list[] = {
-	FEAT_LCDENABLEPOL,
-	FEAT_LCDENABLESIGNAL,
-	FEAT_PCKFREEENABLE,
-	FEAT_FUNCGATED,
-	FEAT_LINEBUFFERSPLIT,
-	FEAT_ROWREPEATENABLE,
-	FEAT_RESIZECONF,
-	FEAT_CPR,
-	FEAT_PRELOAD,
-	FEAT_FIR_COEF_V,
-	FEAT_ALPHA_FIXED_ZORDER,
-	FEAT_FIFO_MERGE,
-};
-
-static const enum dss_feat_id omap3630_dss_feat_list[] = {
-	FEAT_LCDENABLEPOL,
-	FEAT_LCDENABLESIGNAL,
-	FEAT_PCKFREEENABLE,
-	FEAT_FUNCGATED,
-	FEAT_LINEBUFFERSPLIT,
-	FEAT_ROWREPEATENABLE,
-	FEAT_RESIZECONF,
-	FEAT_DSI_PLL_PWR_BUG,
-	FEAT_CPR,
-	FEAT_PRELOAD,
-	FEAT_FIR_COEF_V,
-	FEAT_ALPHA_FIXED_ZORDER,
-	FEAT_FIFO_MERGE,
-	FEAT_OMAP3_DSI_FIFO_BUG,
-	FEAT_DPI_USES_VDDS_DSI,
-};
-
-static const enum dss_feat_id omap4430_es1_0_dss_feat_list[] = {
-	FEAT_MGR_LCD2,
-	FEAT_CORE_CLK_DIV,
-	FEAT_LCD_CLK_SRC,
-	FEAT_DSI_DCS_CMD_CONFIG_VC,
-	FEAT_DSI_VC_OCP_WIDTH,
-	FEAT_DSI_GNQ,
-	FEAT_HANDLE_UV_SEPARATE,
-	FEAT_ATTR2,
-	FEAT_CPR,
-	FEAT_PRELOAD,
-	FEAT_FIR_COEF_V,
-	FEAT_ALPHA_FREE_ZORDER,
-	FEAT_FIFO_MERGE,
-	FEAT_BURST_2D,
-};
-
-static const enum dss_feat_id omap4430_es2_0_1_2_dss_feat_list[] = {
-	FEAT_MGR_LCD2,
-	FEAT_CORE_CLK_DIV,
-	FEAT_LCD_CLK_SRC,
-	FEAT_DSI_DCS_CMD_CONFIG_VC,
-	FEAT_DSI_VC_OCP_WIDTH,
-	FEAT_DSI_GNQ,
-	FEAT_HDMI_CTS_SWMODE,
-	FEAT_HANDLE_UV_SEPARATE,
-	FEAT_ATTR2,
-	FEAT_CPR,
-	FEAT_PRELOAD,
-	FEAT_FIR_COEF_V,
-	FEAT_ALPHA_FREE_ZORDER,
-	FEAT_FIFO_MERGE,
-	FEAT_BURST_2D,
-};
-
-static const enum dss_feat_id omap4_dss_feat_list[] = {
-	FEAT_MGR_LCD2,
-	FEAT_CORE_CLK_DIV,
-	FEAT_LCD_CLK_SRC,
-	FEAT_DSI_DCS_CMD_CONFIG_VC,
-	FEAT_DSI_VC_OCP_WIDTH,
-	FEAT_DSI_GNQ,
-	FEAT_HDMI_CTS_SWMODE,
-	FEAT_HDMI_AUDIO_USE_MCLK,
-	FEAT_HANDLE_UV_SEPARATE,
-	FEAT_ATTR2,
-	FEAT_CPR,
-	FEAT_PRELOAD,
-	FEAT_FIR_COEF_V,
-	FEAT_ALPHA_FREE_ZORDER,
-	FEAT_FIFO_MERGE,
-	FEAT_BURST_2D,
-};
-
-static const enum dss_feat_id omap5_dss_feat_list[] = {
-	FEAT_MGR_LCD2,
-	FEAT_MGR_LCD3,
-	FEAT_CORE_CLK_DIV,
-	FEAT_LCD_CLK_SRC,
-	FEAT_DSI_DCS_CMD_CONFIG_VC,
-	FEAT_DSI_VC_OCP_WIDTH,
-	FEAT_DSI_GNQ,
-	FEAT_HDMI_CTS_SWMODE,
-	FEAT_HDMI_AUDIO_USE_MCLK,
-	FEAT_HANDLE_UV_SEPARATE,
-	FEAT_ATTR2,
-	FEAT_CPR,
-	FEAT_PRELOAD,
-	FEAT_FIR_COEF_V,
-	FEAT_ALPHA_FREE_ZORDER,
-	FEAT_FIFO_MERGE,
-	FEAT_BURST_2D,
-	FEAT_DSI_PHY_DCC,
-	FEAT_MFLAG,
-};
-
-/* OMAP2 DSS Features */
-static const struct omap_dss_features omap2_dss_features = {
-	.reg_fields = omap2_dss_reg_fields,
-	.num_reg_fields = ARRAY_SIZE(omap2_dss_reg_fields),
-
-	.features = omap2_dss_feat_list,
-	.num_features = ARRAY_SIZE(omap2_dss_feat_list),
-
-	.num_mgrs = 2,
-	.num_ovls = 3,
-	.supported_displays = omap2_dss_supported_displays,
-	.supported_outputs = omap2_dss_supported_outputs,
-	.supported_color_modes = omap2_dss_supported_color_modes,
-	.overlay_caps = omap2_dss_overlay_caps,
-	.dss_params = omap2_dss_param_range,
-	.buffer_size_unit = 1,
-	.burst_size_unit = 8,
-};
-
-/* OMAP3 DSS Features */
-static const struct omap_dss_features omap3430_dss_features = {
-	.reg_fields = omap3_dss_reg_fields,
-	.num_reg_fields = ARRAY_SIZE(omap3_dss_reg_fields),
-
-	.features = omap3430_dss_feat_list,
-	.num_features = ARRAY_SIZE(omap3430_dss_feat_list),
-
-	.num_mgrs = 2,
-	.num_ovls = 3,
-	.supported_displays = omap3430_dss_supported_displays,
-	.supported_outputs = omap3430_dss_supported_outputs,
-	.supported_color_modes = omap3_dss_supported_color_modes,
-	.overlay_caps = omap3430_dss_overlay_caps,
-	.dss_params = omap3_dss_param_range,
-	.buffer_size_unit = 1,
-	.burst_size_unit = 8,
-};
-
-/*
- * AM35xx DSS Features. This is basically OMAP3 DSS Features without the
- * vdds_dsi regulator.
- */
-static const struct omap_dss_features am35xx_dss_features = {
-	.reg_fields = omap3_dss_reg_fields,
-	.num_reg_fields = ARRAY_SIZE(omap3_dss_reg_fields),
-
-	.features = am35xx_dss_feat_list,
-	.num_features = ARRAY_SIZE(am35xx_dss_feat_list),
-
-	.num_mgrs = 2,
-	.num_ovls = 3,
-	.supported_displays = omap3430_dss_supported_displays,
-	.supported_outputs = omap3430_dss_supported_outputs,
-	.supported_color_modes = omap3_dss_supported_color_modes,
-	.overlay_caps = omap3430_dss_overlay_caps,
-	.dss_params = omap3_dss_param_range,
-	.buffer_size_unit = 1,
-	.burst_size_unit = 8,
-};
-
-static const struct omap_dss_features am43xx_dss_features = {
-	.reg_fields = am43xx_dss_reg_fields,
-	.num_reg_fields = ARRAY_SIZE(am43xx_dss_reg_fields),
-
-	.features = am43xx_dss_feat_list,
-	.num_features = ARRAY_SIZE(am43xx_dss_feat_list),
-
-	.num_mgrs = 1,
-	.num_ovls = 3,
-	.supported_displays = am43xx_dss_supported_displays,
-	.supported_outputs = am43xx_dss_supported_outputs,
-	.supported_color_modes = omap3_dss_supported_color_modes,
-	.overlay_caps = omap3430_dss_overlay_caps,
-	.dss_params = am43xx_dss_param_range,
-	.buffer_size_unit = 1,
-	.burst_size_unit = 8,
-};
-
-static const struct omap_dss_features omap3630_dss_features = {
-	.reg_fields = omap3_dss_reg_fields,
-	.num_reg_fields = ARRAY_SIZE(omap3_dss_reg_fields),
-
-	.features = omap3630_dss_feat_list,
-	.num_features = ARRAY_SIZE(omap3630_dss_feat_list),
-
-	.num_mgrs = 2,
-	.num_ovls = 3,
-	.supported_displays = omap3630_dss_supported_displays,
-	.supported_outputs = omap3630_dss_supported_outputs,
-	.supported_color_modes = omap3_dss_supported_color_modes,
-	.overlay_caps = omap3630_dss_overlay_caps,
-	.dss_params = omap3_dss_param_range,
-	.buffer_size_unit = 1,
-	.burst_size_unit = 8,
-};
-
-/* OMAP4 DSS Features */
-/* For OMAP4430 ES 1.0 revision */
-static const struct omap_dss_features omap4430_es1_0_dss_features  = {
-	.reg_fields = omap4_dss_reg_fields,
-	.num_reg_fields = ARRAY_SIZE(omap4_dss_reg_fields),
-
-	.features = omap4430_es1_0_dss_feat_list,
-	.num_features = ARRAY_SIZE(omap4430_es1_0_dss_feat_list),
-
-	.num_mgrs = 3,
-	.num_ovls = 4,
-	.supported_displays = omap4_dss_supported_displays,
-	.supported_outputs = omap4_dss_supported_outputs,
-	.supported_color_modes = omap4_dss_supported_color_modes,
-	.overlay_caps = omap4_dss_overlay_caps,
-	.dss_params = omap4_dss_param_range,
-	.buffer_size_unit = 16,
-	.burst_size_unit = 16,
-};
-
-/* For OMAP4430 ES 2.0, 2.1 and 2.2 revisions */
-static const struct omap_dss_features omap4430_es2_0_1_2_dss_features = {
-	.reg_fields = omap4_dss_reg_fields,
-	.num_reg_fields = ARRAY_SIZE(omap4_dss_reg_fields),
-
-	.features = omap4430_es2_0_1_2_dss_feat_list,
-	.num_features = ARRAY_SIZE(omap4430_es2_0_1_2_dss_feat_list),
-
-	.num_mgrs = 3,
-	.num_ovls = 4,
-	.supported_displays = omap4_dss_supported_displays,
-	.supported_outputs = omap4_dss_supported_outputs,
-	.supported_color_modes = omap4_dss_supported_color_modes,
-	.overlay_caps = omap4_dss_overlay_caps,
-	.dss_params = omap4_dss_param_range,
-	.buffer_size_unit = 16,
-	.burst_size_unit = 16,
-};
-
-/* For all the other OMAP4 versions */
-static const struct omap_dss_features omap4_dss_features = {
-	.reg_fields = omap4_dss_reg_fields,
-	.num_reg_fields = ARRAY_SIZE(omap4_dss_reg_fields),
-
-	.features = omap4_dss_feat_list,
-	.num_features = ARRAY_SIZE(omap4_dss_feat_list),
-
-	.num_mgrs = 3,
-	.num_ovls = 4,
-	.supported_displays = omap4_dss_supported_displays,
-	.supported_outputs = omap4_dss_supported_outputs,
-	.supported_color_modes = omap4_dss_supported_color_modes,
-	.overlay_caps = omap4_dss_overlay_caps,
-	.dss_params = omap4_dss_param_range,
-	.buffer_size_unit = 16,
-	.burst_size_unit = 16,
-};
-
-/* OMAP5 DSS Features */
-static const struct omap_dss_features omap5_dss_features = {
-	.reg_fields = omap5_dss_reg_fields,
-	.num_reg_fields = ARRAY_SIZE(omap5_dss_reg_fields),
-
-	.features = omap5_dss_feat_list,
-	.num_features = ARRAY_SIZE(omap5_dss_feat_list),
-
-	.num_mgrs = 4,
-	.num_ovls = 4,
-	.supported_displays = omap5_dss_supported_displays,
-	.supported_outputs = omap5_dss_supported_outputs,
-	.supported_color_modes = omap4_dss_supported_color_modes,
-	.overlay_caps = omap4_dss_overlay_caps,
-	.dss_params = omap5_dss_param_range,
-	.buffer_size_unit = 16,
-	.burst_size_unit = 16,
-};
-
-/* Functions returning values related to a DSS feature */
-int dss_feat_get_num_mgrs(void)
-{
-	return omap_current_dss_features->num_mgrs;
-}
-
-int dss_feat_get_num_ovls(void)
-{
-	return omap_current_dss_features->num_ovls;
-}
-
-unsigned long dss_feat_get_param_min(enum dss_range_param param)
-{
-	return omap_current_dss_features->dss_params[param].min;
-}
-
-unsigned long dss_feat_get_param_max(enum dss_range_param param)
-{
-	return omap_current_dss_features->dss_params[param].max;
-}
-
-enum omap_display_type dss_feat_get_supported_displays(enum omap_channel channel)
-{
-	return omap_current_dss_features->supported_displays[channel];
-}
-
-enum omap_dss_output_id dss_feat_get_supported_outputs(enum omap_channel channel)
-{
-	return omap_current_dss_features->supported_outputs[channel];
-}
-
-const u32 *dss_feat_get_supported_color_modes(enum omap_plane_id plane)
-{
-	return omap_current_dss_features->supported_color_modes[plane];
-}
-
-enum omap_overlay_caps dss_feat_get_overlay_caps(enum omap_plane_id plane)
-{
-	return omap_current_dss_features->overlay_caps[plane];
-}
-
-bool dss_feat_color_mode_supported(enum omap_plane_id plane, u32 fourcc)
-{
-	const u32 *modes;
-	unsigned int i;
-
-	modes = omap_current_dss_features->supported_color_modes[plane];
-
-	for (i = 0; modes[i]; ++i) {
-		if (modes[i] == fourcc)
-			return true;
-	}
-
-	return false;
-}
-
-u32 dss_feat_get_buffer_size_unit(void)
-{
-	return omap_current_dss_features->buffer_size_unit;
-}
-
-u32 dss_feat_get_burst_size_unit(void)
-{
-	return omap_current_dss_features->burst_size_unit;
-}
-
-/* DSS has_feature check */
-bool dss_has_feature(enum dss_feat_id id)
-{
-	int i;
-	const enum dss_feat_id *features = omap_current_dss_features->features;
-	const int num_features = omap_current_dss_features->num_features;
-
-	for (i = 0; i < num_features; i++) {
-		if (features[i] == id)
-			return true;
-	}
-
-	return false;
-}
-
-void dss_feat_get_reg_field(enum dss_feat_reg_field id, u8 *start, u8 *end)
-{
-	if (id >= omap_current_dss_features->num_reg_fields)
-		BUG();
-
-	*start = omap_current_dss_features->reg_fields[id].start;
-	*end = omap_current_dss_features->reg_fields[id].end;
-}
-
-void dss_features_init(enum omapdss_version version)
-{
-	switch (version) {
-	case OMAPDSS_VER_OMAP24xx:
-		omap_current_dss_features = &omap2_dss_features;
-		break;
-
-	case OMAPDSS_VER_OMAP34xx_ES1:
-	case OMAPDSS_VER_OMAP34xx_ES3:
-		omap_current_dss_features = &omap3430_dss_features;
-		break;
-
-	case OMAPDSS_VER_OMAP3630:
-		omap_current_dss_features = &omap3630_dss_features;
-		break;
-
-	case OMAPDSS_VER_OMAP4430_ES1:
-		omap_current_dss_features = &omap4430_es1_0_dss_features;
-		break;
-
-	case OMAPDSS_VER_OMAP4430_ES2:
-		omap_current_dss_features = &omap4430_es2_0_1_2_dss_features;
-		break;
-
-	case OMAPDSS_VER_OMAP4:
-		omap_current_dss_features = &omap4_dss_features;
-		break;
-
-	case OMAPDSS_VER_OMAP5:
-	case OMAPDSS_VER_DRA7xx:
-		omap_current_dss_features = &omap5_dss_features;
-		break;
-
-	case OMAPDSS_VER_AM35xx:
-		omap_current_dss_features = &am35xx_dss_features;
-		break;
-
-	case OMAPDSS_VER_AM43xx:
-		omap_current_dss_features = &am43xx_dss_features;
-		break;
-
-	default:
-		DSSWARN("Unsupported OMAP version");
-		break;
-	}
-}
diff --git a/drivers/gpu/drm/omapdrm/dss/dss_features.h b/drivers/gpu/drm/omapdrm/dss/dss_features.h
deleted file mode 100644
index c36436d27ff5..000000000000
--- a/drivers/gpu/drm/omapdrm/dss/dss_features.h
+++ /dev/null
@@ -1,109 +0,0 @@
-/*
- * linux/drivers/video/omap2/dss/dss_features.h
- *
- * Copyright (C) 2010 Texas Instruments
- * Author: Archit Taneja <archit@ti.com>
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License version 2 as published by
- * the Free Software Foundation.
- *
- * This program is distributed in the hope that it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along with
- * this program.  If not, see <http://www.gnu.org/licenses/>.
- */
-
-#ifndef __OMAP2_DSS_FEATURES_H
-#define __OMAP2_DSS_FEATURES_H
-
-#define MAX_DSS_MANAGERS	4
-#define MAX_DSS_OVERLAYS	4
-#define MAX_DSS_LCD_MANAGERS	3
-#define MAX_NUM_DSI		2
-
-/* DSS has feature id */
-enum dss_feat_id {
-	FEAT_LCDENABLEPOL,
-	FEAT_LCDENABLESIGNAL,
-	FEAT_PCKFREEENABLE,
-	FEAT_FUNCGATED,
-	FEAT_MGR_LCD2,
-	FEAT_MGR_LCD3,
-	FEAT_LINEBUFFERSPLIT,
-	FEAT_ROWREPEATENABLE,
-	FEAT_RESIZECONF,
-	/* Independent core clk divider */
-	FEAT_CORE_CLK_DIV,
-	FEAT_LCD_CLK_SRC,
-	/* DSI-PLL power command 0x3 is not working */
-	FEAT_DSI_PLL_PWR_BUG,
-	FEAT_DSI_DCS_CMD_CONFIG_VC,
-	FEAT_DSI_VC_OCP_WIDTH,
-	FEAT_DSI_REVERSE_TXCLKESC,
-	FEAT_DSI_GNQ,
-	FEAT_DPI_USES_VDDS_DSI,
-	FEAT_HDMI_CTS_SWMODE,
-	FEAT_HDMI_AUDIO_USE_MCLK,
-	FEAT_HANDLE_UV_SEPARATE,
-	FEAT_ATTR2,
-	FEAT_VENC_REQUIRES_TV_DAC_CLK,
-	FEAT_CPR,
-	FEAT_PRELOAD,
-	FEAT_FIR_COEF_V,
-	FEAT_ALPHA_FIXED_ZORDER,
-	FEAT_ALPHA_FREE_ZORDER,
-	FEAT_FIFO_MERGE,
-	/* An unknown HW bug causing the normal FIFO thresholds not to work */
-	FEAT_OMAP3_DSI_FIFO_BUG,
-	FEAT_BURST_2D,
-	FEAT_DSI_PHY_DCC,
-	FEAT_MFLAG,
-};
-
-/* DSS register field id */
-enum dss_feat_reg_field {
-	FEAT_REG_FIRHINC,
-	FEAT_REG_FIRVINC,
-	FEAT_REG_FIFOHIGHTHRESHOLD,
-	FEAT_REG_FIFOLOWTHRESHOLD,
-	FEAT_REG_FIFOSIZE,
-	FEAT_REG_HORIZONTALACCU,
-	FEAT_REG_VERTICALACCU,
-	FEAT_REG_DISPC_CLK_SWITCH,
-};
-
-enum dss_range_param {
-	FEAT_PARAM_DSS_FCK,
-	FEAT_PARAM_DSS_PCD,
-	FEAT_PARAM_DSIPLL_LPDIV,
-	FEAT_PARAM_DSI_FCK,
-	FEAT_PARAM_DOWNSCALE,
-	FEAT_PARAM_LINEWIDTH,
-};
-
-/* DSS Feature Functions */
-unsigned long dss_feat_get_param_min(enum dss_range_param param);
-unsigned long dss_feat_get_param_max(enum dss_range_param param);
-enum omap_overlay_caps dss_feat_get_overlay_caps(enum omap_plane_id plane);
-bool dss_feat_color_mode_supported(enum omap_plane_id plane,
-		u32 fourcc);
-
-u32 dss_feat_get_buffer_size_unit(void);	/* in bytes */
-u32 dss_feat_get_burst_size_unit(void);		/* in bytes */
-
-bool dss_has_feature(enum dss_feat_id id);
-void dss_feat_get_reg_field(enum dss_feat_reg_field id, u8 *start, u8 *end);
-void dss_features_init(enum omapdss_version version);
-
-enum omap_display_type dss_feat_get_supported_displays(enum omap_channel channel);
-enum omap_dss_output_id dss_feat_get_supported_outputs(enum omap_channel channel);
-
-int dss_feat_get_num_mgrs(void);
-int dss_feat_get_num_ovls(void);
-const u32 *dss_feat_get_supported_color_modes(enum omap_plane_id plane);
-
-#endif
diff --git a/drivers/gpu/drm/omapdrm/dss/hdmi.h b/drivers/gpu/drm/omapdrm/dss/hdmi.h
index fb6cccd02374..a820b394af09 100644
--- a/drivers/gpu/drm/omapdrm/dss/hdmi.h
+++ b/drivers/gpu/drm/omapdrm/dss/hdmi.h
@@ -234,6 +234,7 @@ struct hdmi_core_audio_config {
 struct hdmi_wp_data {
 	void __iomem *base;
 	phys_addr_t phys_base;
+	unsigned int version;
 };
 
 struct hdmi_pll_data {
@@ -245,15 +246,24 @@ struct hdmi_pll_data {
 	struct hdmi_wp_data *wp;
 };
 
+struct hdmi_phy_features {
+	bool bist_ctrl;
+	bool ldo_voltage;
+	unsigned long max_phy;
+};
+
 struct hdmi_phy_data {
 	void __iomem *base;
 
+	const struct hdmi_phy_features *features;
 	u8 lane_function[4];
 	u8 lane_polarity[4];
 };
 
 struct hdmi_core_data {
 	void __iomem *base;
+	bool cts_swmode;
+	bool audio_use_mclk;
 };
 
 static inline void hdmi_write_reg(void __iomem *base_addr, const u32 idx,
@@ -303,7 +313,8 @@ void hdmi_wp_video_config_timing(struct hdmi_wp_data *wp,
 		struct videomode *vm);
 void hdmi_wp_init_vid_fmt_timings(struct hdmi_video_format *video_fmt,
 		struct videomode *vm, struct hdmi_config *param);
-int hdmi_wp_init(struct platform_device *pdev, struct hdmi_wp_data *wp);
+int hdmi_wp_init(struct platform_device *pdev, struct hdmi_wp_data *wp,
+		 unsigned int version);
 phys_addr_t hdmi_wp_get_audio_dma_addr(struct hdmi_wp_data *wp);
 
 /* HDMI PLL funcs */
@@ -316,7 +327,8 @@ void hdmi_pll_uninit(struct hdmi_pll_data *hpll);
 int hdmi_phy_configure(struct hdmi_phy_data *phy, unsigned long hfbitclk,
 	unsigned long lfbitclk);
 void hdmi_phy_dump(struct hdmi_phy_data *phy, struct seq_file *s);
-int hdmi_phy_init(struct platform_device *pdev, struct hdmi_phy_data *phy);
+int hdmi_phy_init(struct platform_device *pdev, struct hdmi_phy_data *phy,
+		  unsigned int version);
 int hdmi_phy_parse_lanes(struct hdmi_phy_data *phy, const u32 *lanes);
 
 /* HDMI common funcs */
diff --git a/drivers/gpu/drm/omapdrm/dss/hdmi4.c b/drivers/gpu/drm/omapdrm/dss/hdmi4.c
index 284b4942b9ac..f169348da377 100644
--- a/drivers/gpu/drm/omapdrm/dss/hdmi4.c
+++ b/drivers/gpu/drm/omapdrm/dss/hdmi4.c
@@ -40,7 +40,6 @@
 #include "omapdss.h"
 #include "hdmi4_core.h"
 #include "dss.h"
-#include "dss_features.h"
 #include "hdmi.h"
 
 static struct omap_hdmi hdmi;
@@ -668,7 +667,7 @@ static int hdmi_audio_register(struct device *dev)
 {
 	struct omap_hdmi_audio_pdata pdata = {
 		.dev = dev,
-		.dss_version = omapdss_get_version(),
+		.version = 4,
 		.audio_dma_addr = hdmi_wp_get_audio_dma_addr(&hdmi.wp),
 		.ops = &hdmi_audio_ops,
 	};
@@ -700,7 +699,7 @@ static int hdmi4_bind(struct device *dev, struct device *master, void *data)
 	if (r)
 		return r;
 
-	r = hdmi_wp_init(pdev, &hdmi.wp);
+	r = hdmi_wp_init(pdev, &hdmi.wp, 4);
 	if (r)
 		return r;
 
@@ -708,7 +707,7 @@ static int hdmi4_bind(struct device *dev, struct device *master, void *data)
 	if (r)
 		return r;
 
-	r = hdmi_phy_init(pdev, &hdmi.phy);
+	r = hdmi_phy_init(pdev, &hdmi.phy, 4);
 	if (r)
 		goto err;
 
diff --git a/drivers/gpu/drm/omapdrm/dss/hdmi4_core.c b/drivers/gpu/drm/omapdrm/dss/hdmi4_core.c
index ed6001613405..365cf07daa01 100644
--- a/drivers/gpu/drm/omapdrm/dss/hdmi4_core.c
+++ b/drivers/gpu/drm/omapdrm/dss/hdmi4_core.c
@@ -31,11 +31,11 @@
 #include <linux/platform_device.h>
 #include <linux/string.h>
 #include <linux/seq_file.h>
+#include <linux/sys_soc.h>
 #include <sound/asound.h>
 #include <sound/asoundef.h>
 
 #include "hdmi4_core.h"
-#include "dss_features.h"
 
 #define HDMI_CORE_AV		0x500
 
@@ -757,10 +757,10 @@ int hdmi4_audio_config(struct hdmi_core_data *core, struct hdmi_wp_data *wp,
 	/* Audio clock regeneration settings */
 	acore.n = n;
 	acore.cts = cts;
-	if (dss_has_feature(FEAT_HDMI_CTS_SWMODE)) {
+	if (core->cts_swmode) {
 		acore.aud_par_busclk = 0;
 		acore.cts_mode = HDMI_AUDIO_CTS_MODE_SW;
-		acore.use_mclk = dss_has_feature(FEAT_HDMI_AUDIO_USE_MCLK);
+		acore.use_mclk = core->audio_use_mclk;
 	} else {
 		acore.aud_par_busclk = (((128 * 31) - 1) << 8);
 		acore.cts_mode = HDMI_AUDIO_CTS_MODE_HW;
@@ -884,10 +884,42 @@ void hdmi4_audio_stop(struct hdmi_core_data *core, struct hdmi_wp_data *wp)
 	hdmi_wp_audio_core_req_enable(wp, false);
 }
 
+struct hdmi4_features {
+	bool cts_swmode;
+	bool audio_use_mclk;
+};
+
+static const struct hdmi4_features hdmi4_es1_features = {
+	.cts_swmode = false,
+	.audio_use_mclk = false,
+};
+
+static const struct hdmi4_features hdmi4_es2_features = {
+	.cts_swmode = true,
+	.audio_use_mclk = false,
+};
+
+static const struct hdmi4_features hdmi4_es3_features = {
+	.cts_swmode = true,
+	.audio_use_mclk = true,
+};
+
+static const struct soc_device_attribute hdmi4_soc_devices[] = {
+	{ .family = "OMAP4", .revision = "ES1.?", .data = &hdmi4_es1_features },
+	{ .family = "OMAP4", .revision = "ES2.?", .data = &hdmi4_es2_features },
+	{ .family = "OMAP4",			  .data = &hdmi4_es3_features },
+	{ /* sentinel */ }
+};
+
 int hdmi4_core_init(struct platform_device *pdev, struct hdmi_core_data *core)
 {
+	const struct hdmi4_features *features;
 	struct resource *res;
 
+	features = soc_device_match(hdmi4_soc_devices)->data;
+	core->cts_swmode = features->cts_swmode;
+	core->audio_use_mclk = features->audio_use_mclk;
+
 	res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "core");
 	core->base = devm_ioremap_resource(&pdev->dev, res);
 	if (IS_ERR(core->base))
diff --git a/drivers/gpu/drm/omapdrm/dss/hdmi5.c b/drivers/gpu/drm/omapdrm/dss/hdmi5.c
index 441e1999d86a..b3221ca5bcd8 100644
--- a/drivers/gpu/drm/omapdrm/dss/hdmi5.c
+++ b/drivers/gpu/drm/omapdrm/dss/hdmi5.c
@@ -45,7 +45,6 @@
 #include "omapdss.h"
 #include "hdmi5_core.h"
 #include "dss.h"
-#include "dss_features.h"
 
 static struct omap_hdmi hdmi;
 
@@ -695,7 +694,7 @@ static int hdmi_audio_register(struct device *dev)
 {
 	struct omap_hdmi_audio_pdata pdata = {
 		.dev = dev,
-		.dss_version = omapdss_get_version(),
+		.version = 5,
 		.audio_dma_addr = hdmi_wp_get_audio_dma_addr(&hdmi.wp),
 		.ops = &hdmi_audio_ops,
 	};
@@ -732,7 +731,7 @@ static int hdmi5_bind(struct device *dev, struct device *master, void *data)
 	if (r)
 		return r;
 
-	r = hdmi_wp_init(pdev, &hdmi.wp);
+	r = hdmi_wp_init(pdev, &hdmi.wp, 5);
 	if (r)
 		return r;
 
@@ -740,7 +739,7 @@ static int hdmi5_bind(struct device *dev, struct device *master, void *data)
 	if (r)
 		return r;
 
-	r = hdmi_phy_init(pdev, &hdmi.phy);
+	r = hdmi_phy_init(pdev, &hdmi.phy, 5);
 	if (r)
 		goto err;
 
diff --git a/drivers/gpu/drm/omapdrm/dss/hdmi_phy.c b/drivers/gpu/drm/omapdrm/dss/hdmi_phy.c
index fb5e4c724b4b..a156292b1820 100644
--- a/drivers/gpu/drm/omapdrm/dss/hdmi_phy.c
+++ b/drivers/gpu/drm/omapdrm/dss/hdmi_phy.c
@@ -19,14 +19,6 @@
 #include "dss.h"
 #include "hdmi.h"
 
-struct hdmi_phy_features {
-	bool bist_ctrl;
-	bool ldo_voltage;
-	unsigned long max_phy;
-};
-
-static const struct hdmi_phy_features *phy_feat;
-
 void hdmi_phy_dump(struct hdmi_phy_data *phy, struct seq_file *s)
 {
 #define DUMPPHY(r) seq_printf(s, "%-35s %08x\n", #r,\
@@ -36,7 +28,7 @@ void hdmi_phy_dump(struct hdmi_phy_data *phy, struct seq_file *s)
 	DUMPPHY(HDMI_TXPHY_DIGITAL_CTRL);
 	DUMPPHY(HDMI_TXPHY_POWER_CTRL);
 	DUMPPHY(HDMI_TXPHY_PAD_CFG_CTRL);
-	if (phy_feat->bist_ctrl)
+	if (phy->features->bist_ctrl)
 		DUMPPHY(HDMI_TXPHY_BIST_CONTROL);
 }
 
@@ -146,7 +138,7 @@ int hdmi_phy_configure(struct hdmi_phy_data *phy, unsigned long hfbitclk,
 	 * In OMAP5+, the HFBITCLK must be divided by 2 before issuing the
 	 * HDMI_PHYPWRCMD_LDOON command.
 	*/
-	if (phy_feat->bist_ctrl)
+	if (phy->features->bist_ctrl)
 		REG_FLD_MOD(phy->base, HDMI_TXPHY_BIST_CONTROL, 1, 11, 11);
 
 	/*
@@ -155,7 +147,7 @@ int hdmi_phy_configure(struct hdmi_phy_data *phy, unsigned long hfbitclk,
 	 */
 	if (hfbitclk != lfbitclk)
 		freqout = 0;
-	else if (hfbitclk / 10 < phy_feat->max_phy)
+	else if (hfbitclk / 10 < phy->features->max_phy)
 		freqout = 1;
 	else
 		freqout = 2;
@@ -170,7 +162,7 @@ int hdmi_phy_configure(struct hdmi_phy_data *phy, unsigned long hfbitclk,
 	hdmi_write_reg(phy->base, HDMI_TXPHY_DIGITAL_CTRL, 0xF0000000);
 
 	/* Setup max LDO voltage */
-	if (phy_feat->ldo_voltage)
+	if (phy->features->ldo_voltage)
 		REG_FLD_MOD(phy->base, HDMI_TXPHY_POWER_CTRL, 0xB, 3, 0);
 
 	hdmi_phy_configure_lanes(phy);
@@ -190,47 +182,15 @@ static const struct hdmi_phy_features omap54xx_phy_feats = {
 	.max_phy	=	186000000,
 };
 
-static int hdmi_phy_init_features(struct platform_device *pdev)
-{
-	struct hdmi_phy_features *dst;
-	const struct hdmi_phy_features *src;
-
-	dst = devm_kzalloc(&pdev->dev, sizeof(*dst), GFP_KERNEL);
-	if (!dst) {
-		dev_err(&pdev->dev, "Failed to allocate HDMI PHY Features\n");
-		return -ENOMEM;
-	}
-
-	switch (omapdss_get_version()) {
-	case OMAPDSS_VER_OMAP4430_ES1:
-	case OMAPDSS_VER_OMAP4430_ES2:
-	case OMAPDSS_VER_OMAP4:
-		src = &omap44xx_phy_feats;
-		break;
-
-	case OMAPDSS_VER_OMAP5:
-	case OMAPDSS_VER_DRA7xx:
-		src = &omap54xx_phy_feats;
-		break;
-
-	default:
-		return -ENODEV;
-	}
-
-	memcpy(dst, src, sizeof(*dst));
-	phy_feat = dst;
-
-	return 0;
-}
-
-int hdmi_phy_init(struct platform_device *pdev, struct hdmi_phy_data *phy)
+int hdmi_phy_init(struct platform_device *pdev, struct hdmi_phy_data *phy,
+		  unsigned int version)
 {
-	int r;
 	struct resource *res;
 
-	r = hdmi_phy_init_features(pdev);
-	if (r)
-		return r;
+	if (version == 4)
+		phy->features = &omap44xx_phy_feats;
+	else
+		phy->features = &omap54xx_phy_feats;
 
 	res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "phy");
 	phy->base = devm_ioremap_resource(&pdev->dev, res);
diff --git a/drivers/gpu/drm/omapdrm/dss/hdmi_pll.c b/drivers/gpu/drm/omapdrm/dss/hdmi_pll.c
index 46239358655a..55bee81f4dd5 100644
--- a/drivers/gpu/drm/omapdrm/dss/hdmi_pll.c
+++ b/drivers/gpu/drm/omapdrm/dss/hdmi_pll.c
@@ -71,7 +71,7 @@ static void hdmi_pll_disable(struct dss_pll *dsspll)
 	WARN_ON(r < 0 && r != -ENOSYS);
 }
 
-static const struct dss_pll_ops dsi_pll_ops = {
+static const struct dss_pll_ops hdmi_pll_ops = {
 	.enable = hdmi_pll_enable,
 	.disable = hdmi_pll_disable,
 	.set_config = dss_pll_write_config_type_b,
@@ -128,7 +128,8 @@ static const struct dss_pll_hw dss_omap5_hdmi_pll_hw = {
 	.has_refsel = true,
 };
 
-static int dsi_init_pll_data(struct platform_device *pdev, struct hdmi_pll_data *hpll)
+static int hdmi_init_pll_data(struct platform_device *pdev,
+			      struct hdmi_pll_data *hpll)
 {
 	struct dss_pll *pll = &hpll->pll;
 	struct clk *clk;
@@ -145,23 +146,12 @@ static int dsi_init_pll_data(struct platform_device *pdev, struct hdmi_pll_data
 	pll->base = hpll->base;
 	pll->clkin = clk;
 
-	switch (omapdss_get_version()) {
-	case OMAPDSS_VER_OMAP4430_ES1:
-	case OMAPDSS_VER_OMAP4430_ES2:
-	case OMAPDSS_VER_OMAP4:
+	if (hpll->wp->version == 4)
 		pll->hw = &dss_omap4_hdmi_pll_hw;
-		break;
-
-	case OMAPDSS_VER_OMAP5:
-	case OMAPDSS_VER_DRA7xx:
+	else
 		pll->hw = &dss_omap5_hdmi_pll_hw;
-		break;
-
-	default:
-		return -ENODEV;
-	}
 
-	pll->ops = &dsi_pll_ops;
+	pll->ops = &hdmi_pll_ops;
 
 	r = dss_pll_register(pll);
 	if (r)
@@ -184,7 +174,7 @@ int hdmi_pll_init(struct platform_device *pdev, struct hdmi_pll_data *pll,
 	if (IS_ERR(pll->base))
 		return PTR_ERR(pll->base);
 
-	r = dsi_init_pll_data(pdev, pll);
+	r = hdmi_init_pll_data(pdev, pll);
 	if (r) {
 		DSSERR("failed to init HDMI PLL\n");
 		return r;
diff --git a/drivers/gpu/drm/omapdrm/dss/hdmi_wp.c b/drivers/gpu/drm/omapdrm/dss/hdmi_wp.c
index ab129df2e310..88034fbe0e9f 100644
--- a/drivers/gpu/drm/omapdrm/dss/hdmi_wp.c
+++ b/drivers/gpu/drm/omapdrm/dss/hdmi_wp.c
@@ -178,9 +178,7 @@ void hdmi_wp_video_config_timing(struct hdmi_wp_data *wp,
 	 * However, we don't support OMAP5 ES1 at all, so we can just check for
 	 * OMAP4 here.
 	 */
-	if (omapdss_get_version() == OMAPDSS_VER_OMAP4430_ES1 ||
-	    omapdss_get_version() == OMAPDSS_VER_OMAP4430_ES2 ||
-	    omapdss_get_version() == OMAPDSS_VER_OMAP4)
+	if (wp->version == 4)
 		hsync_len_offset = 0;
 
 	timing_h |= FLD_VAL(vm->hback_porch, 31, 20);
@@ -235,9 +233,7 @@ void hdmi_wp_audio_config_format(struct hdmi_wp_data *wp,
 	DSSDBG("Enter hdmi_wp_audio_config_format\n");
 
 	r = hdmi_read_reg(wp->base, HDMI_WP_AUDIO_CFG);
-	if (omapdss_get_version() == OMAPDSS_VER_OMAP4430_ES1 ||
-	    omapdss_get_version() == OMAPDSS_VER_OMAP4430_ES2 ||
-	    omapdss_get_version() == OMAPDSS_VER_OMAP4) {
+	if (wp->version == 4) {
 		r = FLD_MOD(r, aud_fmt->stereo_channels, 26, 24);
 		r = FLD_MOD(r, aud_fmt->active_chnnls_msk, 23, 16);
 	}
@@ -282,7 +278,8 @@ int hdmi_wp_audio_core_req_enable(struct hdmi_wp_data *wp, bool enable)
 	return 0;
 }
 
-int hdmi_wp_init(struct platform_device *pdev, struct hdmi_wp_data *wp)
+int hdmi_wp_init(struct platform_device *pdev, struct hdmi_wp_data *wp,
+		 unsigned int version)
 {
 	struct resource *res;
 
@@ -292,6 +289,7 @@ int hdmi_wp_init(struct platform_device *pdev, struct hdmi_wp_data *wp)
 		return PTR_ERR(wp->base);
 
 	wp->phys_base = res->start;
+	wp->version = version;
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/omapdrm/dss/omapdss.h b/drivers/gpu/drm/omapdrm/dss/omapdss.h
index 85953a0bc7c2..47a331670963 100644
--- a/drivers/gpu/drm/omapdrm/dss/omapdss.h
+++ b/drivers/gpu/drm/omapdrm/dss/omapdss.h
@@ -25,6 +25,7 @@
 #include <video/videomode.h>
 #include <linux/platform_data/omapdss.h>
 #include <uapi/drm/drm_mode.h>
+#include <drm/drm_crtc.h>
 
 #define DISPC_IRQ_FRAMEDONE		(1 << 0)
 #define DISPC_IRQ_VSYNC			(1 << 1)
@@ -241,13 +242,6 @@ struct omap_dss_dsi_config {
 	enum omap_dss_dsi_trans_mode trans_mode;
 };
 
-/* Hardcoded videomodes for tv. Venc only uses these to
- * identify the mode, and does not actually use the configs
- * itself. However, the configs should be something that
- * a normal monitor can also show */
-extern const struct videomode omap_dss_pal_vm;
-extern const struct videomode omap_dss_ntsc_vm;
-
 struct omap_dss_cpr_coefs {
 	s16 rr, rg, rb;
 	s16 gr, gg, gb;
@@ -403,6 +397,14 @@ struct omapdss_hdmi_ops {
 	int (*read_edid)(struct omap_dss_device *dssdev, u8 *buf, int len);
 	bool (*detect)(struct omap_dss_device *dssdev);
 
+	int (*register_hpd_cb)(struct omap_dss_device *dssdev,
+			       void (*cb)(void *cb_data,
+					  enum drm_connector_status status),
+			       void *cb_data);
+	void (*unregister_hpd_cb)(struct omap_dss_device *dssdev);
+	void (*enable_hpd)(struct omap_dss_device *dssdev);
+	void (*disable_hpd)(struct omap_dss_device *dssdev);
+
 	int (*set_hdmi_mode)(struct omap_dss_device *dssdev, bool hdmi_mode);
 	int (*set_infoframe)(struct omap_dss_device *dssdev,
 		const struct hdmi_avi_infoframe *avi);
@@ -567,12 +569,19 @@ struct omap_dss_driver {
 	int (*read_edid)(struct omap_dss_device *dssdev, u8 *buf, int len);
 	bool (*detect)(struct omap_dss_device *dssdev);
 
+	int (*register_hpd_cb)(struct omap_dss_device *dssdev,
+			       void (*cb)(void *cb_data,
+					  enum drm_connector_status status),
+			       void *cb_data);
+	void (*unregister_hpd_cb)(struct omap_dss_device *dssdev);
+	void (*enable_hpd)(struct omap_dss_device *dssdev);
+	void (*disable_hpd)(struct omap_dss_device *dssdev);
+
 	int (*set_hdmi_mode)(struct omap_dss_device *dssdev, bool hdmi_mode);
 	int (*set_hdmi_infoframe)(struct omap_dss_device *dssdev,
 		const struct hdmi_avi_infoframe *avi);
 };
 
-enum omapdss_version omapdss_get_version(void);
 bool omapdss_is_initialized(void);
 
 int omap_dss_register_driver(struct omap_dss_driver *);
diff --git a/drivers/gpu/drm/omapdrm/dss/pll.c b/drivers/gpu/drm/omapdrm/dss/pll.c
index 5e221302768b..9d9d9d42009b 100644
--- a/drivers/gpu/drm/omapdrm/dss/pll.c
+++ b/drivers/gpu/drm/omapdrm/dss/pll.c
@@ -215,8 +215,8 @@ bool dss_pll_calc_a(const struct dss_pll *pll, unsigned long clkin,
 		dss_pll_calc_func func, void *data)
 {
 	const struct dss_pll_hw *hw = pll->hw;
-	int n, n_min, n_max;
-	int m, m_min, m_max;
+	int n, n_start, n_stop, n_inc;
+	int m, m_start, m_stop, m_inc;
 	unsigned long fint, clkdco;
 	unsigned long pll_hw_max;
 	unsigned long fint_hw_min, fint_hw_max;
@@ -226,22 +226,33 @@ bool dss_pll_calc_a(const struct dss_pll *pll, unsigned long clkin,
 	fint_hw_min = hw->fint_min;
 	fint_hw_max = hw->fint_max;
 
-	n_min = max(DIV_ROUND_UP(clkin, fint_hw_max), 1ul);
-	n_max = min((unsigned)(clkin / fint_hw_min), hw->n_max);
+	n_start = max(DIV_ROUND_UP(clkin, fint_hw_max), 1ul);
+	n_stop = min((unsigned)(clkin / fint_hw_min), hw->n_max);
+	n_inc = 1;
+
+	if (hw->errata_i886) {
+		swap(n_start, n_stop);
+		n_inc = -1;
+	}
 
 	pll_max = pll_max ? pll_max : ULONG_MAX;
 
-	/* Try to find high N & M to avoid jitter (DRA7 errata i886) */
-	for (n = n_max; n >= n_min; --n) {
+	for (n = n_start; n != n_stop; n += n_inc) {
 		fint = clkin / n;
 
-		m_min = max(DIV_ROUND_UP(DIV_ROUND_UP(pll_min, fint), 2),
+		m_start = max(DIV_ROUND_UP(DIV_ROUND_UP(pll_min, fint), 2),
 				1ul);
-		m_max = min3((unsigned)(pll_max / fint / 2),
+		m_stop = min3((unsigned)(pll_max / fint / 2),
 				(unsigned)(pll_hw_max / fint / 2),
 				hw->m_max);
+		m_inc = 1;
+
+		if (hw->errata_i886) {
+			swap(m_start, m_stop);
+			m_inc = -1;
+		}
 
-		for (m = m_max; m >= m_min; --m) {
+		for (m = m_start; m != m_stop; m += m_inc) {
 			clkdco = 2 * m * fint;
 
 			if (func(n, m, fint, clkdco, data))
diff --git a/drivers/gpu/drm/omapdrm/dss/venc.c b/drivers/gpu/drm/omapdrm/dss/venc.c
index a6bfb3918b8d..d58da6f32693 100644
--- a/drivers/gpu/drm/omapdrm/dss/venc.c
+++ b/drivers/gpu/drm/omapdrm/dss/venc.c
@@ -37,10 +37,10 @@
 #include <linux/of.h>
 #include <linux/of_graph.h>
 #include <linux/component.h>
+#include <linux/sys_soc.h>
 
 #include "omapdss.h"
 #include "dss.h"
-#include "dss_features.h"
 
 /* Venc registers */
 #define VENC_REV_ID				0x00
@@ -263,7 +263,13 @@ static const struct venc_config venc_config_pal_bdghi = {
 	.fid_ext_start_y__fid_ext_offset_y	= 0x01380005,
 };
 
-const struct videomode omap_dss_pal_vm = {
+enum venc_videomode {
+	VENC_MODE_UNKNOWN,
+	VENC_MODE_PAL,
+	VENC_MODE_NTSC,
+};
+
+static const struct videomode omap_dss_pal_vm = {
 	.hactive	= 720,
 	.vactive	= 574,
 	.pixelclock	= 13500000,
@@ -279,9 +285,8 @@ const struct videomode omap_dss_pal_vm = {
 			  DISPLAY_FLAGS_PIXDATA_POSEDGE |
 			  DISPLAY_FLAGS_SYNC_NEGEDGE,
 };
-EXPORT_SYMBOL(omap_dss_pal_vm);
 
-const struct videomode omap_dss_ntsc_vm = {
+static const struct videomode omap_dss_ntsc_vm = {
 	.hactive	= 720,
 	.vactive	= 482,
 	.pixelclock	= 13500000,
@@ -297,7 +302,24 @@ const struct videomode omap_dss_ntsc_vm = {
 			  DISPLAY_FLAGS_PIXDATA_POSEDGE |
 			  DISPLAY_FLAGS_SYNC_NEGEDGE,
 };
-EXPORT_SYMBOL(omap_dss_ntsc_vm);
+
+static enum venc_videomode venc_get_videomode(const struct videomode *vm)
+{
+	if (!(vm->flags & DISPLAY_FLAGS_INTERLACED))
+		return VENC_MODE_UNKNOWN;
+
+	if (vm->pixelclock == omap_dss_pal_vm.pixelclock &&
+	    vm->hactive == omap_dss_pal_vm.hactive &&
+	    vm->vactive == omap_dss_pal_vm.vactive)
+		return VENC_MODE_PAL;
+
+	if (vm->pixelclock == omap_dss_ntsc_vm.pixelclock &&
+	    vm->hactive == omap_dss_ntsc_vm.hactive &&
+	    vm->vactive == omap_dss_ntsc_vm.vactive)
+		return VENC_MODE_NTSC;
+
+	return VENC_MODE_UNKNOWN;
+}
 
 static struct {
 	struct platform_device *pdev;
@@ -311,6 +333,7 @@ static struct {
 	struct videomode vm;
 	enum omap_dss_venc_type type;
 	bool invert_polarity;
+	bool requires_tv_dac_clk;
 
 	struct omap_dss_device output;
 } venc;
@@ -424,14 +447,14 @@ static void venc_runtime_put(void)
 
 static const struct venc_config *venc_timings_to_config(struct videomode *vm)
 {
-	if (memcmp(&omap_dss_pal_vm, vm, sizeof(*vm)) == 0)
+	switch (venc_get_videomode(vm)) {
+	default:
+		WARN_ON_ONCE(1);
+	case VENC_MODE_PAL:
 		return &venc_config_pal_trm;
-
-	if (memcmp(&omap_dss_ntsc_vm, vm, sizeof(*vm)) == 0)
+	case VENC_MODE_NTSC:
 		return &venc_config_ntsc_trm;
-
-	BUG();
-	return NULL;
+	}
 }
 
 static int venc_power_on(struct omap_dss_device *dssdev)
@@ -542,15 +565,28 @@ static void venc_display_disable(struct omap_dss_device *dssdev)
 static void venc_set_timings(struct omap_dss_device *dssdev,
 			     struct videomode *vm)
 {
+	struct videomode actual_vm;
+
 	DSSDBG("venc_set_timings\n");
 
 	mutex_lock(&venc.venc_lock);
 
+	switch (venc_get_videomode(vm)) {
+	default:
+		WARN_ON_ONCE(1);
+	case VENC_MODE_PAL:
+		actual_vm = omap_dss_pal_vm;
+		break;
+	case VENC_MODE_NTSC:
+		actual_vm = omap_dss_ntsc_vm;
+		break;
+	}
+
 	/* Reset WSS data when the TV standard changes. */
-	if (memcmp(&venc.vm, vm, sizeof(*vm)))
+	if (memcmp(&venc.vm, &actual_vm, sizeof(actual_vm)))
 		venc.wss_data = 0;
 
-	venc.vm = *vm;
+	venc.vm = actual_vm;
 
 	dispc_set_tv_pclk(13500000);
 
@@ -562,13 +598,13 @@ static int venc_check_timings(struct omap_dss_device *dssdev,
 {
 	DSSDBG("venc_check_timings\n");
 
-	if (memcmp(&omap_dss_pal_vm, vm, sizeof(*vm)) == 0)
-		return 0;
-
-	if (memcmp(&omap_dss_ntsc_vm, vm, sizeof(*vm)) == 0)
+	switch (venc_get_videomode(vm)) {
+	case VENC_MODE_PAL:
+	case VENC_MODE_NTSC:
 		return 0;
-
-	return -EINVAL;
+	default:
+		return -EINVAL;
+	}
 }
 
 static void venc_get_timings(struct omap_dss_device *dssdev,
@@ -693,7 +729,7 @@ static int venc_get_clocks(struct platform_device *pdev)
 {
 	struct clk *clk;
 
-	if (dss_has_feature(FEAT_VENC_REQUIRES_TV_DAC_CLK)) {
+	if (venc.requires_tv_dac_clk) {
 		clk = devm_clk_get(&pdev->dev, "tv_dac_clk");
 		if (IS_ERR(clk)) {
 			DSSERR("can't get tv_dac_clk\n");
@@ -828,6 +864,12 @@ err:
 }
 
 /* VENC HW IP initialisation */
+static const struct soc_device_attribute venc_soc_devices[] = {
+	{ .machine = "OMAP3[45]*" },
+	{ .machine = "AM35*" },
+	{ /* sentinel */ }
+};
+
 static int venc_bind(struct device *dev, struct device *master, void *data)
 {
 	struct platform_device *pdev = to_platform_device(dev);
@@ -837,6 +879,10 @@ static int venc_bind(struct device *dev, struct device *master, void *data)
 
 	venc.pdev = pdev;
 
+	/* The OMAP34xx, OMAP35xx and AM35xx VENC require the TV DAC clock. */
+	if (soc_device_match(venc_soc_devices))
+		venc.requires_tv_dac_clk = true;
+
 	mutex_init(&venc.venc_lock);
 
 	venc.wss_data = 0;
diff --git a/drivers/gpu/drm/omapdrm/dss/video-pll.c b/drivers/gpu/drm/omapdrm/dss/video-pll.c
index fbd1263a29a4..38a239cc5e04 100644
--- a/drivers/gpu/drm/omapdrm/dss/video-pll.c
+++ b/drivers/gpu/drm/omapdrm/dss/video-pll.c
@@ -19,7 +19,6 @@
 
 #include "omapdss.h"
 #include "dss.h"
-#include "dss_features.h"
 
 struct dss_video_pll {
 	struct dss_pll pll;
@@ -131,6 +130,8 @@ static const struct dss_pll_hw dss_dra7_video_pll_hw = {
 	.mX_lsb[3] = 5,
 
 	.has_refsel = true,
+
+	.errata_i886 = true,
 };
 
 struct dss_pll *dss_video_pll_init(struct platform_device *pdev, int id,
diff --git a/drivers/gpu/drm/omapdrm/omap_connector.c b/drivers/gpu/drm/omapdrm/omap_connector.c
index d1ec76ef5cc6..aa5ba9ae2191 100644
--- a/drivers/gpu/drm/omapdrm/omap_connector.c
+++ b/drivers/gpu/drm/omapdrm/omap_connector.c
@@ -35,6 +35,23 @@ struct omap_connector {
 	bool hdmi_mode;
 };
 
+static void omap_connector_hpd_cb(void *cb_data,
+				  enum drm_connector_status status)
+{
+	struct omap_connector *omap_connector = cb_data;
+	struct drm_connector *connector = &omap_connector->base;
+	struct drm_device *dev = connector->dev;
+	enum drm_connector_status old_status;
+
+	mutex_lock(&dev->mode_config.mutex);
+	old_status = connector->status;
+	connector->status = status;
+	mutex_unlock(&dev->mode_config.mutex);
+
+	if (old_status != status)
+		drm_kms_helper_hotplug_event(dev);
+}
+
 bool omap_connector_get_hdmi_mode(struct drm_connector *connector)
 {
 	struct omap_connector *omap_connector = to_omap_connector(connector);
@@ -75,6 +92,10 @@ static void omap_connector_destroy(struct drm_connector *connector)
 	struct omap_dss_device *dssdev = omap_connector->dssdev;
 
 	DBG("%s", omap_connector->dssdev->name);
+	if (connector->polled == DRM_CONNECTOR_POLL_HPD &&
+	    dssdev->driver->unregister_hpd_cb) {
+		dssdev->driver->unregister_hpd_cb(dssdev);
+	}
 	drm_connector_unregister(connector);
 	drm_connector_cleanup(connector);
 	kfree(omap_connector);
@@ -215,6 +236,7 @@ struct drm_connector *omap_connector_init(struct drm_device *dev,
 {
 	struct drm_connector *connector = NULL;
 	struct omap_connector *omap_connector;
+	bool hpd_supported = false;
 
 	DBG("%s", dssdev->name);
 
@@ -232,7 +254,20 @@ struct drm_connector *omap_connector_init(struct drm_device *dev,
 				connector_type);
 	drm_connector_helper_add(connector, &omap_connector_helper_funcs);
 
-	if (dssdev->driver->detect)
+	if (dssdev->driver->register_hpd_cb) {
+		int ret = dssdev->driver->register_hpd_cb(dssdev,
+							  omap_connector_hpd_cb,
+							  omap_connector);
+		if (!ret)
+			hpd_supported = true;
+		else if (ret != -ENOTSUPP)
+			DBG("%s: Failed to register HPD callback (%d).",
+			    dssdev->name, ret);
+	}
+
+	if (hpd_supported)
+		connector->polled = DRM_CONNECTOR_POLL_HPD;
+	else if (dssdev->driver->detect)
 		connector->polled = DRM_CONNECTOR_POLL_CONNECT |
 				    DRM_CONNECTOR_POLL_DISCONNECT;
 	else
diff --git a/drivers/gpu/drm/omapdrm/omap_crtc.c b/drivers/gpu/drm/omapdrm/omap_crtc.c
index 400d0d2f6790..cc85c16cbc2a 100644
--- a/drivers/gpu/drm/omapdrm/omap_crtc.c
+++ b/drivers/gpu/drm/omapdrm/omap_crtc.c
@@ -589,8 +589,10 @@ omap_crtc_duplicate_state(struct drm_crtc *crtc)
 	current_state = to_omap_crtc_state(crtc->state);
 
 	state = kmalloc(sizeof(*state), GFP_KERNEL);
-	if (state)
-		__drm_atomic_helper_crtc_duplicate_state(crtc, &state->base);
+	if (!state)
+		return NULL;
+
+	__drm_atomic_helper_crtc_duplicate_state(crtc, &state->base);
 
 	state->zpos = current_state->zpos;
 	state->rotation = current_state->rotation;
diff --git a/drivers/gpu/drm/omapdrm/omap_drv.c b/drivers/gpu/drm/omapdrm/omap_drv.c
index 721a358531b0..cdf5b0601eba 100644
--- a/drivers/gpu/drm/omapdrm/omap_drv.c
+++ b/drivers/gpu/drm/omapdrm/omap_drv.c
@@ -84,23 +84,36 @@ static void omap_atomic_commit_tail(struct drm_atomic_state *old_state)
 	/* Apply the atomic update. */
 	drm_atomic_helper_commit_modeset_disables(dev, old_state);
 
-	/* With the current dss dispc implementation we have to enable
-	 * the new modeset before we can commit planes. The dispc ovl
-	 * configuration relies on the video mode configuration been
-	 * written into the HW when the ovl configuration is
-	 * calculated.
-	 *
-	 * This approach is not ideal because after a mode change the
-	 * plane update is executed only after the first vblank
-	 * interrupt. The dispc implementation should be fixed so that
-	 * it is able use uncommitted drm state information.
-	 */
-	drm_atomic_helper_commit_modeset_enables(dev, old_state);
-	omap_atomic_wait_for_completion(dev, old_state);
-
-	drm_atomic_helper_commit_planes(dev, old_state, 0);
-
-	drm_atomic_helper_commit_hw_done(old_state);
+	if (priv->omaprev != 0x3430) {
+		/* With the current dss dispc implementation we have to enable
+		 * the new modeset before we can commit planes. The dispc ovl
+		 * configuration relies on the video mode configuration been
+		 * written into the HW when the ovl configuration is
+		 * calculated.
+		 *
+		 * This approach is not ideal because after a mode change the
+		 * plane update is executed only after the first vblank
+		 * interrupt. The dispc implementation should be fixed so that
+		 * it is able use uncommitted drm state information.
+		 */
+		drm_atomic_helper_commit_modeset_enables(dev, old_state);
+		omap_atomic_wait_for_completion(dev, old_state);
+
+		drm_atomic_helper_commit_planes(dev, old_state, 0);
+
+		drm_atomic_helper_commit_hw_done(old_state);
+	} else {
+		/*
+		 * OMAP3 DSS seems to have issues with the work-around above,
+		 * resulting in endless sync losts if a crtc is enabled without
+		 * a plane. For now, skip the WA for OMAP3.
+		 */
+		drm_atomic_helper_commit_planes(dev, old_state, 0);
+
+		drm_atomic_helper_commit_modeset_enables(dev, old_state);
+
+		drm_atomic_helper_commit_hw_done(old_state);
+	}
 
 	/*
 	 * Wait for completion of the page flips to ensure that old buffers
@@ -324,6 +337,32 @@ static int omap_modeset_init(struct drm_device *dev)
 }
 
 /*
+ * Enable the HPD in external components if supported
+ */
+static void omap_modeset_enable_external_hpd(void)
+{
+	struct omap_dss_device *dssdev = NULL;
+
+	for_each_dss_dev(dssdev) {
+		if (dssdev->driver->enable_hpd)
+			dssdev->driver->enable_hpd(dssdev);
+	}
+}
+
+/*
+ * Disable the HPD in external components if supported
+ */
+static void omap_modeset_disable_external_hpd(void)
+{
+	struct omap_dss_device *dssdev = NULL;
+
+	for_each_dss_dev(dssdev) {
+		if (dssdev->driver->disable_hpd)
+			dssdev->driver->disable_hpd(dssdev);
+	}
+}
+
+/*
  * drm ioctl funcs
  */
 
@@ -438,44 +477,11 @@ static int dev_open(struct drm_device *dev, struct drm_file *file)
  */
 static void dev_lastclose(struct drm_device *dev)
 {
-	int i;
-
-	/* we don't support vga_switcheroo.. so just make sure the fbdev
-	 * mode is active
-	 */
 	struct omap_drm_private *priv = dev->dev_private;
 	int ret;
 
 	DBG("lastclose: dev=%p", dev);
 
-	/* need to restore default rotation state.. not sure
-	 * if there is a cleaner way to restore properties to
-	 * default state?  Maybe a flag that properties should
-	 * automatically be restored to default state on
-	 * lastclose?
-	 */
-	for (i = 0; i < priv->num_crtcs; i++) {
-		struct drm_crtc *crtc = priv->crtcs[i];
-
-		if (!crtc->primary->rotation_property)
-			continue;
-
-		drm_object_property_set_value(&crtc->base,
-					      crtc->primary->rotation_property,
-					      DRM_MODE_ROTATE_0);
-	}
-
-	for (i = 0; i < priv->num_planes; i++) {
-		struct drm_plane *plane = priv->planes[i];
-
-		if (!plane->rotation_property)
-			continue;
-
-		drm_object_property_set_value(&plane->base,
-					      plane->rotation_property,
-					      DRM_MODE_ROTATE_0);
-	}
-
 	if (priv->fbdev) {
 		ret = drm_fb_helper_restore_fbdev_mode_unlocked(priv->fbdev);
 		if (ret)
@@ -549,6 +555,12 @@ static int pdev_probe(struct platform_device *pdev)
 	if (omapdss_is_initialized() == false)
 		return -EPROBE_DEFER;
 
+	ret = dma_set_coherent_mask(&pdev->dev, DMA_BIT_MASK(32));
+	if (ret) {
+		dev_err(&pdev->dev, "Failed to set the DMA mask\n");
+		return ret;
+	}
+
 	omap_crtc_pre_init();
 
 	ret = omap_connect_dssdevs();
@@ -602,6 +614,7 @@ static int pdev_probe(struct platform_device *pdev)
 	priv->fbdev = omap_fbdev_init(ddev);
 
 	drm_kms_helper_poll_init(ddev);
+	omap_modeset_enable_external_hpd();
 
 	/*
 	 * Register the DRM device with the core and the connectors with
@@ -614,6 +627,7 @@ static int pdev_probe(struct platform_device *pdev)
 	return 0;
 
 err_cleanup_helpers:
+	omap_modeset_disable_external_hpd();
 	drm_kms_helper_poll_fini(ddev);
 	if (priv->fbdev)
 		omap_fbdev_free(ddev);
@@ -642,6 +656,7 @@ static int pdev_remove(struct platform_device *pdev)
 
 	drm_dev_unregister(ddev);
 
+	omap_modeset_disable_external_hpd();
 	drm_kms_helper_poll_fini(ddev);
 
 	if (priv->fbdev)
@@ -733,7 +748,7 @@ static SIMPLE_DEV_PM_OPS(omapdrm_pm_ops, omap_drm_suspend, omap_drm_resume);
 
 static struct platform_driver pdev = {
 	.driver = {
-		.name = DRIVER_NAME,
+		.name = "omapdrm",
 		.pm = &omapdrm_pm_ops,
 	},
 	.probe = pdev_probe,
diff --git a/drivers/gpu/drm/omapdrm/omap_fb.c b/drivers/gpu/drm/omapdrm/omap_fb.c
index ddf7a457951b..b1a762b70cbf 100644
--- a/drivers/gpu/drm/omapdrm/omap_fb.c
+++ b/drivers/gpu/drm/omapdrm/omap_fb.c
@@ -379,7 +379,7 @@ struct drm_framebuffer *omap_framebuffer_create(struct drm_device *dev,
 	return fb;
 
 error:
-	while (--i > 0)
+	while (--i >= 0)
 		drm_gem_object_unreference_unlocked(bos[i]);
 
 	return fb;
diff --git a/drivers/gpu/drm/omapdrm/omap_gem_dmabuf.c b/drivers/gpu/drm/omapdrm/omap_gem_dmabuf.c
index 863a881dd7cd..afdbad5c866a 100644
--- a/drivers/gpu/drm/omapdrm/omap_gem_dmabuf.c
+++ b/drivers/gpu/drm/omapdrm/omap_gem_dmabuf.c
@@ -144,7 +144,7 @@ static int omap_gem_dmabuf_mmap(struct dma_buf *buffer,
 	return omap_gem_mmap_obj(obj, vma);
 }
 
-static struct dma_buf_ops omap_dmabuf_ops = {
+static const struct dma_buf_ops omap_dmabuf_ops = {
 	.map_dma_buf = omap_gem_map_dma_buf,
 	.unmap_dma_buf = omap_gem_unmap_dma_buf,
 	.release = drm_gem_dmabuf_release,
diff --git a/drivers/gpu/drm/pl111/pl111_display.c b/drivers/gpu/drm/pl111/pl111_display.c
index f0139fa58d55..b58c988d9da0 100644
--- a/drivers/gpu/drm/pl111/pl111_display.c
+++ b/drivers/gpu/drm/pl111/pl111_display.c
@@ -23,6 +23,7 @@
 #include <drm/drmP.h>
 #include <drm/drm_panel.h>
 #include <drm/drm_gem_cma_helper.h>
+#include <drm/drm_gem_framebuffer_helper.h>
 #include <drm/drm_fb_cma_helper.h>
 
 #include "pl111_drm.h"
@@ -274,7 +275,7 @@ void pl111_disable_vblank(struct drm_device *drm, unsigned int crtc)
 static int pl111_display_prepare_fb(struct drm_simple_display_pipe *pipe,
 				    struct drm_plane_state *plane_state)
 {
-	return drm_fb_cma_prepare_fb(&pipe->plane, plane_state);
+	return drm_gem_fb_prepare_fb(&pipe->plane, plane_state);
 }
 
 static const struct drm_simple_display_pipe_funcs pl111_display_funcs = {
diff --git a/drivers/gpu/drm/pl111/pl111_drv.c b/drivers/gpu/drm/pl111/pl111_drv.c
index 29653fe5285c..581c452cede1 100644
--- a/drivers/gpu/drm/pl111/pl111_drv.c
+++ b/drivers/gpu/drm/pl111/pl111_drv.c
@@ -66,14 +66,15 @@
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc_helper.h>
 #include <drm/drm_gem_cma_helper.h>
+#include <drm/drm_gem_framebuffer_helper.h>
 #include <drm/drm_fb_cma_helper.h>
 
 #include "pl111_drm.h"
 
 #define DRIVER_DESC      "DRM module for PL111"
 
-static struct drm_mode_config_funcs mode_config_funcs = {
-	.fb_create = drm_fb_cma_create,
+static const struct drm_mode_config_funcs mode_config_funcs = {
+	.fb_create = drm_gem_fb_create,
 	.atomic_check = drm_atomic_helper_check,
 	.atomic_commit = drm_atomic_helper_commit,
 };
diff --git a/drivers/gpu/drm/qxl/qxl_drv.c b/drivers/gpu/drm/qxl/qxl_drv.c
index 403e135895bf..2445e75cf7ea 100644
--- a/drivers/gpu/drm/qxl/qxl_drv.c
+++ b/drivers/gpu/drm/qxl/qxl_drv.c
@@ -263,7 +263,6 @@ static struct drm_driver qxl_driver = {
 
 	.dumb_create = qxl_mode_dumb_create,
 	.dumb_map_offset = qxl_mode_dumb_mmap,
-	.dumb_destroy = drm_gem_dumb_destroy,
 #if defined(CONFIG_DEBUG_FS)
 	.debugfs_init = qxl_debugfs_init,
 #endif
diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h
index 5008f3d4cccc..ec63bc5e9de7 100644
--- a/drivers/gpu/drm/radeon/radeon.h
+++ b/drivers/gpu/drm/radeon/radeon.h
@@ -464,7 +464,7 @@ struct radeon_bo_list {
 	struct radeon_bo		*robj;
 	struct ttm_validate_buffer	tv;
 	uint64_t			gpu_offset;
-	unsigned			prefered_domains;
+	unsigned			preferred_domains;
 	unsigned			allowed_domains;
 	uint32_t			tiling_flags;
 };
@@ -2327,7 +2327,7 @@ struct radeon_device {
 	uint8_t				*bios;
 	bool				is_atom_bios;
 	uint16_t			bios_header_start;
-	struct radeon_bo		*stollen_vga_memory;
+	struct radeon_bo		*stolen_vga_memory;
 	/* Register mmio */
 	resource_size_t			rmmio_base;
 	resource_size_t			rmmio_size;
diff --git a/drivers/gpu/drm/radeon/radeon_acpi.c b/drivers/gpu/drm/radeon/radeon_acpi.c
index 6efbd65c929e..8d3251a10cd4 100644
--- a/drivers/gpu/drm/radeon/radeon_acpi.c
+++ b/drivers/gpu/drm/radeon/radeon_acpi.c
@@ -351,7 +351,7 @@ out:
  * handles it.
  * Returns NOTIFY code
  */
-int radeon_atif_handler(struct radeon_device *rdev,
+static int radeon_atif_handler(struct radeon_device *rdev,
 		struct acpi_bus_event *event)
 {
 	struct radeon_atif *atif = &rdev->atif;
diff --git a/drivers/gpu/drm/radeon/radeon_acpi.h b/drivers/gpu/drm/radeon/radeon_acpi.h
index 7af1977c2c68..35202a453e66 100644
--- a/drivers/gpu/drm/radeon/radeon_acpi.h
+++ b/drivers/gpu/drm/radeon/radeon_acpi.h
@@ -27,9 +27,6 @@
 struct radeon_device;
 struct acpi_bus_event;
 
-int radeon_atif_handler(struct radeon_device *rdev,
-		struct acpi_bus_event *event);
-
 /* AMD hw uses four ACPI control methods:
  * 1. ATIF
  * ARG0: (ACPI_INTEGER) function code
diff --git a/drivers/gpu/drm/radeon/radeon_cs.c b/drivers/gpu/drm/radeon/radeon_cs.c
index 00b22af70f5c..1ae31dbc61c6 100644
--- a/drivers/gpu/drm/radeon/radeon_cs.c
+++ b/drivers/gpu/drm/radeon/radeon_cs.c
@@ -130,7 +130,7 @@ static int radeon_cs_parser_relocs(struct radeon_cs_parser *p)
 		     p->rdev->family == CHIP_RS880)) {
 
 			/* TODO: is this still needed for NI+ ? */
-			p->relocs[i].prefered_domains =
+			p->relocs[i].preferred_domains =
 				RADEON_GEM_DOMAIN_VRAM;
 
 			p->relocs[i].allowed_domains =
@@ -148,14 +148,14 @@ static int radeon_cs_parser_relocs(struct radeon_cs_parser *p)
 				return -EINVAL;
 			}
 
-			p->relocs[i].prefered_domains = domain;
+			p->relocs[i].preferred_domains = domain;
 			if (domain == RADEON_GEM_DOMAIN_VRAM)
 				domain |= RADEON_GEM_DOMAIN_GTT;
 			p->relocs[i].allowed_domains = domain;
 		}
 
 		if (radeon_ttm_tt_has_userptr(p->relocs[i].robj->tbo.ttm)) {
-			uint32_t domain = p->relocs[i].prefered_domains;
+			uint32_t domain = p->relocs[i].preferred_domains;
 			if (!(domain & RADEON_GEM_DOMAIN_GTT)) {
 				DRM_ERROR("Only RADEON_GEM_DOMAIN_GTT is "
 					  "allowed for userptr BOs\n");
@@ -163,7 +163,7 @@ static int radeon_cs_parser_relocs(struct radeon_cs_parser *p)
 			}
 			need_mmap_lock = true;
 			domain = RADEON_GEM_DOMAIN_GTT;
-			p->relocs[i].prefered_domains = domain;
+			p->relocs[i].preferred_domains = domain;
 			p->relocs[i].allowed_domains = domain;
 		}
 
@@ -437,7 +437,7 @@ static void radeon_cs_parser_fini(struct radeon_cs_parser *parser, int error, bo
 			if (bo == NULL)
 				continue;
 
-			drm_gem_object_unreference_unlocked(&bo->gem_base);
+			drm_gem_object_put_unlocked(&bo->gem_base);
 		}
 	}
 	kfree(parser->track);
diff --git a/drivers/gpu/drm/radeon/radeon_cursor.c b/drivers/gpu/drm/radeon/radeon_cursor.c
index 4a4f9533c53b..91952277557e 100644
--- a/drivers/gpu/drm/radeon/radeon_cursor.c
+++ b/drivers/gpu/drm/radeon/radeon_cursor.c
@@ -307,7 +307,7 @@ int radeon_crtc_cursor_set2(struct drm_crtc *crtc,
 	robj = gem_to_radeon_bo(obj);
 	ret = radeon_bo_reserve(robj, false);
 	if (ret != 0) {
-		drm_gem_object_unreference_unlocked(obj);
+		drm_gem_object_put_unlocked(obj);
 		return ret;
 	}
 	/* Only 27 bit offset for legacy cursor */
@@ -317,7 +317,7 @@ int radeon_crtc_cursor_set2(struct drm_crtc *crtc,
 	radeon_bo_unreserve(robj);
 	if (ret) {
 		DRM_ERROR("Failed to pin new cursor BO (%d)\n", ret);
-		drm_gem_object_unreference_unlocked(obj);
+		drm_gem_object_put_unlocked(obj);
 		return ret;
 	}
 
@@ -352,7 +352,7 @@ unpin:
 			radeon_bo_unpin(robj);
 			radeon_bo_unreserve(robj);
 		}
-		drm_gem_object_unreference_unlocked(radeon_crtc->cursor_bo);
+		drm_gem_object_put_unlocked(radeon_crtc->cursor_bo);
 	}
 
 	radeon_crtc->cursor_bo = obj;
diff --git a/drivers/gpu/drm/radeon/radeon_display.c b/drivers/gpu/drm/radeon/radeon_display.c
index ee274c6e374d..ddfe91efa61e 100644
--- a/drivers/gpu/drm/radeon/radeon_display.c
+++ b/drivers/gpu/drm/radeon/radeon_display.c
@@ -267,7 +267,7 @@ static void radeon_unpin_work_func(struct work_struct *__work)
 	} else
 		DRM_ERROR("failed to reserve buffer after flip\n");
 
-	drm_gem_object_unreference_unlocked(&work->old_rbo->gem_base);
+	drm_gem_object_put_unlocked(&work->old_rbo->gem_base);
 	kfree(work);
 }
 
@@ -504,7 +504,7 @@ static int radeon_crtc_page_flip_target(struct drm_crtc *crtc,
 	obj = old_radeon_fb->obj;
 
 	/* take a reference to the old object */
-	drm_gem_object_reference(obj);
+	drm_gem_object_get(obj);
 	work->old_rbo = gem_to_radeon_bo(obj);
 
 	new_radeon_fb = to_radeon_framebuffer(fb);
@@ -603,7 +603,7 @@ pflip_cleanup:
 	radeon_bo_unreserve(new_rbo);
 
 cleanup:
-	drm_gem_object_unreference_unlocked(&work->old_rbo->gem_base);
+	drm_gem_object_put_unlocked(&work->old_rbo->gem_base);
 	dma_fence_put(work->fence);
 	kfree(work);
 	return r;
@@ -1288,7 +1288,7 @@ static void radeon_user_framebuffer_destroy(struct drm_framebuffer *fb)
 {
 	struct radeon_framebuffer *radeon_fb = to_radeon_framebuffer(fb);
 
-	drm_gem_object_unreference_unlocked(radeon_fb->obj);
+	drm_gem_object_put_unlocked(radeon_fb->obj);
 	drm_framebuffer_cleanup(fb);
 	kfree(radeon_fb);
 }
@@ -1348,14 +1348,14 @@ radeon_user_framebuffer_create(struct drm_device *dev,
 
 	radeon_fb = kzalloc(sizeof(*radeon_fb), GFP_KERNEL);
 	if (radeon_fb == NULL) {
-		drm_gem_object_unreference_unlocked(obj);
+		drm_gem_object_put_unlocked(obj);
 		return ERR_PTR(-ENOMEM);
 	}
 
 	ret = radeon_framebuffer_init(dev, radeon_fb, mode_cmd, obj);
 	if (ret) {
 		kfree(radeon_fb);
-		drm_gem_object_unreference_unlocked(obj);
+		drm_gem_object_put_unlocked(obj);
 		return ERR_PTR(ret);
 	}
 
diff --git a/drivers/gpu/drm/radeon/radeon_fb.c b/drivers/gpu/drm/radeon/radeon_fb.c
index af6ee7d9b465..fd25361ac681 100644
--- a/drivers/gpu/drm/radeon/radeon_fb.c
+++ b/drivers/gpu/drm/radeon/radeon_fb.c
@@ -118,7 +118,7 @@ static void radeonfb_destroy_pinned_object(struct drm_gem_object *gobj)
 		radeon_bo_unpin(rbo);
 		radeon_bo_unreserve(rbo);
 	}
-	drm_gem_object_unreference_unlocked(gobj);
+	drm_gem_object_put_unlocked(gobj);
 }
 
 static int radeonfb_create_pinned_object(struct radeon_fbdev *rfbdev,
@@ -299,7 +299,7 @@ out:
 
 	}
 	if (fb && ret) {
-		drm_gem_object_unreference_unlocked(gobj);
+		drm_gem_object_put_unlocked(gobj);
 		drm_framebuffer_unregister_private(fb);
 		drm_framebuffer_cleanup(fb);
 		kfree(fb);
diff --git a/drivers/gpu/drm/radeon/radeon_gem.c b/drivers/gpu/drm/radeon/radeon_gem.c
index 574bf7e6b118..3386452bd2f0 100644
--- a/drivers/gpu/drm/radeon/radeon_gem.c
+++ b/drivers/gpu/drm/radeon/radeon_gem.c
@@ -271,7 +271,7 @@ int radeon_gem_create_ioctl(struct drm_device *dev, void *data,
 	}
 	r = drm_gem_handle_create(filp, gobj, &handle);
 	/* drop reference from allocate - handle holds it now */
-	drm_gem_object_unreference_unlocked(gobj);
+	drm_gem_object_put_unlocked(gobj);
 	if (r) {
 		up_read(&rdev->exclusive_lock);
 		r = radeon_gem_handle_lockup(rdev, r);
@@ -352,7 +352,7 @@ int radeon_gem_userptr_ioctl(struct drm_device *dev, void *data,
 
 	r = drm_gem_handle_create(filp, gobj, &handle);
 	/* drop reference from allocate - handle holds it now */
-	drm_gem_object_unreference_unlocked(gobj);
+	drm_gem_object_put_unlocked(gobj);
 	if (r)
 		goto handle_lockup;
 
@@ -361,7 +361,7 @@ int radeon_gem_userptr_ioctl(struct drm_device *dev, void *data,
 	return 0;
 
 release_object:
-	drm_gem_object_unreference_unlocked(gobj);
+	drm_gem_object_put_unlocked(gobj);
 
 handle_lockup:
 	up_read(&rdev->exclusive_lock);
@@ -395,7 +395,7 @@ int radeon_gem_set_domain_ioctl(struct drm_device *dev, void *data,
 
 	r = radeon_gem_set_domain(gobj, args->read_domains, args->write_domain);
 
-	drm_gem_object_unreference_unlocked(gobj);
+	drm_gem_object_put_unlocked(gobj);
 	up_read(&rdev->exclusive_lock);
 	r = radeon_gem_handle_lockup(robj->rdev, r);
 	return r;
@@ -414,11 +414,11 @@ int radeon_mode_dumb_mmap(struct drm_file *filp,
 	}
 	robj = gem_to_radeon_bo(gobj);
 	if (radeon_ttm_tt_has_userptr(robj->tbo.ttm)) {
-		drm_gem_object_unreference_unlocked(gobj);
+		drm_gem_object_put_unlocked(gobj);
 		return -EPERM;
 	}
 	*offset_p = radeon_bo_mmap_offset(robj);
-	drm_gem_object_unreference_unlocked(gobj);
+	drm_gem_object_put_unlocked(gobj);
 	return 0;
 }
 
@@ -453,7 +453,7 @@ int radeon_gem_busy_ioctl(struct drm_device *dev, void *data,
 
 	cur_placement = ACCESS_ONCE(robj->tbo.mem.mem_type);
 	args->domain = radeon_mem_type_to_domain(cur_placement);
-	drm_gem_object_unreference_unlocked(gobj);
+	drm_gem_object_put_unlocked(gobj);
 	return r;
 }
 
@@ -485,7 +485,7 @@ int radeon_gem_wait_idle_ioctl(struct drm_device *dev, void *data,
 	if (rdev->asic->mmio_hdp_flush &&
 	    radeon_mem_type_to_domain(cur_placement) == RADEON_GEM_DOMAIN_VRAM)
 		robj->rdev->asic->mmio_hdp_flush(rdev);
-	drm_gem_object_unreference_unlocked(gobj);
+	drm_gem_object_put_unlocked(gobj);
 	r = radeon_gem_handle_lockup(rdev, r);
 	return r;
 }
@@ -504,7 +504,7 @@ int radeon_gem_set_tiling_ioctl(struct drm_device *dev, void *data,
 		return -ENOENT;
 	robj = gem_to_radeon_bo(gobj);
 	r = radeon_bo_set_tiling_flags(robj, args->tiling_flags, args->pitch);
-	drm_gem_object_unreference_unlocked(gobj);
+	drm_gem_object_put_unlocked(gobj);
 	return r;
 }
 
@@ -527,7 +527,7 @@ int radeon_gem_get_tiling_ioctl(struct drm_device *dev, void *data,
 	radeon_bo_get_tiling_flags(rbo, &args->tiling_flags, &args->pitch);
 	radeon_bo_unreserve(rbo);
 out:
-	drm_gem_object_unreference_unlocked(gobj);
+	drm_gem_object_put_unlocked(gobj);
 	return r;
 }
 
@@ -661,14 +661,14 @@ int radeon_gem_va_ioctl(struct drm_device *dev, void *data,
 	r = radeon_bo_reserve(rbo, false);
 	if (r) {
 		args->operation = RADEON_VA_RESULT_ERROR;
-		drm_gem_object_unreference_unlocked(gobj);
+		drm_gem_object_put_unlocked(gobj);
 		return r;
 	}
 	bo_va = radeon_vm_bo_find(&fpriv->vm, rbo);
 	if (!bo_va) {
 		args->operation = RADEON_VA_RESULT_ERROR;
 		radeon_bo_unreserve(rbo);
-		drm_gem_object_unreference_unlocked(gobj);
+		drm_gem_object_put_unlocked(gobj);
 		return -ENOENT;
 	}
 
@@ -695,7 +695,7 @@ int radeon_gem_va_ioctl(struct drm_device *dev, void *data,
 		args->operation = RADEON_VA_RESULT_ERROR;
 	}
 out:
-	drm_gem_object_unreference_unlocked(gobj);
+	drm_gem_object_put_unlocked(gobj);
 	return r;
 }
 
@@ -736,7 +736,7 @@ int radeon_gem_op_ioctl(struct drm_device *dev, void *data,
 
 	radeon_bo_unreserve(robj);
 out:
-	drm_gem_object_unreference_unlocked(gobj);
+	drm_gem_object_put_unlocked(gobj);
 	return r;
 }
 
@@ -762,7 +762,7 @@ int radeon_mode_dumb_create(struct drm_file *file_priv,
 
 	r = drm_gem_handle_create(file_priv, gobj, &handle);
 	/* drop reference from allocate - handle holds it now */
-	drm_gem_object_unreference_unlocked(gobj);
+	drm_gem_object_put_unlocked(gobj);
 	if (r) {
 		return r;
 	}
diff --git a/drivers/gpu/drm/radeon/radeon_kfd.c b/drivers/gpu/drm/radeon/radeon_kfd.c
index a2ab6dcdf4a2..f6578c96925c 100644
--- a/drivers/gpu/drm/radeon/radeon_kfd.c
+++ b/drivers/gpu/drm/radeon/radeon_kfd.c
@@ -75,12 +75,14 @@ static int kgd_init_pipeline(struct kgd_dev *kgd, uint32_t pipe_id,
 				uint32_t hpd_size, uint64_t hpd_gpu_addr);
 static int kgd_init_interrupts(struct kgd_dev *kgd, uint32_t pipe_id);
 static int kgd_hqd_load(struct kgd_dev *kgd, void *mqd, uint32_t pipe_id,
-			uint32_t queue_id, uint32_t __user *wptr);
+			uint32_t queue_id, uint32_t __user *wptr,
+			uint32_t wptr_shift, uint32_t wptr_mask,
+			struct mm_struct *mm);
 static int kgd_hqd_sdma_load(struct kgd_dev *kgd, void *mqd);
 static bool kgd_hqd_is_occupied(struct kgd_dev *kgd, uint64_t queue_address,
 				uint32_t pipe_id, uint32_t queue_id);
 
-static int kgd_hqd_destroy(struct kgd_dev *kgd, uint32_t reset_type,
+static int kgd_hqd_destroy(struct kgd_dev *kgd, void *mqd, uint32_t reset_type,
 				unsigned int timeout, uint32_t pipe_id,
 				uint32_t queue_id);
 static bool kgd_hqd_sdma_is_occupied(struct kgd_dev *kgd, void *mqd);
@@ -482,7 +484,9 @@ static inline struct cik_sdma_rlc_registers *get_sdma_mqd(void *mqd)
 }
 
 static int kgd_hqd_load(struct kgd_dev *kgd, void *mqd, uint32_t pipe_id,
-			uint32_t queue_id, uint32_t __user *wptr)
+			uint32_t queue_id, uint32_t __user *wptr,
+			uint32_t wptr_shift, uint32_t wptr_mask,
+			struct mm_struct *mm)
 {
 	uint32_t wptr_shadow, is_wptr_shadow_valid;
 	struct cik_mqd *m;
@@ -636,7 +640,7 @@ static bool kgd_hqd_sdma_is_occupied(struct kgd_dev *kgd, void *mqd)
 	return false;
 }
 
-static int kgd_hqd_destroy(struct kgd_dev *kgd, uint32_t reset_type,
+static int kgd_hqd_destroy(struct kgd_dev *kgd, void *mqd, uint32_t reset_type,
 				unsigned int timeout, uint32_t pipe_id,
 				uint32_t queue_id)
 {
@@ -785,7 +789,8 @@ static uint32_t kgd_address_watch_get_offset(struct kgd_dev *kgd,
 					unsigned int watch_point_id,
 					unsigned int reg_offset)
 {
-	return watchRegs[watch_point_id * ADDRESS_WATCH_REG_MAX + reg_offset];
+	return watchRegs[watch_point_id * ADDRESS_WATCH_REG_MAX + reg_offset]
+		/ 4;
 }
 
 static bool get_atc_vmid_pasid_mapping_valid(struct kgd_dev *kgd, uint8_t vmid)
diff --git a/drivers/gpu/drm/radeon/radeon_object.c b/drivers/gpu/drm/radeon/radeon_object.c
index 8b722297a05c..093594976126 100644
--- a/drivers/gpu/drm/radeon/radeon_object.c
+++ b/drivers/gpu/drm/radeon/radeon_object.c
@@ -445,7 +445,7 @@ void radeon_bo_force_delete(struct radeon_device *rdev)
 		list_del_init(&bo->list);
 		mutex_unlock(&bo->rdev->gem.mutex);
 		/* this should unref the ttm bo */
-		drm_gem_object_unreference_unlocked(&bo->gem_base);
+		drm_gem_object_put_unlocked(&bo->gem_base);
 	}
 }
 
@@ -546,7 +546,7 @@ int radeon_bo_list_validate(struct radeon_device *rdev,
 	list_for_each_entry(lobj, head, tv.head) {
 		struct radeon_bo *bo = lobj->robj;
 		if (!bo->pin_count) {
-			u32 domain = lobj->prefered_domains;
+			u32 domain = lobj->preferred_domains;
 			u32 allowed = lobj->allowed_domains;
 			u32 current_domain =
 				radeon_mem_type_to_domain(bo->tbo.mem.mem_type);
diff --git a/drivers/gpu/drm/radeon/radeon_ttm.c b/drivers/gpu/drm/radeon/radeon_ttm.c
index 2804b4a15896..bf69bf9086bf 100644
--- a/drivers/gpu/drm/radeon/radeon_ttm.c
+++ b/drivers/gpu/drm/radeon/radeon_ttm.c
@@ -907,17 +907,17 @@ int radeon_ttm_init(struct radeon_device *rdev)
 
 	r = radeon_bo_create(rdev, 256 * 1024, PAGE_SIZE, true,
 			     RADEON_GEM_DOMAIN_VRAM, 0, NULL,
-			     NULL, &rdev->stollen_vga_memory);
+			     NULL, &rdev->stolen_vga_memory);
 	if (r) {
 		return r;
 	}
-	r = radeon_bo_reserve(rdev->stollen_vga_memory, false);
+	r = radeon_bo_reserve(rdev->stolen_vga_memory, false);
 	if (r)
 		return r;
-	r = radeon_bo_pin(rdev->stollen_vga_memory, RADEON_GEM_DOMAIN_VRAM, NULL);
-	radeon_bo_unreserve(rdev->stollen_vga_memory);
+	r = radeon_bo_pin(rdev->stolen_vga_memory, RADEON_GEM_DOMAIN_VRAM, NULL);
+	radeon_bo_unreserve(rdev->stolen_vga_memory);
 	if (r) {
-		radeon_bo_unref(&rdev->stollen_vga_memory);
+		radeon_bo_unref(&rdev->stolen_vga_memory);
 		return r;
 	}
 	DRM_INFO("radeon: %uM of VRAM memory ready\n",
@@ -946,13 +946,13 @@ void radeon_ttm_fini(struct radeon_device *rdev)
 	if (!rdev->mman.initialized)
 		return;
 	radeon_ttm_debugfs_fini(rdev);
-	if (rdev->stollen_vga_memory) {
-		r = radeon_bo_reserve(rdev->stollen_vga_memory, false);
+	if (rdev->stolen_vga_memory) {
+		r = radeon_bo_reserve(rdev->stolen_vga_memory, false);
 		if (r == 0) {
-			radeon_bo_unpin(rdev->stollen_vga_memory);
-			radeon_bo_unreserve(rdev->stollen_vga_memory);
+			radeon_bo_unpin(rdev->stolen_vga_memory);
+			radeon_bo_unreserve(rdev->stolen_vga_memory);
 		}
-		radeon_bo_unref(&rdev->stollen_vga_memory);
+		radeon_bo_unref(&rdev->stolen_vga_memory);
 	}
 	ttm_bo_clean_mm(&rdev->mman.bdev, TTM_PL_VRAM);
 	ttm_bo_clean_mm(&rdev->mman.bdev, TTM_PL_TT);
@@ -1030,19 +1030,17 @@ int radeon_mmap(struct file *filp, struct vm_area_struct *vma)
 static int radeon_mm_dump_table(struct seq_file *m, void *data)
 {
 	struct drm_info_node *node = (struct drm_info_node *)m->private;
-	unsigned ttm_pl = *(int *)node->info_ent->data;
+	unsigned ttm_pl = *(int*)node->info_ent->data;
 	struct drm_device *dev = node->minor->dev;
 	struct radeon_device *rdev = dev->dev_private;
-	struct drm_mm *mm = (struct drm_mm *)rdev->mman.bdev.man[ttm_pl].priv;
-	struct ttm_bo_global *glob = rdev->mman.bdev.glob;
+	struct ttm_mem_type_manager *man = &rdev->mman.bdev.man[ttm_pl];
 	struct drm_printer p = drm_seq_file_printer(m);
 
-	spin_lock(&glob->lru_lock);
-	drm_mm_print(mm, &p);
-	spin_unlock(&glob->lru_lock);
+	man->func->debug(man, &p);
 	return 0;
 }
 
+
 static int ttm_pl_vram = TTM_PL_VRAM;
 static int ttm_pl_tt = TTM_PL_TT;
 
diff --git a/drivers/gpu/drm/radeon/radeon_vm.c b/drivers/gpu/drm/radeon/radeon_vm.c
index 5f68245579a3..5e82b408d522 100644
--- a/drivers/gpu/drm/radeon/radeon_vm.c
+++ b/drivers/gpu/drm/radeon/radeon_vm.c
@@ -139,7 +139,7 @@ struct radeon_bo_list *radeon_vm_get_bos(struct radeon_device *rdev,
 
 	/* add the vm page table to the list */
 	list[0].robj = vm->page_directory;
-	list[0].prefered_domains = RADEON_GEM_DOMAIN_VRAM;
+	list[0].preferred_domains = RADEON_GEM_DOMAIN_VRAM;
 	list[0].allowed_domains = RADEON_GEM_DOMAIN_VRAM;
 	list[0].tv.bo = &vm->page_directory->tbo;
 	list[0].tv.shared = true;
@@ -151,7 +151,7 @@ struct radeon_bo_list *radeon_vm_get_bos(struct radeon_device *rdev,
 			continue;
 
 		list[idx].robj = vm->page_tables[i].bo;
-		list[idx].prefered_domains = RADEON_GEM_DOMAIN_VRAM;
+		list[idx].preferred_domains = RADEON_GEM_DOMAIN_VRAM;
 		list[idx].allowed_domains = RADEON_GEM_DOMAIN_VRAM;
 		list[idx].tv.bo = &list[idx].robj->tbo;
 		list[idx].tv.shared = true;
diff --git a/drivers/gpu/drm/rockchip/Kconfig b/drivers/gpu/drm/rockchip/Kconfig
index 50c41c0a50ef..dcc539ba85d6 100644
--- a/drivers/gpu/drm/rockchip/Kconfig
+++ b/drivers/gpu/drm/rockchip/Kconfig
@@ -5,6 +5,10 @@ config DRM_ROCKCHIP
 	select DRM_KMS_HELPER
 	select DRM_PANEL
 	select VIDEOMODE_HELPERS
+	select DRM_ANALOGIX_DP if ROCKCHIP_ANALOGIX_DP
+	select DRM_DW_HDMI if ROCKCHIP_DW_HDMI
+	select DRM_MIPI_DSI if ROCKCHIP_DW_MIPI_DSI
+	select SND_SOC_HDMI_CODEC if ROCKCHIP_CDN_DP && SND_SOC
 	help
 	  Choose this option if you have a Rockchip soc chipset.
 	  This driver provides kernel mode setting and buffer
@@ -12,10 +16,10 @@ config DRM_ROCKCHIP
 	  2D or 3D acceleration; acceleration is performed by other
 	  IP found on the SoC.
 
+if DRM_ROCKCHIP
+
 config ROCKCHIP_ANALOGIX_DP
 	bool "Rockchip specific extensions for Analogix DP driver"
-	depends on DRM_ROCKCHIP
-	select DRM_ANALOGIX_DP
 	help
 	  This selects support for Rockchip SoC specific extensions
 	  for the Analogix Core DP driver. If you want to enable DP
@@ -23,9 +27,7 @@ config ROCKCHIP_ANALOGIX_DP
 
 config ROCKCHIP_CDN_DP
         bool "Rockchip cdn DP"
-        depends on DRM_ROCKCHIP
-	depends on EXTCON
-	select SND_SOC_HDMI_CODEC if SND_SOC
+	depends on EXTCON=y || (EXTCON=m && DRM_ROCKCHIP=m)
         help
 	  This selects support for Rockchip SoC specific extensions
 	  for the cdn DP driver. If you want to enable Dp on
@@ -34,8 +36,6 @@ config ROCKCHIP_CDN_DP
 
 config ROCKCHIP_DW_HDMI
         bool "Rockchip specific extensions for Synopsys DW HDMI"
-        depends on DRM_ROCKCHIP
-        select DRM_DW_HDMI
         help
 	  This selects support for Rockchip SoC specific extensions
 	  for the Synopsys DesignWare HDMI driver. If you want to
@@ -44,8 +44,6 @@ config ROCKCHIP_DW_HDMI
 
 config ROCKCHIP_DW_MIPI_DSI
 	bool "Rockchip specific extensions for Synopsys DW MIPI DSI"
-	depends on DRM_ROCKCHIP
-	select DRM_MIPI_DSI
 	help
 	 This selects support for Rockchip SoC specific extensions
 	 for the Synopsys DesignWare HDMI driver. If you want to
@@ -54,8 +52,9 @@ config ROCKCHIP_DW_MIPI_DSI
 
 config ROCKCHIP_INNO_HDMI
 	bool "Rockchip specific extensions for Innosilicon HDMI"
-	depends on DRM_ROCKCHIP
 	help
 	  This selects support for Rockchip SoC specific extensions
 	  for the Innosilicon HDMI driver. If you want to enable
 	  HDMI on RK3036 based SoC, you should select this option.
+
+endif
diff --git a/drivers/gpu/drm/rockchip/inno_hdmi.c b/drivers/gpu/drm/rockchip/inno_hdmi.c
index bd87768dd549..7a251a54e792 100644
--- a/drivers/gpu/drm/rockchip/inno_hdmi.c
+++ b/drivers/gpu/drm/rockchip/inno_hdmi.c
@@ -592,7 +592,7 @@ static void inno_hdmi_connector_destroy(struct drm_connector *connector)
 	drm_connector_cleanup(connector);
 }
 
-static struct drm_connector_funcs inno_hdmi_connector_funcs = {
+static const struct drm_connector_funcs inno_hdmi_connector_funcs = {
 	.fill_modes = inno_hdmi_probe_single_connector_modes,
 	.detect = inno_hdmi_connector_detect,
 	.destroy = inno_hdmi_connector_destroy,
diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_fb.c b/drivers/gpu/drm/rockchip/rockchip_drm_fb.c
index 8a0f75612d4b..70773041785b 100644
--- a/drivers/gpu/drm/rockchip/rockchip_drm_fb.c
+++ b/drivers/gpu/drm/rockchip/rockchip_drm_fb.c
@@ -48,7 +48,7 @@ static void rockchip_drm_fb_destroy(struct drm_framebuffer *fb)
 	int i;
 
 	for (i = 0; i < ROCKCHIP_MAX_FB_BUFFER; i++)
-		drm_gem_object_unreference_unlocked(rockchip_fb->obj[i]);
+		drm_gem_object_put_unlocked(rockchip_fb->obj[i]);
 
 	drm_framebuffer_cleanup(fb);
 	kfree(rockchip_fb);
@@ -144,7 +144,7 @@ rockchip_user_fb_create(struct drm_device *dev, struct drm_file *file_priv,
 			width * drm_format_plane_cpp(mode_cmd->pixel_format, i);
 
 		if (obj->size < min_size) {
-			drm_gem_object_unreference_unlocked(obj);
+			drm_gem_object_put_unlocked(obj);
 			ret = -EINVAL;
 			goto err_gem_object_unreference;
 		}
@@ -161,7 +161,7 @@ rockchip_user_fb_create(struct drm_device *dev, struct drm_file *file_priv,
 
 err_gem_object_unreference:
 	for (i--; i >= 0; i--)
-		drm_gem_object_unreference_unlocked(objs[i]);
+		drm_gem_object_put_unlocked(objs[i]);
 	return ERR_PTR(ret);
 }
 
diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_fbdev.c b/drivers/gpu/drm/rockchip/rockchip_drm_fbdev.c
index ce946b9c57a9..724579ebf947 100644
--- a/drivers/gpu/drm/rockchip/rockchip_drm_fbdev.c
+++ b/drivers/gpu/drm/rockchip/rockchip_drm_fbdev.c
@@ -173,7 +173,7 @@ void rockchip_drm_fbdev_fini(struct drm_device *dev)
 	drm_fb_helper_unregister_fbi(helper);
 
 	if (helper->fb)
-		drm_framebuffer_unreference(helper->fb);
+		drm_framebuffer_put(helper->fb);
 
 	drm_fb_helper_fini(helper);
 }
diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_gem.c b/drivers/gpu/drm/rockchip/rockchip_drm_gem.c
index f74333efe4bb..1869c8bb76c8 100644
--- a/drivers/gpu/drm/rockchip/rockchip_drm_gem.c
+++ b/drivers/gpu/drm/rockchip/rockchip_drm_gem.c
@@ -383,7 +383,7 @@ rockchip_gem_create_with_handle(struct drm_file *file_priv,
 		goto err_handle_create;
 
 	/* drop reference from allocate - handle holds it now. */
-	drm_gem_object_unreference_unlocked(obj);
+	drm_gem_object_put_unlocked(obj);
 
 	return rk_obj;
 
diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_vop.c b/drivers/gpu/drm/rockchip/rockchip_drm_vop.c
index 948719dddc36..bf9ed0e63973 100644
--- a/drivers/gpu/drm/rockchip/rockchip_drm_vop.c
+++ b/drivers/gpu/drm/rockchip/rockchip_drm_vop.c
@@ -1026,7 +1026,7 @@ static void vop_crtc_atomic_flush(struct drm_crtc *crtc,
 		if (old_plane_state->fb == new_plane_state->fb)
 			continue;
 
-		drm_framebuffer_reference(old_plane_state->fb);
+		drm_framebuffer_get(old_plane_state->fb);
 		drm_flip_work_queue(&vop->fb_unref_work, old_plane_state->fb);
 		set_bit(VOP_PENDING_FB_UNREF, &vop->pending);
 		WARN_ON(drm_crtc_vblank_get(crtc) != 0);
@@ -1150,7 +1150,7 @@ static void vop_fb_unref_worker(struct drm_flip_work *work, void *val)
 	struct drm_framebuffer *fb = val;
 
 	drm_crtc_vblank_put(&vop->crtc);
-	drm_framebuffer_unreference(fb);
+	drm_framebuffer_put(fb);
 }
 
 static void vop_handle_vblank(struct vop *vop)
diff --git a/drivers/gpu/drm/sun4i/Kconfig b/drivers/gpu/drm/sun4i/Kconfig
index 5bcad8f5fb4f..06f05302ee75 100644
--- a/drivers/gpu/drm/sun4i/Kconfig
+++ b/drivers/gpu/drm/sun4i/Kconfig
@@ -13,17 +13,26 @@ config DRM_SUN4I
 	  Display Engine. If M is selected the module will be called
 	  sun4i-drm.
 
+if DRM_SUN4I
+
 config DRM_SUN4I_HDMI
        tristate "Allwinner A10 HDMI Controller Support"
-       depends on DRM_SUN4I
        default DRM_SUN4I
        help
 	  Choose this option if you have an Allwinner SoC with an HDMI
 	  controller.
 
+config DRM_SUN4I_HDMI_CEC
+       bool "Allwinner A10 HDMI CEC Support"
+       depends on DRM_SUN4I_HDMI
+       select CEC_CORE
+       depends on CEC_PIN
+       help
+	  Choose this option if you have an Allwinner SoC with an HDMI
+	  controller and want to use CEC.
+
 config DRM_SUN4I_BACKEND
 	tristate "Support for Allwinner A10 Display Engine Backend"
-	depends on DRM_SUN4I
 	default DRM_SUN4I
 	help
 	  Choose this option if you have an Allwinner SoC with the
@@ -33,10 +42,11 @@ config DRM_SUN4I_BACKEND
 
 config DRM_SUN8I_MIXER
 	tristate "Support for Allwinner Display Engine 2.0 Mixer"
-	depends on DRM_SUN4I
 	default MACH_SUN8I
 	help
 	  Choose this option if you have an Allwinner SoC with the
 	  Allwinner Display Engine 2.0, which has a mixer to do some
 	  graphics mixture and feed graphics to TCON, If M is
 	  selected the module will be called sun8i-mixer.
+
+endif
diff --git a/drivers/gpu/drm/sun4i/Makefile b/drivers/gpu/drm/sun4i/Makefile
index e29fd3a2ba9c..43c753cafc88 100644
--- a/drivers/gpu/drm/sun4i/Makefile
+++ b/drivers/gpu/drm/sun4i/Makefile
@@ -2,6 +2,7 @@ sun4i-drm-y += sun4i_drv.o
 sun4i-drm-y += sun4i_framebuffer.o
 
 sun4i-drm-hdmi-y += sun4i_hdmi_enc.o
+sun4i-drm-hdmi-y += sun4i_hdmi_i2c.o
 sun4i-drm-hdmi-y += sun4i_hdmi_ddc_clk.o
 sun4i-drm-hdmi-y += sun4i_hdmi_tmds_clk.o
 
diff --git a/drivers/gpu/drm/sun4i/sun4i_backend.c b/drivers/gpu/drm/sun4i/sun4i_backend.c
index cf480218daa5..ec5943627aa5 100644
--- a/drivers/gpu/drm/sun4i/sun4i_backend.c
+++ b/drivers/gpu/drm/sun4i/sun4i_backend.c
@@ -312,7 +312,7 @@ static int sun4i_backend_of_get_id(struct device_node *node)
 		struct device_node *remote;
 		u32 reg;
 
-		remote = of_parse_phandle(ep, "remote-endpoint", 0);
+		remote = of_graph_get_remote_endpoint(ep);
 		if (!remote)
 			continue;
 
diff --git a/drivers/gpu/drm/sun4i/sun4i_hdmi.h b/drivers/gpu/drm/sun4i/sun4i_hdmi.h
index 2f2f2ff1ea63..1457750988da 100644
--- a/drivers/gpu/drm/sun4i/sun4i_hdmi.h
+++ b/drivers/gpu/drm/sun4i/sun4i_hdmi.h
@@ -15,6 +15,8 @@
 #include <drm/drm_connector.h>
 #include <drm/drm_encoder.h>
 
+#include <media/cec.h>
+
 #define SUN4I_HDMI_CTRL_REG		0x004
 #define SUN4I_HDMI_CTRL_ENABLE			BIT(31)
 
@@ -86,6 +88,11 @@
 #define SUN4I_HDMI_PLL_DBG0_TMDS_PARENT_MASK	BIT(21)
 #define SUN4I_HDMI_PLL_DBG0_TMDS_PARENT_SHIFT	21
 
+#define SUN4I_HDMI_CEC			0x214
+#define SUN4I_HDMI_CEC_ENABLE			BIT(11)
+#define SUN4I_HDMI_CEC_TX			BIT(9)
+#define SUN4I_HDMI_CEC_RX			BIT(8)
+
 #define SUN4I_HDMI_PKT_CTRL_REG(n)	(0x2f0 + (4 * (n)))
 #define SUN4I_HDMI_PKT_CTRL_TYPE(n, t)		((t) << (((n) % 4) * 4))
 
@@ -96,6 +103,7 @@
 #define SUN4I_HDMI_DDC_CTRL_ENABLE		BIT(31)
 #define SUN4I_HDMI_DDC_CTRL_START_CMD		BIT(30)
 #define SUN4I_HDMI_DDC_CTRL_FIFO_DIR_MASK	BIT(8)
+#define SUN4I_HDMI_DDC_CTRL_FIFO_DIR_WRITE	(1 << 8)
 #define SUN4I_HDMI_DDC_CTRL_FIFO_DIR_READ	(0 << 8)
 #define SUN4I_HDMI_DDC_CTRL_RESET		BIT(0)
 
@@ -105,14 +113,34 @@
 #define SUN4I_HDMI_DDC_ADDR_OFFSET(off)		(((off) & 0xff) << 8)
 #define SUN4I_HDMI_DDC_ADDR_SLAVE(addr)		((addr) & 0xff)
 
+#define SUN4I_HDMI_DDC_INT_STATUS_REG		0x50c
+#define SUN4I_HDMI_DDC_INT_STATUS_ILLEGAL_FIFO_OPERATION	BIT(7)
+#define SUN4I_HDMI_DDC_INT_STATUS_DDC_RX_FIFO_UNDERFLOW		BIT(6)
+#define SUN4I_HDMI_DDC_INT_STATUS_DDC_TX_FIFO_OVERFLOW		BIT(5)
+#define SUN4I_HDMI_DDC_INT_STATUS_FIFO_REQUEST			BIT(4)
+#define SUN4I_HDMI_DDC_INT_STATUS_ARBITRATION_ERROR		BIT(3)
+#define SUN4I_HDMI_DDC_INT_STATUS_ACK_ERROR			BIT(2)
+#define SUN4I_HDMI_DDC_INT_STATUS_BUS_ERROR			BIT(1)
+#define SUN4I_HDMI_DDC_INT_STATUS_TRANSFER_COMPLETE		BIT(0)
+
 #define SUN4I_HDMI_DDC_FIFO_CTRL_REG	0x510
 #define SUN4I_HDMI_DDC_FIFO_CTRL_CLEAR		BIT(31)
+#define SUN4I_HDMI_DDC_FIFO_CTRL_RX_THRES(n)	(((n) & 0xf) << 4)
+#define SUN4I_HDMI_DDC_FIFO_CTRL_RX_THRES_MASK	GENMASK(7, 4)
+#define SUN4I_HDMI_DDC_FIFO_CTRL_RX_THRES_MAX	(BIT(4) - 1)
+#define SUN4I_HDMI_DDC_FIFO_CTRL_TX_THRES(n)	((n) & 0xf)
+#define SUN4I_HDMI_DDC_FIFO_CTRL_TX_THRES_MASK	GENMASK(3, 0)
+#define SUN4I_HDMI_DDC_FIFO_CTRL_TX_THRES_MAX	(BIT(4) - 1)
 
 #define SUN4I_HDMI_DDC_FIFO_DATA_REG	0x518
+
 #define SUN4I_HDMI_DDC_BYTE_COUNT_REG	0x51c
+#define SUN4I_HDMI_DDC_BYTE_COUNT_MAX		(BIT(10) - 1)
 
 #define SUN4I_HDMI_DDC_CMD_REG		0x520
 #define SUN4I_HDMI_DDC_CMD_EXPLICIT_EDDC_READ	6
+#define SUN4I_HDMI_DDC_CMD_IMPLICIT_READ	5
+#define SUN4I_HDMI_DDC_CMD_IMPLICIT_WRITE	3
 
 #define SUN4I_HDMI_DDC_CLK_REG		0x528
 #define SUN4I_HDMI_DDC_CLK_M(m)			(((m) & 0x7) << 3)
@@ -146,12 +174,16 @@ struct sun4i_hdmi {
 	struct clk		*ddc_clk;
 	struct clk		*tmds_clk;
 
+	struct i2c_adapter	*i2c;
+
 	struct sun4i_drv	*drv;
 
 	bool			hdmi_monitor;
+	struct cec_adapter	*cec_adap;
 };
 
 int sun4i_ddc_create(struct sun4i_hdmi *hdmi, struct clk *clk);
 int sun4i_tmds_create(struct sun4i_hdmi *hdmi);
+int sun4i_hdmi_i2c_create(struct device *dev, struct sun4i_hdmi *hdmi);
 
 #endif /* _SUN4I_HDMI_H_ */
diff --git a/drivers/gpu/drm/sun4i/sun4i_hdmi_enc.c b/drivers/gpu/drm/sun4i/sun4i_hdmi_enc.c
index f5d0d6bd1084..9ea6cd5a1370 100644
--- a/drivers/gpu/drm/sun4i/sun4i_hdmi_enc.c
+++ b/drivers/gpu/drm/sun4i/sun4i_hdmi_enc.c
@@ -29,8 +29,6 @@
 #include "sun4i_hdmi.h"
 #include "sun4i_tcon.h"
 
-#define DDC_SEGMENT_ADDR	0x30
-
 static inline struct sun4i_hdmi *
 drm_encoder_to_sun4i_hdmi(struct drm_encoder *encoder)
 {
@@ -184,93 +182,13 @@ static const struct drm_encoder_funcs sun4i_hdmi_funcs = {
 	.destroy	= drm_encoder_cleanup,
 };
 
-static int sun4i_hdmi_read_sub_block(struct sun4i_hdmi *hdmi,
-				     unsigned int blk, unsigned int offset,
-				     u8 *buf, unsigned int count)
-{
-	unsigned long reg;
-	int i;
-
-	reg = readl(hdmi->base + SUN4I_HDMI_DDC_CTRL_REG);
-	reg &= ~SUN4I_HDMI_DDC_CTRL_FIFO_DIR_MASK;
-	writel(reg | SUN4I_HDMI_DDC_CTRL_FIFO_DIR_READ,
-	       hdmi->base + SUN4I_HDMI_DDC_CTRL_REG);
-
-	writel(SUN4I_HDMI_DDC_ADDR_SEGMENT(offset >> 8) |
-	       SUN4I_HDMI_DDC_ADDR_EDDC(DDC_SEGMENT_ADDR << 1) |
-	       SUN4I_HDMI_DDC_ADDR_OFFSET(offset) |
-	       SUN4I_HDMI_DDC_ADDR_SLAVE(DDC_ADDR),
-	       hdmi->base + SUN4I_HDMI_DDC_ADDR_REG);
-
-	reg = readl(hdmi->base + SUN4I_HDMI_DDC_FIFO_CTRL_REG);
-	writel(reg | SUN4I_HDMI_DDC_FIFO_CTRL_CLEAR,
-	       hdmi->base + SUN4I_HDMI_DDC_FIFO_CTRL_REG);
-
-	writel(count, hdmi->base + SUN4I_HDMI_DDC_BYTE_COUNT_REG);
-	writel(SUN4I_HDMI_DDC_CMD_EXPLICIT_EDDC_READ,
-	       hdmi->base + SUN4I_HDMI_DDC_CMD_REG);
-
-	reg = readl(hdmi->base + SUN4I_HDMI_DDC_CTRL_REG);
-	writel(reg | SUN4I_HDMI_DDC_CTRL_START_CMD,
-	       hdmi->base + SUN4I_HDMI_DDC_CTRL_REG);
-
-	if (readl_poll_timeout(hdmi->base + SUN4I_HDMI_DDC_CTRL_REG, reg,
-			       !(reg & SUN4I_HDMI_DDC_CTRL_START_CMD),
-			       100, 100000))
-		return -EIO;
-
-	for (i = 0; i < count; i++)
-		buf[i] = readb(hdmi->base + SUN4I_HDMI_DDC_FIFO_DATA_REG);
-
-	return 0;
-}
-
-static int sun4i_hdmi_read_edid_block(void *data, u8 *buf, unsigned int blk,
-				      size_t length)
-{
-	struct sun4i_hdmi *hdmi = data;
-	int retry = 2, i;
-
-	do {
-		for (i = 0; i < length; i += SUN4I_HDMI_DDC_FIFO_SIZE) {
-			unsigned char offset = blk * EDID_LENGTH + i;
-			unsigned int count = min((unsigned int)SUN4I_HDMI_DDC_FIFO_SIZE,
-						 length - i);
-			int ret;
-
-			ret = sun4i_hdmi_read_sub_block(hdmi, blk, offset,
-							buf + i, count);
-			if (ret)
-				return ret;
-		}
-	} while (!drm_edid_block_valid(buf, blk, true, NULL) && (retry--));
-
-	return 0;
-}
-
 static int sun4i_hdmi_get_modes(struct drm_connector *connector)
 {
 	struct sun4i_hdmi *hdmi = drm_connector_to_sun4i_hdmi(connector);
-	unsigned long reg;
 	struct edid *edid;
 	int ret;
 
-	/* Reset i2c controller */
-	writel(SUN4I_HDMI_DDC_CTRL_ENABLE | SUN4I_HDMI_DDC_CTRL_RESET,
-	       hdmi->base + SUN4I_HDMI_DDC_CTRL_REG);
-	if (readl_poll_timeout(hdmi->base + SUN4I_HDMI_DDC_CTRL_REG, reg,
-			       !(reg & SUN4I_HDMI_DDC_CTRL_RESET),
-			       100, 2000))
-		return -EIO;
-
-	writel(SUN4I_HDMI_DDC_LINE_CTRL_SDA_ENABLE |
-	       SUN4I_HDMI_DDC_LINE_CTRL_SCL_ENABLE,
-	       hdmi->base + SUN4I_HDMI_DDC_LINE_CTRL_REG);
-
-	clk_prepare_enable(hdmi->ddc_clk);
-	clk_set_rate(hdmi->ddc_clk, 100000);
-
-	edid = drm_do_get_edid(connector, sun4i_hdmi_read_edid_block, hdmi);
+	edid = drm_get_edid(connector, hdmi->i2c);
 	if (!edid)
 		return 0;
 
@@ -279,11 +197,10 @@ static int sun4i_hdmi_get_modes(struct drm_connector *connector)
 			 hdmi->hdmi_monitor ? "an HDMI" : "a DVI");
 
 	drm_mode_connector_update_edid_property(connector, edid);
+	cec_s_phys_addr_from_edid(hdmi->cec_adap, edid);
 	ret = drm_add_edid_modes(connector, edid);
 	kfree(edid);
 
-	clk_disable_unprepare(hdmi->ddc_clk);
-
 	return ret;
 }
 
@@ -299,8 +216,10 @@ sun4i_hdmi_connector_detect(struct drm_connector *connector, bool force)
 
 	if (readl_poll_timeout(hdmi->base + SUN4I_HDMI_HPD_REG, reg,
 			       reg & SUN4I_HDMI_HPD_HIGH,
-			       0, 500000))
+			       0, 500000)) {
+		cec_phys_addr_invalidate(hdmi->cec_adap);
 		return connector_status_disconnected;
+	}
 
 	return connector_status_connected;
 }
@@ -314,6 +233,40 @@ static const struct drm_connector_funcs sun4i_hdmi_connector_funcs = {
 	.atomic_destroy_state	= drm_atomic_helper_connector_destroy_state,
 };
 
+#ifdef CONFIG_DRM_SUN4I_HDMI_CEC
+static bool sun4i_hdmi_cec_pin_read(struct cec_adapter *adap)
+{
+	struct sun4i_hdmi *hdmi = cec_get_drvdata(adap);
+
+	return readl(hdmi->base + SUN4I_HDMI_CEC) & SUN4I_HDMI_CEC_RX;
+}
+
+static void sun4i_hdmi_cec_pin_low(struct cec_adapter *adap)
+{
+	struct sun4i_hdmi *hdmi = cec_get_drvdata(adap);
+
+	/* Start driving the CEC pin low */
+	writel(SUN4I_HDMI_CEC_ENABLE, hdmi->base + SUN4I_HDMI_CEC);
+}
+
+static void sun4i_hdmi_cec_pin_high(struct cec_adapter *adap)
+{
+	struct sun4i_hdmi *hdmi = cec_get_drvdata(adap);
+
+	/*
+	 * Stop driving the CEC pin, the pull up will take over
+	 * unless another CEC device is driving the pin low.
+	 */
+	writel(0, hdmi->base + SUN4I_HDMI_CEC);
+}
+
+static const struct cec_pin_ops sun4i_hdmi_cec_pin_ops = {
+	.read = sun4i_hdmi_cec_pin_read,
+	.low = sun4i_hdmi_cec_pin_low,
+	.high = sun4i_hdmi_cec_pin_high,
+};
+#endif
+
 static int sun4i_hdmi_bind(struct device *dev, struct device *master,
 			   void *data)
 {
@@ -406,9 +359,9 @@ static int sun4i_hdmi_bind(struct device *dev, struct device *master,
 		SUN4I_HDMI_PLL_CTRL_PLL_EN;
 	writel(reg, hdmi->base + SUN4I_HDMI_PLL_CTRL_REG);
 
-	ret = sun4i_ddc_create(hdmi, hdmi->tmds_clk);
+	ret = sun4i_hdmi_i2c_create(dev, hdmi);
 	if (ret) {
-		dev_err(dev, "Couldn't create the DDC clock\n");
+		dev_err(dev, "Couldn't create the HDMI I2C adapter\n");
 		return ret;
 	}
 
@@ -421,13 +374,26 @@ static int sun4i_hdmi_bind(struct device *dev, struct device *master,
 			       NULL);
 	if (ret) {
 		dev_err(dev, "Couldn't initialise the HDMI encoder\n");
-		return ret;
+		goto err_del_i2c_adapter;
 	}
 
 	hdmi->encoder.possible_crtcs = drm_of_find_possible_crtcs(drm,
 								  dev->of_node);
-	if (!hdmi->encoder.possible_crtcs)
-		return -EPROBE_DEFER;
+	if (!hdmi->encoder.possible_crtcs) {
+		ret = -EPROBE_DEFER;
+		goto err_del_i2c_adapter;
+	}
+
+#ifdef CONFIG_DRM_SUN4I_HDMI_CEC
+	hdmi->cec_adap = cec_pin_allocate_adapter(&sun4i_hdmi_cec_pin_ops,
+		hdmi, "sun4i", CEC_CAP_TRANSMIT | CEC_CAP_LOG_ADDRS |
+		CEC_CAP_PASSTHROUGH | CEC_CAP_RC);
+	ret = PTR_ERR_OR_ZERO(hdmi->cec_adap);
+	if (ret < 0)
+		goto err_cleanup_connector;
+	writel(readl(hdmi->base + SUN4I_HDMI_CEC) & ~SUN4I_HDMI_CEC_TX,
+	       hdmi->base + SUN4I_HDMI_CEC);
+#endif
 
 	drm_connector_helper_add(&hdmi->connector,
 				 &sun4i_hdmi_connector_helper_funcs);
@@ -444,12 +410,18 @@ static int sun4i_hdmi_bind(struct device *dev, struct device *master,
 	hdmi->connector.polled = DRM_CONNECTOR_POLL_CONNECT |
 		DRM_CONNECTOR_POLL_DISCONNECT;
 
+	ret = cec_register_adapter(hdmi->cec_adap, dev);
+	if (ret < 0)
+		goto err_cleanup_connector;
 	drm_mode_connector_attach_encoder(&hdmi->connector, &hdmi->encoder);
 
 	return 0;
 
 err_cleanup_connector:
+	cec_delete_adapter(hdmi->cec_adap);
 	drm_encoder_cleanup(&hdmi->encoder);
+err_del_i2c_adapter:
+	i2c_del_adapter(hdmi->i2c);
 	return ret;
 }
 
@@ -458,8 +430,10 @@ static void sun4i_hdmi_unbind(struct device *dev, struct device *master,
 {
 	struct sun4i_hdmi *hdmi = dev_get_drvdata(dev);
 
+	cec_unregister_adapter(hdmi->cec_adap);
 	drm_connector_cleanup(&hdmi->connector);
 	drm_encoder_cleanup(&hdmi->encoder);
+	i2c_del_adapter(hdmi->i2c);
 }
 
 static const struct component_ops sun4i_hdmi_ops = {
diff --git a/drivers/gpu/drm/sun4i/sun4i_hdmi_i2c.c b/drivers/gpu/drm/sun4i/sun4i_hdmi_i2c.c
new file mode 100644
index 000000000000..2e42d09ab42e
--- /dev/null
+++ b/drivers/gpu/drm/sun4i/sun4i_hdmi_i2c.c
@@ -0,0 +1,220 @@
+/*
+ * Copyright (C) 2016 Maxime Ripard <maxime.ripard@free-electrons.com>
+ * Copyright (C) 2017 Jonathan Liu <net147@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; either version 2 of
+ * the License, or (at your option) any later version.
+ */
+
+#include <linux/clk.h>
+#include <linux/i2c.h>
+#include <linux/iopoll.h>
+
+#include "sun4i_hdmi.h"
+
+#define SUN4I_HDMI_DDC_INT_STATUS_ERROR_MASK ( \
+	SUN4I_HDMI_DDC_INT_STATUS_ILLEGAL_FIFO_OPERATION | \
+	SUN4I_HDMI_DDC_INT_STATUS_DDC_RX_FIFO_UNDERFLOW | \
+	SUN4I_HDMI_DDC_INT_STATUS_DDC_TX_FIFO_OVERFLOW | \
+	SUN4I_HDMI_DDC_INT_STATUS_ARBITRATION_ERROR | \
+	SUN4I_HDMI_DDC_INT_STATUS_ACK_ERROR | \
+	SUN4I_HDMI_DDC_INT_STATUS_BUS_ERROR \
+)
+
+/* FIFO request bit is set when FIFO level is above RX_THRESHOLD during read */
+#define RX_THRESHOLD SUN4I_HDMI_DDC_FIFO_CTRL_RX_THRES_MAX
+/* FIFO request bit is set when FIFO level is below TX_THRESHOLD during write */
+#define TX_THRESHOLD 1
+
+static int fifo_transfer(struct sun4i_hdmi *hdmi, u8 *buf, int len, bool read)
+{
+	/*
+	 * 1 byte takes 9 clock cycles (8 bits + 1 ACK) = 90 us for 100 kHz
+	 * clock. As clock rate is fixed, just round it up to 100 us.
+	 */
+	const unsigned long byte_time_ns = 100;
+	const u32 mask = SUN4I_HDMI_DDC_INT_STATUS_ERROR_MASK |
+			 SUN4I_HDMI_DDC_INT_STATUS_FIFO_REQUEST |
+			 SUN4I_HDMI_DDC_INT_STATUS_TRANSFER_COMPLETE;
+	u32 reg;
+
+	/* Limit transfer length by FIFO threshold */
+	len = min_t(int, len, read ? (RX_THRESHOLD + 1) :
+			      (SUN4I_HDMI_DDC_FIFO_SIZE - TX_THRESHOLD + 1));
+
+	/* Wait until error, FIFO request bit set or transfer complete */
+	if (readl_poll_timeout(hdmi->base + SUN4I_HDMI_DDC_INT_STATUS_REG, reg,
+			       reg & mask, len * byte_time_ns, 100000))
+		return -ETIMEDOUT;
+
+	if (reg & SUN4I_HDMI_DDC_INT_STATUS_ERROR_MASK)
+		return -EIO;
+
+	if (read)
+		readsb(hdmi->base + SUN4I_HDMI_DDC_FIFO_DATA_REG, buf, len);
+	else
+		writesb(hdmi->base + SUN4I_HDMI_DDC_FIFO_DATA_REG, buf, len);
+
+	/* Clear FIFO request bit */
+	writel(SUN4I_HDMI_DDC_INT_STATUS_FIFO_REQUEST,
+	       hdmi->base + SUN4I_HDMI_DDC_INT_STATUS_REG);
+
+	return len;
+}
+
+static int xfer_msg(struct sun4i_hdmi *hdmi, struct i2c_msg *msg)
+{
+	int i, len;
+	u32 reg;
+
+	/* Set FIFO direction */
+	reg = readl(hdmi->base + SUN4I_HDMI_DDC_CTRL_REG);
+	reg &= ~SUN4I_HDMI_DDC_CTRL_FIFO_DIR_MASK;
+	reg |= (msg->flags & I2C_M_RD) ?
+	       SUN4I_HDMI_DDC_CTRL_FIFO_DIR_READ :
+	       SUN4I_HDMI_DDC_CTRL_FIFO_DIR_WRITE;
+	writel(reg, hdmi->base + SUN4I_HDMI_DDC_CTRL_REG);
+
+	/* Set I2C address */
+	writel(SUN4I_HDMI_DDC_ADDR_SLAVE(msg->addr),
+	       hdmi->base + SUN4I_HDMI_DDC_ADDR_REG);
+
+	/* Set FIFO RX/TX thresholds and clear FIFO */
+	reg = readl(hdmi->base + SUN4I_HDMI_DDC_FIFO_CTRL_REG);
+	reg |= SUN4I_HDMI_DDC_FIFO_CTRL_CLEAR;
+	reg &= ~SUN4I_HDMI_DDC_FIFO_CTRL_RX_THRES_MASK;
+	reg |= SUN4I_HDMI_DDC_FIFO_CTRL_RX_THRES(RX_THRESHOLD);
+	reg &= ~SUN4I_HDMI_DDC_FIFO_CTRL_TX_THRES_MASK;
+	reg |= SUN4I_HDMI_DDC_FIFO_CTRL_TX_THRES(TX_THRESHOLD);
+	writel(reg, hdmi->base + SUN4I_HDMI_DDC_FIFO_CTRL_REG);
+	if (readl_poll_timeout(hdmi->base + SUN4I_HDMI_DDC_FIFO_CTRL_REG,
+			       reg,
+			       !(reg & SUN4I_HDMI_DDC_FIFO_CTRL_CLEAR),
+			       100, 2000))
+		return -EIO;
+
+	/* Set transfer length */
+	writel(msg->len, hdmi->base + SUN4I_HDMI_DDC_BYTE_COUNT_REG);
+
+	/* Set command */
+	writel(msg->flags & I2C_M_RD ?
+	       SUN4I_HDMI_DDC_CMD_IMPLICIT_READ :
+	       SUN4I_HDMI_DDC_CMD_IMPLICIT_WRITE,
+	       hdmi->base + SUN4I_HDMI_DDC_CMD_REG);
+
+	/* Clear interrupt status bits */
+	writel(SUN4I_HDMI_DDC_INT_STATUS_ERROR_MASK |
+	       SUN4I_HDMI_DDC_INT_STATUS_FIFO_REQUEST |
+	       SUN4I_HDMI_DDC_INT_STATUS_TRANSFER_COMPLETE,
+	       hdmi->base + SUN4I_HDMI_DDC_INT_STATUS_REG);
+
+	/* Start command */
+	reg = readl(hdmi->base + SUN4I_HDMI_DDC_CTRL_REG);
+	writel(reg | SUN4I_HDMI_DDC_CTRL_START_CMD,
+	       hdmi->base + SUN4I_HDMI_DDC_CTRL_REG);
+
+	/* Transfer bytes */
+	for (i = 0; i < msg->len; i += len) {
+		len = fifo_transfer(hdmi, msg->buf + i, msg->len - i,
+				    msg->flags & I2C_M_RD);
+		if (len <= 0)
+			return len;
+	}
+
+	/* Wait for command to finish */
+	if (readl_poll_timeout(hdmi->base + SUN4I_HDMI_DDC_CTRL_REG,
+			       reg,
+			       !(reg & SUN4I_HDMI_DDC_CTRL_START_CMD),
+			       100, 100000))
+		return -EIO;
+
+	/* Check for errors */
+	reg = readl(hdmi->base + SUN4I_HDMI_DDC_INT_STATUS_REG);
+	if ((reg & SUN4I_HDMI_DDC_INT_STATUS_ERROR_MASK) ||
+	    !(reg & SUN4I_HDMI_DDC_INT_STATUS_TRANSFER_COMPLETE)) {
+		return -EIO;
+	}
+
+	return 0;
+}
+
+static int sun4i_hdmi_i2c_xfer(struct i2c_adapter *adap,
+			       struct i2c_msg *msgs, int num)
+{
+	struct sun4i_hdmi *hdmi = i2c_get_adapdata(adap);
+	u32 reg;
+	int err, i, ret = num;
+
+	for (i = 0; i < num; i++) {
+		if (!msgs[i].len)
+			return -EINVAL;
+		if (msgs[i].len > SUN4I_HDMI_DDC_BYTE_COUNT_MAX)
+			return -EINVAL;
+	}
+
+	/* Reset I2C controller */
+	writel(SUN4I_HDMI_DDC_CTRL_ENABLE | SUN4I_HDMI_DDC_CTRL_RESET,
+	       hdmi->base + SUN4I_HDMI_DDC_CTRL_REG);
+	if (readl_poll_timeout(hdmi->base + SUN4I_HDMI_DDC_CTRL_REG, reg,
+			       !(reg & SUN4I_HDMI_DDC_CTRL_RESET),
+			       100, 2000))
+		return -EIO;
+
+	writel(SUN4I_HDMI_DDC_LINE_CTRL_SDA_ENABLE |
+	       SUN4I_HDMI_DDC_LINE_CTRL_SCL_ENABLE,
+	       hdmi->base + SUN4I_HDMI_DDC_LINE_CTRL_REG);
+
+	clk_prepare_enable(hdmi->ddc_clk);
+	clk_set_rate(hdmi->ddc_clk, 100000);
+
+	for (i = 0; i < num; i++) {
+		err = xfer_msg(hdmi, &msgs[i]);
+		if (err) {
+			ret = err;
+			break;
+		}
+	}
+
+	clk_disable_unprepare(hdmi->ddc_clk);
+	return ret;
+}
+
+static u32 sun4i_hdmi_i2c_func(struct i2c_adapter *adap)
+{
+	return I2C_FUNC_I2C | I2C_FUNC_SMBUS_EMUL;
+}
+
+static const struct i2c_algorithm sun4i_hdmi_i2c_algorithm = {
+	.master_xfer	= sun4i_hdmi_i2c_xfer,
+	.functionality	= sun4i_hdmi_i2c_func,
+};
+
+int sun4i_hdmi_i2c_create(struct device *dev, struct sun4i_hdmi *hdmi)
+{
+	struct i2c_adapter *adap;
+	int ret = 0;
+
+	ret = sun4i_ddc_create(hdmi, hdmi->tmds_clk);
+	if (ret)
+		return ret;
+
+	adap = devm_kzalloc(dev, sizeof(*adap), GFP_KERNEL);
+	if (!adap)
+		return -ENOMEM;
+
+	adap->owner = THIS_MODULE;
+	adap->class = I2C_CLASS_DDC;
+	adap->algo = &sun4i_hdmi_i2c_algorithm;
+	strlcpy(adap->name, "sun4i_hdmi_i2c adapter", sizeof(adap->name));
+	i2c_set_adapdata(adap, hdmi);
+
+	ret = i2c_add_adapter(adap);
+	if (ret)
+		return ret;
+
+	hdmi->i2c = adap;
+
+	return ret;
+}
diff --git a/drivers/gpu/drm/sun4i/sun4i_layer.c b/drivers/gpu/drm/sun4i/sun4i_layer.c
index d45f3a1a0a29..7bddf12548d3 100644
--- a/drivers/gpu/drm/sun4i/sun4i_layer.c
+++ b/drivers/gpu/drm/sun4i/sun4i_layer.c
@@ -25,12 +25,6 @@ struct sun4i_plane_desc {
 	       uint32_t                nformats;
 };
 
-static int sun4i_backend_layer_atomic_check(struct drm_plane *plane,
-					    struct drm_plane_state *state)
-{
-	return 0;
-}
-
 static void sun4i_backend_layer_atomic_disable(struct drm_plane *plane,
 					       struct drm_plane_state *old_state)
 {
@@ -52,8 +46,7 @@ static void sun4i_backend_layer_atomic_update(struct drm_plane *plane,
 	sun4i_backend_layer_enable(backend, layer->id, true);
 }
 
-static struct drm_plane_helper_funcs sun4i_backend_layer_helper_funcs = {
-	.atomic_check	= sun4i_backend_layer_atomic_check,
+static const struct drm_plane_helper_funcs sun4i_backend_layer_helper_funcs = {
 	.atomic_disable	= sun4i_backend_layer_atomic_disable,
 	.atomic_update	= sun4i_backend_layer_atomic_update,
 };
diff --git a/drivers/gpu/drm/sun4i/sun4i_rgb.c b/drivers/gpu/drm/sun4i/sun4i_rgb.c
index 550bb262943f..7cd7090ad63a 100644
--- a/drivers/gpu/drm/sun4i/sun4i_rgb.c
+++ b/drivers/gpu/drm/sun4i/sun4i_rgb.c
@@ -119,7 +119,7 @@ sun4i_rgb_connector_destroy(struct drm_connector *connector)
 	drm_connector_cleanup(connector);
 }
 
-static struct drm_connector_funcs sun4i_rgb_con_funcs = {
+static const struct drm_connector_funcs sun4i_rgb_con_funcs = {
 	.fill_modes		= drm_helper_probe_single_connector_modes,
 	.destroy		= sun4i_rgb_connector_destroy,
 	.reset			= drm_atomic_helper_connector_reset,
@@ -127,13 +127,6 @@ static struct drm_connector_funcs sun4i_rgb_con_funcs = {
 	.atomic_destroy_state	= drm_atomic_helper_connector_destroy_state,
 };
 
-static int sun4i_rgb_atomic_check(struct drm_encoder *encoder,
-				  struct drm_crtc_state *crtc_state,
-				  struct drm_connector_state *conn_state)
-{
-	return 0;
-}
-
 static void sun4i_rgb_encoder_enable(struct drm_encoder *encoder)
 {
 	struct sun4i_rgb *rgb = drm_encoder_to_sun4i_rgb(encoder);
@@ -181,7 +174,6 @@ static void sun4i_rgb_encoder_mode_set(struct drm_encoder *encoder,
 }
 
 static struct drm_encoder_helper_funcs sun4i_rgb_enc_helper_funcs = {
-	.atomic_check	= sun4i_rgb_atomic_check,
 	.mode_set	= sun4i_rgb_encoder_mode_set,
 	.disable	= sun4i_rgb_encoder_disable,
 	.enable		= sun4i_rgb_encoder_enable,
diff --git a/drivers/gpu/drm/sun4i/sun4i_tcon.h b/drivers/gpu/drm/sun4i/sun4i_tcon.h
index e3c50ecdcd04..552c88ec16be 100644
--- a/drivers/gpu/drm/sun4i/sun4i_tcon.h
+++ b/drivers/gpu/drm/sun4i/sun4i_tcon.h
@@ -194,8 +194,6 @@ void sun4i_tcon_channel_enable(struct sun4i_tcon *tcon, int channel);
 void sun4i_tcon_enable_vblank(struct sun4i_tcon *tcon, bool enable);
 
 /* Mode Related Controls */
-void sun4i_tcon_switch_interlace(struct sun4i_tcon *tcon,
-				 bool enable);
 void sun4i_tcon_set_mux(struct sun4i_tcon *tcon, int channel,
 			struct drm_encoder *encoder);
 void sun4i_tcon0_mode_set(struct sun4i_tcon *tcon,
diff --git a/drivers/gpu/drm/sun4i/sun4i_tv.c b/drivers/gpu/drm/sun4i/sun4i_tv.c
index 7b45ac9383ea..050cfd43c7a0 100644
--- a/drivers/gpu/drm/sun4i/sun4i_tv.c
+++ b/drivers/gpu/drm/sun4i/sun4i_tv.c
@@ -341,13 +341,6 @@ static void sun4i_tv_mode_to_drm_mode(const struct tv_mode *tv_mode,
 	mode->vtotal = mode->vsync_end  + tv_mode->vback_porch;
 }
 
-static int sun4i_tv_atomic_check(struct drm_encoder *encoder,
-				 struct drm_crtc_state *crtc_state,
-				 struct drm_connector_state *conn_state)
-{
-	return 0;
-}
-
 static void sun4i_tv_disable(struct drm_encoder *encoder)
 {
 	struct sun4i_tv *tv = drm_encoder_to_sun4i_tv(encoder);
@@ -489,7 +482,6 @@ static void sun4i_tv_mode_set(struct drm_encoder *encoder,
 }
 
 static struct drm_encoder_helper_funcs sun4i_tv_helper_funcs = {
-	.atomic_check	= sun4i_tv_atomic_check,
 	.disable	= sun4i_tv_disable,
 	.enable		= sun4i_tv_enable,
 	.mode_set	= sun4i_tv_mode_set,
@@ -545,7 +537,7 @@ sun4i_tv_comp_connector_destroy(struct drm_connector *connector)
 	drm_connector_cleanup(connector);
 }
 
-static struct drm_connector_funcs sun4i_tv_comp_connector_funcs = {
+static const struct drm_connector_funcs sun4i_tv_comp_connector_funcs = {
 	.fill_modes		= drm_helper_probe_single_connector_modes,
 	.destroy		= sun4i_tv_comp_connector_destroy,
 	.reset			= drm_atomic_helper_connector_reset,
diff --git a/drivers/gpu/drm/tegra/Kconfig b/drivers/gpu/drm/tegra/Kconfig
index 2db29d67193d..dc58ab140151 100644
--- a/drivers/gpu/drm/tegra/Kconfig
+++ b/drivers/gpu/drm/tegra/Kconfig
@@ -3,6 +3,7 @@ config DRM_TEGRA
 	depends on ARCH_TEGRA || (ARM && COMPILE_TEST)
 	depends on COMMON_CLK
 	depends on DRM
+	depends on OF
 	select DRM_KMS_HELPER
 	select DRM_MIPI_DSI
 	select DRM_PANEL
diff --git a/drivers/gpu/drm/tegra/Makefile b/drivers/gpu/drm/tegra/Makefile
index 6af3a9ad6565..8927784396e8 100644
--- a/drivers/gpu/drm/tegra/Makefile
+++ b/drivers/gpu/drm/tegra/Makefile
@@ -17,4 +17,6 @@ tegra-drm-y := \
 	falcon.o \
 	vic.o
 
+tegra-drm-y += trace.o
+
 obj-$(CONFIG_DRM_TEGRA) += tegra-drm.o
diff --git a/drivers/gpu/drm/tegra/dpaux.c b/drivers/gpu/drm/tegra/dpaux.c
index 2fde44c3a1b3..e4da041ba89b 100644
--- a/drivers/gpu/drm/tegra/dpaux.c
+++ b/drivers/gpu/drm/tegra/dpaux.c
@@ -25,6 +25,7 @@
 
 #include "dpaux.h"
 #include "drm.h"
+#include "trace.h"
 
 static DEFINE_MUTEX(dpaux_lock);
 static LIST_HEAD(dpaux_list);
@@ -65,14 +66,19 @@ static inline struct tegra_dpaux *work_to_dpaux(struct work_struct *work)
 }
 
 static inline u32 tegra_dpaux_readl(struct tegra_dpaux *dpaux,
-				    unsigned long offset)
+				    unsigned int offset)
 {
-	return readl(dpaux->regs + (offset << 2));
+	u32 value = readl(dpaux->regs + (offset << 2));
+
+	trace_dpaux_readl(dpaux->dev, offset, value);
+
+	return value;
 }
 
 static inline void tegra_dpaux_writel(struct tegra_dpaux *dpaux,
-				      u32 value, unsigned long offset)
+				      u32 value, unsigned int offset)
 {
+	trace_dpaux_writel(dpaux->dev, offset, value);
 	writel(value, dpaux->regs + (offset << 2));
 }
 
diff --git a/drivers/gpu/drm/tegra/drm.c b/drivers/gpu/drm/tegra/drm.c
index 3ba659a5940d..597d563d636a 100644
--- a/drivers/gpu/drm/tegra/drm.c
+++ b/drivers/gpu/drm/tegra/drm.c
@@ -306,8 +306,6 @@ host1x_bo_lookup(struct drm_file *file, u32 handle)
 	if (!gem)
 		return NULL;
 
-	drm_gem_object_unreference_unlocked(gem);
-
 	bo = to_tegra_bo(gem);
 	return &bo->base;
 }
@@ -396,8 +394,10 @@ int tegra_drm_submit(struct tegra_drm_context *context,
 		(void __user *)(uintptr_t)args->waitchks;
 	struct drm_tegra_syncpt syncpt;
 	struct host1x *host1x = dev_get_drvdata(drm->dev->parent);
+	struct drm_gem_object **refs;
 	struct host1x_syncpt *sp;
 	struct host1x_job *job;
+	unsigned int num_refs;
 	int err;
 
 	/* We don't yet support other than one syncpt_incr struct per submit */
@@ -419,6 +419,21 @@ int tegra_drm_submit(struct tegra_drm_context *context,
 	job->class = context->client->base.class;
 	job->serialize = true;
 
+	/*
+	 * Track referenced BOs so that they can be unreferenced after the
+	 * submission is complete.
+	 */
+	num_refs = num_cmdbufs + num_relocs * 2 + num_waitchks;
+
+	refs = kmalloc_array(num_refs, sizeof(*refs), GFP_KERNEL);
+	if (!refs) {
+		err = -ENOMEM;
+		goto put;
+	}
+
+	/* reuse as an iterator later */
+	num_refs = 0;
+
 	while (num_cmdbufs) {
 		struct drm_tegra_cmdbuf cmdbuf;
 		struct host1x_bo *bo;
@@ -447,6 +462,7 @@ int tegra_drm_submit(struct tegra_drm_context *context,
 
 		offset = (u64)cmdbuf.offset + (u64)cmdbuf.words * sizeof(u32);
 		obj = host1x_to_tegra_bo(bo);
+		refs[num_refs++] = &obj->gem;
 
 		/*
 		 * Gather buffer base address must be 4-bytes aligned,
@@ -476,6 +492,7 @@ int tegra_drm_submit(struct tegra_drm_context *context,
 
 		reloc = &job->relocarray[num_relocs];
 		obj = host1x_to_tegra_bo(reloc->cmdbuf.bo);
+		refs[num_refs++] = &obj->gem;
 
 		/*
 		 * The unaligned cmdbuf offset will cause an unaligned write
@@ -489,6 +506,7 @@ int tegra_drm_submit(struct tegra_drm_context *context,
 		}
 
 		obj = host1x_to_tegra_bo(reloc->target.bo);
+		refs[num_refs++] = &obj->gem;
 
 		if (reloc->target.offset >= obj->gem.size) {
 			err = -EINVAL;
@@ -508,6 +526,7 @@ int tegra_drm_submit(struct tegra_drm_context *context,
 			goto fail;
 
 		obj = host1x_to_tegra_bo(wait->bo);
+		refs[num_refs++] = &obj->gem;
 
 		/*
 		 * The unaligned offset will cause an unaligned write during
@@ -547,17 +566,20 @@ int tegra_drm_submit(struct tegra_drm_context *context,
 		goto fail;
 
 	err = host1x_job_submit(job);
-	if (err)
-		goto fail_submit;
+	if (err) {
+		host1x_job_unpin(job);
+		goto fail;
+	}
 
 	args->fence = job->syncpt_end;
 
-	host1x_job_put(job);
-	return 0;
-
-fail_submit:
-	host1x_job_unpin(job);
 fail:
+	while (num_refs--)
+		drm_gem_object_put_unlocked(refs[num_refs]);
+
+	kfree(refs);
+
+put:
 	host1x_job_put(job);
 	return err;
 }
@@ -593,7 +615,7 @@ static int tegra_gem_mmap(struct drm_device *drm, void *data,
 
 	args->offset = drm_vma_node_offset_addr(&bo->gem.vma_node);
 
-	drm_gem_object_unreference_unlocked(gem);
+	drm_gem_object_put_unlocked(gem);
 
 	return 0;
 }
@@ -860,7 +882,7 @@ static int tegra_gem_set_tiling(struct drm_device *drm, void *data,
 	bo->tiling.mode = mode;
 	bo->tiling.value = value;
 
-	drm_gem_object_unreference_unlocked(gem);
+	drm_gem_object_put_unlocked(gem);
 
 	return 0;
 }
@@ -900,7 +922,7 @@ static int tegra_gem_get_tiling(struct drm_device *drm, void *data,
 		break;
 	}
 
-	drm_gem_object_unreference_unlocked(gem);
+	drm_gem_object_put_unlocked(gem);
 
 	return err;
 }
@@ -925,7 +947,7 @@ static int tegra_gem_set_flags(struct drm_device *drm, void *data,
 	if (args->flags & DRM_TEGRA_GEM_BOTTOM_UP)
 		bo->flags |= TEGRA_BO_BOTTOM_UP;
 
-	drm_gem_object_unreference_unlocked(gem);
+	drm_gem_object_put_unlocked(gem);
 
 	return 0;
 }
@@ -947,7 +969,7 @@ static int tegra_gem_get_flags(struct drm_device *drm, void *data,
 	if (bo->flags & TEGRA_BO_BOTTOM_UP)
 		args->flags |= DRM_TEGRA_GEM_BOTTOM_UP;
 
-	drm_gem_object_unreference_unlocked(gem);
+	drm_gem_object_put_unlocked(gem);
 
 	return 0;
 }
@@ -955,20 +977,34 @@ static int tegra_gem_get_flags(struct drm_device *drm, void *data,
 
 static const struct drm_ioctl_desc tegra_drm_ioctls[] = {
 #ifdef CONFIG_DRM_TEGRA_STAGING
-	DRM_IOCTL_DEF_DRV(TEGRA_GEM_CREATE, tegra_gem_create, 0),
-	DRM_IOCTL_DEF_DRV(TEGRA_GEM_MMAP, tegra_gem_mmap, 0),
-	DRM_IOCTL_DEF_DRV(TEGRA_SYNCPT_READ, tegra_syncpt_read, 0),
-	DRM_IOCTL_DEF_DRV(TEGRA_SYNCPT_INCR, tegra_syncpt_incr, 0),
-	DRM_IOCTL_DEF_DRV(TEGRA_SYNCPT_WAIT, tegra_syncpt_wait, 0),
-	DRM_IOCTL_DEF_DRV(TEGRA_OPEN_CHANNEL, tegra_open_channel, 0),
-	DRM_IOCTL_DEF_DRV(TEGRA_CLOSE_CHANNEL, tegra_close_channel, 0),
-	DRM_IOCTL_DEF_DRV(TEGRA_GET_SYNCPT, tegra_get_syncpt, 0),
-	DRM_IOCTL_DEF_DRV(TEGRA_SUBMIT, tegra_submit, 0),
-	DRM_IOCTL_DEF_DRV(TEGRA_GET_SYNCPT_BASE, tegra_get_syncpt_base, 0),
-	DRM_IOCTL_DEF_DRV(TEGRA_GEM_SET_TILING, tegra_gem_set_tiling, 0),
-	DRM_IOCTL_DEF_DRV(TEGRA_GEM_GET_TILING, tegra_gem_get_tiling, 0),
-	DRM_IOCTL_DEF_DRV(TEGRA_GEM_SET_FLAGS, tegra_gem_set_flags, 0),
-	DRM_IOCTL_DEF_DRV(TEGRA_GEM_GET_FLAGS, tegra_gem_get_flags, 0),
+	DRM_IOCTL_DEF_DRV(TEGRA_GEM_CREATE, tegra_gem_create,
+			  DRM_UNLOCKED | DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(TEGRA_GEM_MMAP, tegra_gem_mmap,
+			  DRM_UNLOCKED | DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(TEGRA_SYNCPT_READ, tegra_syncpt_read,
+			  DRM_UNLOCKED | DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(TEGRA_SYNCPT_INCR, tegra_syncpt_incr,
+			  DRM_UNLOCKED | DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(TEGRA_SYNCPT_WAIT, tegra_syncpt_wait,
+			  DRM_UNLOCKED | DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(TEGRA_OPEN_CHANNEL, tegra_open_channel,
+			  DRM_UNLOCKED | DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(TEGRA_CLOSE_CHANNEL, tegra_close_channel,
+			  DRM_UNLOCKED | DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(TEGRA_GET_SYNCPT, tegra_get_syncpt,
+			  DRM_UNLOCKED | DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(TEGRA_SUBMIT, tegra_submit,
+			  DRM_UNLOCKED | DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(TEGRA_GET_SYNCPT_BASE, tegra_get_syncpt_base,
+			  DRM_UNLOCKED | DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(TEGRA_GEM_SET_TILING, tegra_gem_set_tiling,
+			  DRM_UNLOCKED | DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(TEGRA_GEM_GET_TILING, tegra_gem_get_tiling,
+			  DRM_UNLOCKED | DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(TEGRA_GEM_SET_FLAGS, tegra_gem_set_flags,
+			  DRM_UNLOCKED | DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(TEGRA_GEM_GET_FLAGS, tegra_gem_get_flags,
+			  DRM_UNLOCKED | DRM_RENDER_ALLOW),
 #endif
 };
 
@@ -1035,9 +1071,11 @@ static int tegra_debugfs_iova(struct seq_file *s, void *data)
 	struct tegra_drm *tegra = drm->dev_private;
 	struct drm_printer p = drm_seq_file_printer(s);
 
-	mutex_lock(&tegra->mm_lock);
-	drm_mm_print(&tegra->mm, &p);
-	mutex_unlock(&tegra->mm_lock);
+	if (tegra->domain) {
+		mutex_lock(&tegra->mm_lock);
+		drm_mm_print(&tegra->mm, &p);
+		mutex_unlock(&tegra->mm_lock);
+	}
 
 	return 0;
 }
@@ -1057,7 +1095,7 @@ static int tegra_debugfs_init(struct drm_minor *minor)
 
 static struct drm_driver tegra_drm_driver = {
 	.driver_features = DRIVER_MODESET | DRIVER_GEM | DRIVER_PRIME |
-			   DRIVER_ATOMIC,
+			   DRIVER_ATOMIC | DRIVER_RENDER,
 	.load = tegra_drm_load,
 	.unload = tegra_drm_unload,
 	.open = tegra_drm_open,
@@ -1077,8 +1115,6 @@ static struct drm_driver tegra_drm_driver = {
 	.gem_prime_import = tegra_gem_prime_import,
 
 	.dumb_create = tegra_bo_dumb_create,
-	.dumb_map_offset = tegra_bo_dumb_map_offset,
-	.dumb_destroy = drm_gem_dumb_destroy,
 
 	.ioctls = tegra_drm_ioctls,
 	.num_ioctls = ARRAY_SIZE(tegra_drm_ioctls),
diff --git a/drivers/gpu/drm/tegra/drm.h b/drivers/gpu/drm/tegra/drm.h
index 6d6da01282f3..063f5d397526 100644
--- a/drivers/gpu/drm/tegra/drm.h
+++ b/drivers/gpu/drm/tegra/drm.h
@@ -23,6 +23,7 @@
 #include <drm/drm_fixed.h>
 
 #include "gem.h"
+#include "trace.h"
 
 struct reset_control;
 
@@ -172,14 +173,19 @@ static inline struct tegra_dc *to_tegra_dc(struct drm_crtc *crtc)
 }
 
 static inline void tegra_dc_writel(struct tegra_dc *dc, u32 value,
-				   unsigned long offset)
+				   unsigned int offset)
 {
+	trace_dc_writel(dc->dev, offset, value);
 	writel(value, dc->regs + (offset << 2));
 }
 
-static inline u32 tegra_dc_readl(struct tegra_dc *dc, unsigned long offset)
+static inline u32 tegra_dc_readl(struct tegra_dc *dc, unsigned int offset)
 {
-	return readl(dc->regs + (offset << 2));
+	u32 value = readl(dc->regs + (offset << 2));
+
+	trace_dc_readl(dc->dev, offset, value);
+
+	return value;
 }
 
 struct tegra_dc_window {
diff --git a/drivers/gpu/drm/tegra/dsi.c b/drivers/gpu/drm/tegra/dsi.c
index e4b5aedfdbd4..046649ec9441 100644
--- a/drivers/gpu/drm/tegra/dsi.c
+++ b/drivers/gpu/drm/tegra/dsi.c
@@ -28,6 +28,7 @@
 #include "drm.h"
 #include "dsi.h"
 #include "mipi-phy.h"
+#include "trace.h"
 
 struct tegra_dsi_state {
 	struct drm_connector_state base;
@@ -105,15 +106,20 @@ static struct tegra_dsi_state *tegra_dsi_get_state(struct tegra_dsi *dsi)
 	return to_dsi_state(dsi->output.connector.state);
 }
 
-static inline u32 tegra_dsi_readl(struct tegra_dsi *dsi, unsigned long reg)
+static inline u32 tegra_dsi_readl(struct tegra_dsi *dsi, unsigned int offset)
 {
-	return readl(dsi->regs + (reg << 2));
+	u32 value = readl(dsi->regs + (offset << 2));
+
+	trace_dsi_readl(dsi->dev, offset, value);
+
+	return value;
 }
 
 static inline void tegra_dsi_writel(struct tegra_dsi *dsi, u32 value,
-				    unsigned long reg)
+				    unsigned int offset)
 {
-	writel(value, dsi->regs + (reg << 2));
+	trace_dsi_writel(dsi->dev, offset, value);
+	writel(value, dsi->regs + (offset << 2));
 }
 
 static int tegra_dsi_show_regs(struct seq_file *s, void *data)
diff --git a/drivers/gpu/drm/tegra/fb.c b/drivers/gpu/drm/tegra/fb.c
index 25acb73ee728..80540c1c66dc 100644
--- a/drivers/gpu/drm/tegra/fb.c
+++ b/drivers/gpu/drm/tegra/fb.c
@@ -88,7 +88,7 @@ static void tegra_fb_destroy(struct drm_framebuffer *framebuffer)
 			if (bo->pages)
 				vunmap(bo->vaddr);
 
-			drm_gem_object_unreference_unlocked(&bo->gem);
+			drm_gem_object_put_unlocked(&bo->gem);
 		}
 	}
 
@@ -195,7 +195,7 @@ struct drm_framebuffer *tegra_fb_create(struct drm_device *drm,
 
 unreference:
 	while (i--)
-		drm_gem_object_unreference_unlocked(&planes[i]->gem);
+		drm_gem_object_put_unlocked(&planes[i]->gem);
 
 	return ERR_PTR(err);
 }
@@ -242,7 +242,7 @@ static int tegra_fbdev_probe(struct drm_fb_helper *helper,
 	info = drm_fb_helper_alloc_fbi(helper);
 	if (IS_ERR(info)) {
 		dev_err(drm->dev, "failed to allocate framebuffer info\n");
-		drm_gem_object_unreference_unlocked(&bo->gem);
+		drm_gem_object_put_unlocked(&bo->gem);
 		return PTR_ERR(info);
 	}
 
@@ -251,7 +251,7 @@ static int tegra_fbdev_probe(struct drm_fb_helper *helper,
 		err = PTR_ERR(fbdev->fb);
 		dev_err(drm->dev, "failed to allocate DRM framebuffer: %d\n",
 			err);
-		drm_gem_object_unreference_unlocked(&bo->gem);
+		drm_gem_object_put_unlocked(&bo->gem);
 		return PTR_ERR(fbdev->fb);
 	}
 
diff --git a/drivers/gpu/drm/tegra/gem.c b/drivers/gpu/drm/tegra/gem.c
index 7a39a355678a..ab1e53d434e8 100644
--- a/drivers/gpu/drm/tegra/gem.c
+++ b/drivers/gpu/drm/tegra/gem.c
@@ -24,7 +24,7 @@ static void tegra_bo_put(struct host1x_bo *bo)
 {
 	struct tegra_bo *obj = host1x_to_tegra_bo(bo);
 
-	drm_gem_object_unreference_unlocked(&obj->gem);
+	drm_gem_object_put_unlocked(&obj->gem);
 }
 
 static dma_addr_t tegra_bo_pin(struct host1x_bo *bo, struct sg_table **sgt)
@@ -95,7 +95,7 @@ static struct host1x_bo *tegra_bo_get(struct host1x_bo *bo)
 {
 	struct tegra_bo *obj = host1x_to_tegra_bo(bo);
 
-	drm_gem_object_reference(&obj->gem);
+	drm_gem_object_get(&obj->gem);
 
 	return bo;
 }
@@ -325,7 +325,7 @@ struct tegra_bo *tegra_bo_create_with_handle(struct drm_file *file,
 		return ERR_PTR(err);
 	}
 
-	drm_gem_object_unreference_unlocked(&bo->gem);
+	drm_gem_object_put_unlocked(&bo->gem);
 
 	return bo;
 }
@@ -423,27 +423,6 @@ int tegra_bo_dumb_create(struct drm_file *file, struct drm_device *drm,
 	return 0;
 }
 
-int tegra_bo_dumb_map_offset(struct drm_file *file, struct drm_device *drm,
-			     u32 handle, u64 *offset)
-{
-	struct drm_gem_object *gem;
-	struct tegra_bo *bo;
-
-	gem = drm_gem_object_lookup(file, handle);
-	if (!gem) {
-		dev_err(drm->dev, "failed to lookup GEM object\n");
-		return -EINVAL;
-	}
-
-	bo = to_tegra_bo(gem);
-
-	*offset = drm_vma_node_offset_addr(&bo->gem.vma_node);
-
-	drm_gem_object_unreference_unlocked(gem);
-
-	return 0;
-}
-
 static int tegra_bo_fault(struct vm_fault *vmf)
 {
 	struct vm_area_struct *vma = vmf->vma;
@@ -481,30 +460,28 @@ const struct vm_operations_struct tegra_bo_vm_ops = {
 	.close = drm_gem_vm_close,
 };
 
-int tegra_drm_mmap(struct file *file, struct vm_area_struct *vma)
+static int tegra_gem_mmap(struct drm_gem_object *gem,
+			  struct vm_area_struct *vma)
 {
-	struct drm_gem_object *gem;
-	struct tegra_bo *bo;
-	int ret;
-
-	ret = drm_gem_mmap(file, vma);
-	if (ret)
-		return ret;
-
-	gem = vma->vm_private_data;
-	bo = to_tegra_bo(gem);
+	struct tegra_bo *bo = to_tegra_bo(gem);
 
 	if (!bo->pages) {
 		unsigned long vm_pgoff = vma->vm_pgoff;
+		int err;
 
+		/*
+		 * Clear the VM_PFNMAP flag that was set by drm_gem_mmap(),
+		 * and set the vm_pgoff (used as a fake buffer offset by DRM)
+		 * to 0 as we want to map the whole buffer.
+		 */
 		vma->vm_flags &= ~VM_PFNMAP;
 		vma->vm_pgoff = 0;
 
-		ret = dma_mmap_wc(gem->dev->dev, vma, bo->vaddr, bo->paddr,
+		err = dma_mmap_wc(gem->dev->dev, vma, bo->vaddr, bo->paddr,
 				  gem->size);
-		if (ret) {
+		if (err < 0) {
 			drm_gem_vm_close(vma);
-			return ret;
+			return err;
 		}
 
 		vma->vm_pgoff = vm_pgoff;
@@ -520,6 +497,20 @@ int tegra_drm_mmap(struct file *file, struct vm_area_struct *vma)
 	return 0;
 }
 
+int tegra_drm_mmap(struct file *file, struct vm_area_struct *vma)
+{
+	struct drm_gem_object *gem;
+	int err;
+
+	err = drm_gem_mmap(file, vma);
+	if (err < 0)
+		return err;
+
+	gem = vma->vm_private_data;
+
+	return tegra_gem_mmap(gem, vma);
+}
+
 static struct sg_table *
 tegra_gem_prime_map_dma_buf(struct dma_buf_attachment *attach,
 			    enum dma_data_direction dir)
@@ -603,7 +594,14 @@ static void tegra_gem_prime_kunmap(struct dma_buf *buf, unsigned long page,
 
 static int tegra_gem_prime_mmap(struct dma_buf *buf, struct vm_area_struct *vma)
 {
-	return -EINVAL;
+	struct drm_gem_object *gem = buf->priv;
+	int err;
+
+	err = drm_gem_mmap_obj(gem, gem->size, vma);
+	if (err < 0)
+		return err;
+
+	return tegra_gem_mmap(gem, vma);
 }
 
 static void *tegra_gem_prime_vmap(struct dma_buf *buf)
@@ -654,7 +652,7 @@ struct drm_gem_object *tegra_gem_prime_import(struct drm_device *drm,
 		struct drm_gem_object *gem = buf->priv;
 
 		if (gem->dev == drm) {
-			drm_gem_object_reference(gem);
+			drm_gem_object_get(gem);
 			return gem;
 		}
 	}
diff --git a/drivers/gpu/drm/tegra/gem.h b/drivers/gpu/drm/tegra/gem.h
index 8b32a6fd586d..8eb9fd24ef0e 100644
--- a/drivers/gpu/drm/tegra/gem.h
+++ b/drivers/gpu/drm/tegra/gem.h
@@ -67,8 +67,6 @@ struct tegra_bo *tegra_bo_create_with_handle(struct drm_file *file,
 void tegra_bo_free_object(struct drm_gem_object *gem);
 int tegra_bo_dumb_create(struct drm_file *file, struct drm_device *drm,
 			 struct drm_mode_create_dumb *args);
-int tegra_bo_dumb_map_offset(struct drm_file *file, struct drm_device *drm,
-			     u32 handle, u64 *offset);
 
 int tegra_drm_mmap(struct file *file, struct vm_area_struct *vma);
 
diff --git a/drivers/gpu/drm/tegra/hdmi.c b/drivers/gpu/drm/tegra/hdmi.c
index a621b0da4092..5b9d83b71943 100644
--- a/drivers/gpu/drm/tegra/hdmi.c
+++ b/drivers/gpu/drm/tegra/hdmi.c
@@ -24,6 +24,7 @@
 #include "hdmi.h"
 #include "drm.h"
 #include "dc.h"
+#include "trace.h"
 
 #define HDMI_ELD_BUFFER_SIZE 96
 
@@ -100,14 +101,19 @@ enum {
 };
 
 static inline u32 tegra_hdmi_readl(struct tegra_hdmi *hdmi,
-				   unsigned long offset)
+				   unsigned int offset)
 {
-	return readl(hdmi->regs + (offset << 2));
+	u32 value = readl(hdmi->regs + (offset << 2));
+
+	trace_hdmi_readl(hdmi->dev, offset, value);
+
+	return value;
 }
 
 static inline void tegra_hdmi_writel(struct tegra_hdmi *hdmi, u32 value,
-				     unsigned long offset)
+				     unsigned int offset)
 {
+	trace_hdmi_writel(hdmi->dev, offset, value);
 	writel(value, hdmi->regs + (offset << 2));
 }
 
diff --git a/drivers/gpu/drm/tegra/sor.c b/drivers/gpu/drm/tegra/sor.c
index e0642d05a8d3..7ab1d1dc7cd7 100644
--- a/drivers/gpu/drm/tegra/sor.c
+++ b/drivers/gpu/drm/tegra/sor.c
@@ -26,6 +26,7 @@
 #include "dc.h"
 #include "drm.h"
 #include "sor.h"
+#include "trace.h"
 
 #define SOR_REKEY 0x38
 
@@ -232,14 +233,19 @@ static inline struct tegra_sor *to_sor(struct tegra_output *output)
 	return container_of(output, struct tegra_sor, output);
 }
 
-static inline u32 tegra_sor_readl(struct tegra_sor *sor, unsigned long offset)
+static inline u32 tegra_sor_readl(struct tegra_sor *sor, unsigned int offset)
 {
-	return readl(sor->regs + (offset << 2));
+	u32 value = readl(sor->regs + (offset << 2));
+
+	trace_sor_readl(sor->dev, offset, value);
+
+	return value;
 }
 
 static inline void tegra_sor_writel(struct tegra_sor *sor, u32 value,
-				    unsigned long offset)
+				    unsigned int offset)
 {
+	trace_sor_writel(sor->dev, offset, value);
 	writel(value, sor->regs + (offset << 2));
 }
 
diff --git a/drivers/gpu/drm/tegra/trace.c b/drivers/gpu/drm/tegra/trace.c
new file mode 100644
index 000000000000..006f65c72a34
--- /dev/null
+++ b/drivers/gpu/drm/tegra/trace.c
@@ -0,0 +1,2 @@
+#define CREATE_TRACE_POINTS
+#include "trace.h"
diff --git a/drivers/gpu/drm/tegra/trace.h b/drivers/gpu/drm/tegra/trace.h
new file mode 100644
index 000000000000..e9b7cdad5c4c
--- /dev/null
+++ b/drivers/gpu/drm/tegra/trace.h
@@ -0,0 +1,68 @@
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM tegra
+
+#if !defined(DRM_TEGRA_TRACE_H) || defined(TRACE_HEADER_MULTI_READ)
+#define DRM_TEGRA_TRACE_H 1
+
+#include <linux/device.h>
+#include <linux/tracepoint.h>
+
+DECLARE_EVENT_CLASS(register_access,
+	TP_PROTO(struct device *dev, unsigned int offset, u32 value),
+	TP_ARGS(dev, offset, value),
+	TP_STRUCT__entry(
+		__field(struct device *, dev)
+		__field(unsigned int, offset)
+		__field(u32, value)
+	),
+	TP_fast_assign(
+		__entry->dev = dev;
+		__entry->offset = offset;
+		__entry->value = value;
+	),
+	TP_printk("%s %04x %08x", dev_name(__entry->dev), __entry->offset,
+		  __entry->value)
+);
+
+DEFINE_EVENT(register_access, dc_writel,
+	TP_PROTO(struct device *dev, unsigned int offset, u32 value),
+	TP_ARGS(dev, offset, value));
+DEFINE_EVENT(register_access, dc_readl,
+	TP_PROTO(struct device *dev, unsigned int offset, u32 value),
+	TP_ARGS(dev, offset, value));
+
+DEFINE_EVENT(register_access, hdmi_writel,
+	TP_PROTO(struct device *dev, unsigned int offset, u32 value),
+	TP_ARGS(dev, offset, value));
+DEFINE_EVENT(register_access, hdmi_readl,
+	TP_PROTO(struct device *dev, unsigned int offset, u32 value),
+	TP_ARGS(dev, offset, value));
+
+DEFINE_EVENT(register_access, dsi_writel,
+	TP_PROTO(struct device *dev, unsigned int offset, u32 value),
+	TP_ARGS(dev, offset, value));
+DEFINE_EVENT(register_access, dsi_readl,
+	TP_PROTO(struct device *dev, unsigned int offset, u32 value),
+	TP_ARGS(dev, offset, value));
+
+DEFINE_EVENT(register_access, dpaux_writel,
+	TP_PROTO(struct device *dev, unsigned int offset, u32 value),
+	TP_ARGS(dev, offset, value));
+DEFINE_EVENT(register_access, dpaux_readl,
+	TP_PROTO(struct device *dev, unsigned int offset, u32 value),
+	TP_ARGS(dev, offset, value));
+
+DEFINE_EVENT(register_access, sor_writel,
+	TP_PROTO(struct device *dev, unsigned int offset, u32 value),
+	TP_ARGS(dev, offset, value));
+DEFINE_EVENT(register_access, sor_readl,
+	TP_PROTO(struct device *dev, unsigned int offset, u32 value),
+	TP_ARGS(dev, offset, value));
+
+#endif /* DRM_TEGRA_TRACE_H */
+
+/* This part must be outside protection */
+#undef TRACE_INCLUDE_PATH
+#define TRACE_INCLUDE_PATH .
+#define TRACE_INCLUDE_FILE trace
+#include <trace/define_trace.h>
diff --git a/drivers/gpu/drm/tegra/vic.c b/drivers/gpu/drm/tegra/vic.c
index 47cb1aaa58b1..2448229fa653 100644
--- a/drivers/gpu/drm/tegra/vic.c
+++ b/drivers/gpu/drm/tegra/vic.c
@@ -258,12 +258,16 @@ static const struct tegra_drm_client_ops vic_ops = {
 	.submit = tegra_drm_submit,
 };
 
+#define NVIDIA_TEGRA_124_VIC_FIRMWARE "nvidia/tegra124/vic03_ucode.bin"
+
 static const struct vic_config vic_t124_config = {
-	.firmware = "nvidia/tegra124/vic03_ucode.bin",
+	.firmware = NVIDIA_TEGRA_124_VIC_FIRMWARE,
 };
 
+#define NVIDIA_TEGRA_210_VIC_FIRMWARE "nvidia/tegra210/vic04_ucode.bin"
+
 static const struct vic_config vic_t210_config = {
-	.firmware = "nvidia/tegra210/vic04_ucode.bin",
+	.firmware = NVIDIA_TEGRA_210_VIC_FIRMWARE,
 };
 
 static const struct of_device_id vic_match[] = {
@@ -394,3 +398,10 @@ struct platform_driver tegra_vic_driver = {
 	.probe = vic_probe,
 	.remove = vic_remove,
 };
+
+#if IS_ENABLED(CONFIG_ARCH_TEGRA_124_SOC)
+MODULE_FIRMWARE(NVIDIA_TEGRA_124_VIC_FIRMWARE);
+#endif
+#if IS_ENABLED(CONFIG_ARCH_TEGRA_210_SOC)
+MODULE_FIRMWARE(NVIDIA_TEGRA_210_VIC_FIRMWARE);
+#endif
diff --git a/drivers/gpu/drm/tinydrm/Kconfig b/drivers/gpu/drm/tinydrm/Kconfig
index f17c3caceab2..2e790e7dced5 100644
--- a/drivers/gpu/drm/tinydrm/Kconfig
+++ b/drivers/gpu/drm/tinydrm/Kconfig
@@ -32,3 +32,13 @@ config TINYDRM_REPAPER
 	  2.71" TFT EPD Panel (E2271CS021)
 
 	  If M is selected the module will be called repaper.
+
+config TINYDRM_ST7586
+	tristate "DRM support for Sitronix ST7586 display panels"
+	depends on DRM_TINYDRM && SPI
+	select TINYDRM_MIPI_DBI
+	help
+	  DRM driver for the following Sitronix ST7586 panels:
+	  * LEGO MINDSTORMS EV3
+
+	  If M is selected the module will be called st7586.
diff --git a/drivers/gpu/drm/tinydrm/Makefile b/drivers/gpu/drm/tinydrm/Makefile
index 95bb4d4fa785..0c184bd1bb59 100644
--- a/drivers/gpu/drm/tinydrm/Makefile
+++ b/drivers/gpu/drm/tinydrm/Makefile
@@ -6,3 +6,4 @@ obj-$(CONFIG_TINYDRM_MIPI_DBI)		+= mipi-dbi.o
 # Displays
 obj-$(CONFIG_TINYDRM_MI0283QT)		+= mi0283qt.o
 obj-$(CONFIG_TINYDRM_REPAPER)		+= repaper.o
+obj-$(CONFIG_TINYDRM_ST7586)		+= st7586.o
diff --git a/drivers/gpu/drm/tinydrm/core/tinydrm-helpers.c b/drivers/gpu/drm/tinydrm/core/tinydrm-helpers.c
index 75808bb84c9a..bd6cce093a85 100644
--- a/drivers/gpu/drm/tinydrm/core/tinydrm-helpers.c
+++ b/drivers/gpu/drm/tinydrm/core/tinydrm-helpers.c
@@ -185,7 +185,9 @@ EXPORT_SYMBOL(tinydrm_xrgb8888_to_rgb565);
 /**
  * tinydrm_xrgb8888_to_gray8 - Convert XRGB8888 to grayscale
  * @dst: 8-bit grayscale destination buffer
+ * @vaddr: XRGB8888 source buffer
  * @fb: DRM framebuffer
+ * @clip: Clip rectangle area to copy
  *
  * Drm doesn't have native monochrome or grayscale support.
  * Such drivers can announce the commonly supported XR24 format to userspace
@@ -195,41 +197,31 @@ EXPORT_SYMBOL(tinydrm_xrgb8888_to_rgb565);
  * where 1 means foreground color and 0 background color.
  *
  * ITU BT.601 is used for the RGB -> luma (brightness) conversion.
- *
- * Returns:
- * Zero on success, negative error code on failure.
  */
-int tinydrm_xrgb8888_to_gray8(u8 *dst, struct drm_framebuffer *fb)
+void tinydrm_xrgb8888_to_gray8(u8 *dst, void *vaddr, struct drm_framebuffer *fb,
+			       struct drm_clip_rect *clip)
 {
-	struct drm_gem_cma_object *cma_obj = drm_fb_cma_get_gem_obj(fb, 0);
-	struct dma_buf_attachment *import_attach = cma_obj->base.import_attach;
-	unsigned int x, y, pitch = fb->pitches[0];
-	int ret = 0;
+	unsigned int len = (clip->x2 - clip->x1) * sizeof(u32);
+	unsigned int x, y;
 	void *buf;
 	u32 *src;
 
 	if (WARN_ON(fb->format->format != DRM_FORMAT_XRGB8888))
-		return -EINVAL;
+		return;
 	/*
 	 * The cma memory is write-combined so reads are uncached.
 	 * Speed up by fetching one line at a time.
 	 */
-	buf = kmalloc(pitch, GFP_KERNEL);
+	buf = kmalloc(len, GFP_KERNEL);
 	if (!buf)
-		return -ENOMEM;
-
-	if (import_attach) {
-		ret = dma_buf_begin_cpu_access(import_attach->dmabuf,
-					       DMA_FROM_DEVICE);
-		if (ret)
-			goto err_free;
-	}
+		return;
 
-	for (y = 0; y < fb->height; y++) {
-		src = cma_obj->vaddr + (y * pitch);
-		memcpy(buf, src, pitch);
+	for (y = clip->y1; y < clip->y2; y++) {
+		src = vaddr + (y * fb->pitches[0]);
+		src += clip->x1;
+		memcpy(buf, src, len);
 		src = buf;
-		for (x = 0; x < fb->width; x++) {
+		for (x = clip->x1; x < clip->x2; x++) {
 			u8 r = (*src & 0x00ff0000) >> 16;
 			u8 g = (*src & 0x0000ff00) >> 8;
 			u8 b =  *src & 0x000000ff;
@@ -240,13 +232,7 @@ int tinydrm_xrgb8888_to_gray8(u8 *dst, struct drm_framebuffer *fb)
 		}
 	}
 
-	if (import_attach)
-		ret = dma_buf_end_cpu_access(import_attach->dmabuf,
-					     DMA_FROM_DEVICE);
-err_free:
 	kfree(buf);
-
-	return ret;
 }
 EXPORT_SYMBOL(tinydrm_xrgb8888_to_gray8);
 
diff --git a/drivers/gpu/drm/tinydrm/core/tinydrm-pipe.c b/drivers/gpu/drm/tinydrm/core/tinydrm-pipe.c
index f224b54a30f6..177e9d861001 100644
--- a/drivers/gpu/drm/tinydrm/core/tinydrm-pipe.c
+++ b/drivers/gpu/drm/tinydrm/core/tinydrm-pipe.c
@@ -56,7 +56,7 @@ static const struct drm_connector_helper_funcs tinydrm_connector_hfuncs = {
 static enum drm_connector_status
 tinydrm_connector_detect(struct drm_connector *connector, bool force)
 {
-	if (drm_device_is_unplugged(connector->dev))
+	if (drm_dev_is_unplugged(connector->dev))
 		return connector_status_disconnected;
 
 	return connector->status;
diff --git a/drivers/gpu/drm/tinydrm/repaper.c b/drivers/gpu/drm/tinydrm/repaper.c
index 3343d3f15a90..30dc97b3ff21 100644
--- a/drivers/gpu/drm/tinydrm/repaper.c
+++ b/drivers/gpu/drm/tinydrm/repaper.c
@@ -18,6 +18,7 @@
  */
 
 #include <linux/delay.h>
+#include <linux/dma-buf.h>
 #include <linux/gpio/consumer.h>
 #include <linux/module.h>
 #include <linux/of_device.h>
@@ -525,11 +526,20 @@ static int repaper_fb_dirty(struct drm_framebuffer *fb,
 			    struct drm_clip_rect *clips,
 			    unsigned int num_clips)
 {
+	struct drm_gem_cma_object *cma_obj = drm_fb_cma_get_gem_obj(fb, 0);
+	struct dma_buf_attachment *import_attach = cma_obj->base.import_attach;
 	struct tinydrm_device *tdev = fb->dev->dev_private;
 	struct repaper_epd *epd = epd_from_tinydrm(tdev);
+	struct drm_clip_rect clip;
 	u8 *buf = NULL;
 	int ret = 0;
 
+	/* repaper can't do partial updates */
+	clip.x1 = 0;
+	clip.x2 = fb->width;
+	clip.y1 = 0;
+	clip.y2 = fb->height;
+
 	mutex_lock(&tdev->dirty_lock);
 
 	if (!epd->enabled)
@@ -550,9 +560,21 @@ static int repaper_fb_dirty(struct drm_framebuffer *fb,
 		goto out_unlock;
 	}
 
-	ret = tinydrm_xrgb8888_to_gray8(buf, fb);
-	if (ret)
-		goto out_unlock;
+	if (import_attach) {
+		ret = dma_buf_begin_cpu_access(import_attach->dmabuf,
+					       DMA_FROM_DEVICE);
+		if (ret)
+			goto out_unlock;
+	}
+
+	tinydrm_xrgb8888_to_gray8(buf, cma_obj->vaddr, fb, &clip);
+
+	if (import_attach) {
+		ret = dma_buf_end_cpu_access(import_attach->dmabuf,
+					     DMA_FROM_DEVICE);
+		if (ret)
+			goto out_unlock;
+	}
 
 	repaper_gray8_to_mono_reversed(buf, fb->width, fb->height);
 
diff --git a/drivers/gpu/drm/tinydrm/st7586.c b/drivers/gpu/drm/tinydrm/st7586.c
new file mode 100644
index 000000000000..b439956a07f4
--- /dev/null
+++ b/drivers/gpu/drm/tinydrm/st7586.c
@@ -0,0 +1,428 @@
+/*
+ * DRM driver for Sitronix ST7586 panels
+ *
+ * Copyright 2017 David Lechner <david@lechnology.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#include <linux/delay.h>
+#include <linux/dma-buf.h>
+#include <linux/gpio/consumer.h>
+#include <linux/module.h>
+#include <linux/property.h>
+#include <linux/spi/spi.h>
+#include <video/mipi_display.h>
+
+#include <drm/tinydrm/mipi-dbi.h>
+#include <drm/tinydrm/tinydrm-helpers.h>
+
+/* controller-specific commands */
+#define ST7586_DISP_MODE_GRAY	0x38
+#define ST7586_DISP_MODE_MONO	0x39
+#define ST7586_ENABLE_DDRAM	0x3a
+#define ST7586_SET_DISP_DUTY	0xb0
+#define ST7586_SET_PART_DISP	0xb4
+#define ST7586_SET_NLINE_INV	0xb5
+#define ST7586_SET_VOP		0xc0
+#define ST7586_SET_BIAS_SYSTEM	0xc3
+#define ST7586_SET_BOOST_LEVEL	0xc4
+#define ST7586_SET_VOP_OFFSET	0xc7
+#define ST7586_ENABLE_ANALOG	0xd0
+#define ST7586_AUTO_READ_CTRL	0xd7
+#define ST7586_OTP_RW_CTRL	0xe0
+#define ST7586_OTP_CTRL_OUT	0xe1
+#define ST7586_OTP_READ		0xe3
+
+#define ST7586_DISP_CTRL_MX	BIT(6)
+#define ST7586_DISP_CTRL_MY	BIT(7)
+
+/*
+ * The ST7586 controller has an unusual pixel format where 2bpp grayscale is
+ * packed 3 pixels per byte with the first two pixels using 3 bits and the 3rd
+ * pixel using only 2 bits.
+ *
+ * |  D7  |  D6  |  D5  ||      |      || 2bpp |
+ * | (D4) | (D3) | (D2) ||  D1  |  D0  || GRAY |
+ * +------+------+------++------+------++------+
+ * |  1   |  1   |  1   ||  1   |  1   || 0  0 | black
+ * |  1   |  0   |  0   ||  1   |  0   || 0  1 | dark gray
+ * |  0   |  1   |  0   ||  0   |  1   || 1  0 | light gray
+ * |  0   |  0   |  0   ||  0   |  0   || 1  1 | white
+ */
+
+static const u8 st7586_lookup[] = { 0x7, 0x4, 0x2, 0x0 };
+
+static void st7586_xrgb8888_to_gray332(u8 *dst, void *vaddr,
+				       struct drm_framebuffer *fb,
+				       struct drm_clip_rect *clip)
+{
+	size_t len = (clip->x2 - clip->x1) * (clip->y2 - clip->y1);
+	unsigned int x, y;
+	u8 *src, *buf, val;
+
+	buf = kmalloc(len, GFP_KERNEL);
+	if (!buf)
+		return;
+
+	tinydrm_xrgb8888_to_gray8(buf, vaddr, fb, clip);
+	src = buf;
+
+	for (y = clip->y1; y < clip->y2; y++) {
+		for (x = clip->x1; x < clip->x2; x += 3) {
+			val = st7586_lookup[*src++ >> 6] << 5;
+			val |= st7586_lookup[*src++ >> 6] << 2;
+			val |= st7586_lookup[*src++ >> 6] >> 1;
+			*dst++ = val;
+		}
+	}
+
+	kfree(buf);
+}
+
+static int st7586_buf_copy(void *dst, struct drm_framebuffer *fb,
+			   struct drm_clip_rect *clip)
+{
+	struct drm_gem_cma_object *cma_obj = drm_fb_cma_get_gem_obj(fb, 0);
+	struct dma_buf_attachment *import_attach = cma_obj->base.import_attach;
+	void *src = cma_obj->vaddr;
+	int ret = 0;
+
+	if (import_attach) {
+		ret = dma_buf_begin_cpu_access(import_attach->dmabuf,
+					       DMA_FROM_DEVICE);
+		if (ret)
+			return ret;
+	}
+
+	st7586_xrgb8888_to_gray332(dst, src, fb, clip);
+
+	if (import_attach)
+		ret = dma_buf_end_cpu_access(import_attach->dmabuf,
+					     DMA_FROM_DEVICE);
+
+	return ret;
+}
+
+static int st7586_fb_dirty(struct drm_framebuffer *fb,
+			   struct drm_file *file_priv, unsigned int flags,
+			   unsigned int color, struct drm_clip_rect *clips,
+			   unsigned int num_clips)
+{
+	struct tinydrm_device *tdev = fb->dev->dev_private;
+	struct mipi_dbi *mipi = mipi_dbi_from_tinydrm(tdev);
+	struct drm_clip_rect clip;
+	int start, end;
+	int ret = 0;
+
+	mutex_lock(&tdev->dirty_lock);
+
+	if (!mipi->enabled)
+		goto out_unlock;
+
+	/* fbdev can flush even when we're not interested */
+	if (tdev->pipe.plane.fb != fb)
+		goto out_unlock;
+
+	tinydrm_merge_clips(&clip, clips, num_clips, flags, fb->width,
+			    fb->height);
+
+	/* 3 pixels per byte, so grow clip to nearest multiple of 3 */
+	clip.x1 = rounddown(clip.x1, 3);
+	clip.x2 = roundup(clip.x2, 3);
+
+	DRM_DEBUG("Flushing [FB:%d] x1=%u, x2=%u, y1=%u, y2=%u\n", fb->base.id,
+		  clip.x1, clip.x2, clip.y1, clip.y2);
+
+	ret = st7586_buf_copy(mipi->tx_buf, fb, &clip);
+	if (ret)
+		goto out_unlock;
+
+	/* Pixels are packed 3 per byte */
+	start = clip.x1 / 3;
+	end = clip.x2 / 3;
+
+	mipi_dbi_command(mipi, MIPI_DCS_SET_COLUMN_ADDRESS,
+			 (start >> 8) & 0xFF, start & 0xFF,
+			 (end >> 8) & 0xFF, (end - 1) & 0xFF);
+	mipi_dbi_command(mipi, MIPI_DCS_SET_PAGE_ADDRESS,
+			 (clip.y1 >> 8) & 0xFF, clip.y1 & 0xFF,
+			 (clip.y2 >> 8) & 0xFF, (clip.y2 - 1) & 0xFF);
+
+	ret = mipi_dbi_command_buf(mipi, MIPI_DCS_WRITE_MEMORY_START,
+				   (u8 *)mipi->tx_buf,
+				   (end - start) * (clip.y2 - clip.y1));
+
+out_unlock:
+	mutex_unlock(&tdev->dirty_lock);
+
+	if (ret)
+		dev_err_once(fb->dev->dev, "Failed to update display %d\n",
+			     ret);
+
+	return ret;
+}
+
+static const struct drm_framebuffer_funcs st7586_fb_funcs = {
+	.destroy	= drm_fb_cma_destroy,
+	.create_handle	= drm_fb_cma_create_handle,
+	.dirty		= st7586_fb_dirty,
+};
+
+static void st7586_pipe_enable(struct drm_simple_display_pipe *pipe,
+			       struct drm_crtc_state *crtc_state)
+{
+	struct tinydrm_device *tdev = pipe_to_tinydrm(pipe);
+	struct mipi_dbi *mipi = mipi_dbi_from_tinydrm(tdev);
+	struct drm_framebuffer *fb = pipe->plane.fb;
+	struct device *dev = tdev->drm->dev;
+	int ret;
+	u8 addr_mode;
+
+	DRM_DEBUG_KMS("\n");
+
+	mipi_dbi_hw_reset(mipi);
+	ret = mipi_dbi_command(mipi, ST7586_AUTO_READ_CTRL, 0x9f);
+	if (ret) {
+		dev_err(dev, "Error sending command %d\n", ret);
+		return;
+	}
+
+	mipi_dbi_command(mipi, ST7586_OTP_RW_CTRL, 0x00);
+
+	msleep(10);
+
+	mipi_dbi_command(mipi, ST7586_OTP_READ);
+
+	msleep(20);
+
+	mipi_dbi_command(mipi, ST7586_OTP_CTRL_OUT);
+	mipi_dbi_command(mipi, MIPI_DCS_EXIT_SLEEP_MODE);
+	mipi_dbi_command(mipi, MIPI_DCS_SET_DISPLAY_OFF);
+
+	msleep(50);
+
+	mipi_dbi_command(mipi, ST7586_SET_VOP_OFFSET, 0x00);
+	mipi_dbi_command(mipi, ST7586_SET_VOP, 0xe3, 0x00);
+	mipi_dbi_command(mipi, ST7586_SET_BIAS_SYSTEM, 0x02);
+	mipi_dbi_command(mipi, ST7586_SET_BOOST_LEVEL, 0x04);
+	mipi_dbi_command(mipi, ST7586_ENABLE_ANALOG, 0x1d);
+	mipi_dbi_command(mipi, ST7586_SET_NLINE_INV, 0x00);
+	mipi_dbi_command(mipi, ST7586_DISP_MODE_GRAY);
+	mipi_dbi_command(mipi, ST7586_ENABLE_DDRAM, 0x02);
+
+	switch (mipi->rotation) {
+	default:
+		addr_mode = 0x00;
+		break;
+	case 90:
+		addr_mode = ST7586_DISP_CTRL_MY;
+		break;
+	case 180:
+		addr_mode = ST7586_DISP_CTRL_MX | ST7586_DISP_CTRL_MY;
+		break;
+	case 270:
+		addr_mode = ST7586_DISP_CTRL_MX;
+		break;
+	}
+	mipi_dbi_command(mipi, MIPI_DCS_SET_ADDRESS_MODE, addr_mode);
+
+	mipi_dbi_command(mipi, ST7586_SET_DISP_DUTY, 0x7f);
+	mipi_dbi_command(mipi, ST7586_SET_PART_DISP, 0xa0);
+	mipi_dbi_command(mipi, MIPI_DCS_SET_PARTIAL_AREA, 0x00, 0x00, 0x00, 0x77);
+	mipi_dbi_command(mipi, MIPI_DCS_EXIT_INVERT_MODE);
+
+	msleep(100);
+
+	mipi_dbi_command(mipi, MIPI_DCS_SET_DISPLAY_ON);
+
+	mipi->enabled = true;
+
+	if (fb)
+		fb->funcs->dirty(fb, NULL, 0, 0, NULL, 0);
+}
+
+static void st7586_pipe_disable(struct drm_simple_display_pipe *pipe)
+{
+	struct tinydrm_device *tdev = pipe_to_tinydrm(pipe);
+	struct mipi_dbi *mipi = mipi_dbi_from_tinydrm(tdev);
+
+	DRM_DEBUG_KMS("\n");
+
+	if (!mipi->enabled)
+		return;
+
+	mipi_dbi_command(mipi, MIPI_DCS_SET_DISPLAY_OFF);
+	mipi->enabled = false;
+}
+
+static const u32 st7586_formats[] = {
+	DRM_FORMAT_XRGB8888,
+};
+
+static int st7586_init(struct device *dev, struct mipi_dbi *mipi,
+		const struct drm_simple_display_pipe_funcs *pipe_funcs,
+		struct drm_driver *driver, const struct drm_display_mode *mode,
+		unsigned int rotation)
+{
+	size_t bufsize = (mode->vdisplay + 2) / 3 * mode->hdisplay;
+	struct tinydrm_device *tdev = &mipi->tinydrm;
+	int ret;
+
+	mutex_init(&mipi->cmdlock);
+
+	mipi->tx_buf = devm_kmalloc(dev, bufsize, GFP_KERNEL);
+	if (!mipi->tx_buf)
+		return -ENOMEM;
+
+	ret = devm_tinydrm_init(dev, tdev, &st7586_fb_funcs, driver);
+	if (ret)
+		return ret;
+
+	ret = tinydrm_display_pipe_init(tdev, pipe_funcs,
+					DRM_MODE_CONNECTOR_VIRTUAL,
+					st7586_formats,
+					ARRAY_SIZE(st7586_formats),
+					mode, rotation);
+	if (ret)
+		return ret;
+
+	tdev->drm->mode_config.preferred_depth = 32;
+	mipi->rotation = rotation;
+
+	drm_mode_config_reset(tdev->drm);
+
+	DRM_DEBUG_KMS("preferred_depth=%u, rotation = %u\n",
+		      tdev->drm->mode_config.preferred_depth, rotation);
+
+	return 0;
+}
+
+static const struct drm_simple_display_pipe_funcs st7586_pipe_funcs = {
+	.enable		= st7586_pipe_enable,
+	.disable	= st7586_pipe_disable,
+	.update		= tinydrm_display_pipe_update,
+	.prepare_fb	= tinydrm_display_pipe_prepare_fb,
+};
+
+static const struct drm_display_mode st7586_mode = {
+	TINYDRM_MODE(178, 128, 37, 27),
+};
+
+DEFINE_DRM_GEM_CMA_FOPS(st7586_fops);
+
+static struct drm_driver st7586_driver = {
+	.driver_features	= DRIVER_GEM | DRIVER_MODESET | DRIVER_PRIME |
+				  DRIVER_ATOMIC,
+	.fops			= &st7586_fops,
+	TINYDRM_GEM_DRIVER_OPS,
+	.lastclose		= tinydrm_lastclose,
+	.debugfs_init		= mipi_dbi_debugfs_init,
+	.name			= "st7586",
+	.desc			= "Sitronix ST7586",
+	.date			= "20170801",
+	.major			= 1,
+	.minor			= 0,
+};
+
+static const struct of_device_id st7586_of_match[] = {
+	{ .compatible = "lego,ev3-lcd" },
+	{},
+};
+MODULE_DEVICE_TABLE(of, st7586_of_match);
+
+static const struct spi_device_id st7586_id[] = {
+	{ "ev3-lcd", 0 },
+	{ },
+};
+MODULE_DEVICE_TABLE(spi, st7586_id);
+
+static int st7586_probe(struct spi_device *spi)
+{
+	struct device *dev = &spi->dev;
+	struct tinydrm_device *tdev;
+	struct mipi_dbi *mipi;
+	struct gpio_desc *a0;
+	u32 rotation = 0;
+	int ret;
+
+	mipi = devm_kzalloc(dev, sizeof(*mipi), GFP_KERNEL);
+	if (!mipi)
+		return -ENOMEM;
+
+	mipi->reset = devm_gpiod_get(dev, "reset", GPIOD_OUT_HIGH);
+	if (IS_ERR(mipi->reset)) {
+		dev_err(dev, "Failed to get gpio 'reset'\n");
+		return PTR_ERR(mipi->reset);
+	}
+
+	a0 = devm_gpiod_get(dev, "a0", GPIOD_OUT_LOW);
+	if (IS_ERR(a0)) {
+		dev_err(dev, "Failed to get gpio 'a0'\n");
+		return PTR_ERR(a0);
+	}
+
+	device_property_read_u32(dev, "rotation", &rotation);
+
+	ret = mipi_dbi_spi_init(spi, mipi, a0);
+	if (ret)
+		return ret;
+
+	/* Cannot read from this controller via SPI */
+	mipi->read_commands = NULL;
+
+	/*
+	 * we are using 8-bit data, so we are not actually swapping anything,
+	 * but setting mipi->swap_bytes makes mipi_dbi_typec3_command() do the
+	 * right thing and not use 16-bit transfers (which results in swapped
+	 * bytes on little-endian systems and causes out of order data to be
+	 * sent to the display).
+	 */
+	mipi->swap_bytes = true;
+
+	ret = st7586_init(&spi->dev, mipi, &st7586_pipe_funcs, &st7586_driver,
+			  &st7586_mode, rotation);
+	if (ret)
+		return ret;
+
+	tdev = &mipi->tinydrm;
+
+	ret = devm_tinydrm_register(tdev);
+	if (ret)
+		return ret;
+
+	spi_set_drvdata(spi, mipi);
+
+	DRM_DEBUG_DRIVER("Initialized %s:%s @%uMHz on minor %d\n",
+			 tdev->drm->driver->name, dev_name(dev),
+			 spi->max_speed_hz / 1000000,
+			 tdev->drm->primary->index);
+
+	return 0;
+}
+
+static void st7586_shutdown(struct spi_device *spi)
+{
+	struct mipi_dbi *mipi = spi_get_drvdata(spi);
+
+	tinydrm_shutdown(&mipi->tinydrm);
+}
+
+static struct spi_driver st7586_spi_driver = {
+	.driver = {
+		.name = "st7586",
+		.owner = THIS_MODULE,
+		.of_match_table = st7586_of_match,
+	},
+	.id_table = st7586_id,
+	.probe = st7586_probe,
+	.shutdown = st7586_shutdown,
+};
+module_spi_driver(st7586_spi_driver);
+
+MODULE_DESCRIPTION("Sitronix ST7586 DRM driver");
+MODULE_AUTHOR("David Lechner <david@lechnology.com>");
+MODULE_LICENSE("GPL");
diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c
index 22b57020790d..cba11f13d994 100644
--- a/drivers/gpu/drm/ttm/ttm_bo.c
+++ b/drivers/gpu/drm/ttm/ttm_bo.c
@@ -70,6 +70,7 @@ static inline int ttm_mem_type_from_place(const struct ttm_place *place,
 static void ttm_mem_type_debug(struct ttm_bo_device *bdev, int mem_type)
 {
 	struct ttm_mem_type_manager *man = &bdev->man[mem_type];
+	struct drm_printer p = drm_debug_printer(TTM_PFX);
 
 	pr_err("    has_type: %d\n", man->has_type);
 	pr_err("    use_type: %d\n", man->use_type);
@@ -79,7 +80,7 @@ static void ttm_mem_type_debug(struct ttm_bo_device *bdev, int mem_type)
 	pr_err("    available_caching: 0x%08X\n", man->available_caching);
 	pr_err("    default_caching: 0x%08X\n", man->default_caching);
 	if (mem_type != TTM_PL_SYSTEM)
-		(*man->func->debug)(man, TTM_PFX);
+		(*man->func->debug)(man, &p);
 }
 
 static void ttm_bo_mem_space_debug(struct ttm_buffer_object *bo,
@@ -394,14 +395,33 @@ static void ttm_bo_cleanup_memtype_use(struct ttm_buffer_object *bo)
 	ww_mutex_unlock (&bo->resv->lock);
 }
 
+static int ttm_bo_individualize_resv(struct ttm_buffer_object *bo)
+{
+	int r;
+
+	if (bo->resv == &bo->ttm_resv)
+		return 0;
+
+	reservation_object_init(&bo->ttm_resv);
+	BUG_ON(!reservation_object_trylock(&bo->ttm_resv));
+
+	r = reservation_object_copy_fences(&bo->ttm_resv, bo->resv);
+	if (r) {
+		reservation_object_unlock(&bo->ttm_resv);
+		reservation_object_fini(&bo->ttm_resv);
+	}
+
+	return r;
+}
+
 static void ttm_bo_flush_all_fences(struct ttm_buffer_object *bo)
 {
 	struct reservation_object_list *fobj;
 	struct dma_fence *fence;
 	int i;
 
-	fobj = reservation_object_get_list(bo->resv);
-	fence = reservation_object_get_excl(bo->resv);
+	fobj = reservation_object_get_list(&bo->ttm_resv);
+	fence = reservation_object_get_excl(&bo->ttm_resv);
 	if (fence && !fence->ops->signaled)
 		dma_fence_enable_sw_signaling(fence);
 
@@ -430,8 +450,19 @@ static void ttm_bo_cleanup_refs_or_queue(struct ttm_buffer_object *bo)
 			ttm_bo_cleanup_memtype_use(bo);
 
 			return;
-		} else
-			ttm_bo_flush_all_fences(bo);
+		}
+
+		ret = ttm_bo_individualize_resv(bo);
+		if (ret) {
+			/* Last resort, if we fail to allocate memory for the
+			 * fences block for the BO to become idle and free it.
+			 */
+			spin_unlock(&glob->lru_lock);
+			ttm_bo_wait(bo, true, true);
+			ttm_bo_cleanup_memtype_use(bo);
+			return;
+		}
+		ttm_bo_flush_all_fences(bo);
 
 		/*
 		 * Make NO_EVICT bos immediately available to
@@ -443,6 +474,8 @@ static void ttm_bo_cleanup_refs_or_queue(struct ttm_buffer_object *bo)
 			ttm_bo_add_to_lru(bo);
 		}
 
+		if (bo->resv != &bo->ttm_resv)
+			reservation_object_unlock(&bo->ttm_resv);
 		__ttm_bo_unreserve(bo);
 	}
 
@@ -471,17 +504,25 @@ static int ttm_bo_cleanup_refs_and_unlock(struct ttm_buffer_object *bo,
 					  bool no_wait_gpu)
 {
 	struct ttm_bo_global *glob = bo->glob;
+	struct reservation_object *resv;
 	int ret;
 
-	ret = ttm_bo_wait(bo, false, true);
+	if (unlikely(list_empty(&bo->ddestroy)))
+		resv = bo->resv;
+	else
+		resv = &bo->ttm_resv;
+
+	if (reservation_object_test_signaled_rcu(resv, true))
+		ret = 0;
+	else
+		ret = -EBUSY;
 
 	if (ret && !no_wait_gpu) {
 		long lret;
 		ww_mutex_unlock(&bo->resv->lock);
 		spin_unlock(&glob->lru_lock);
 
-		lret = reservation_object_wait_timeout_rcu(bo->resv,
-							   true,
+		lret = reservation_object_wait_timeout_rcu(resv, true,
 							   interruptible,
 							   30 * HZ);
 
@@ -505,13 +546,6 @@ static int ttm_bo_cleanup_refs_and_unlock(struct ttm_buffer_object *bo,
 			spin_unlock(&glob->lru_lock);
 			return 0;
 		}
-
-		/*
-		 * remove sync_obj with ttm_bo_wait, the wait should be
-		 * finished, and no new wait object should have been added.
-		 */
-		ret = ttm_bo_wait(bo, false, true);
-		WARN_ON(ret);
 	}
 
 	if (ret || unlikely(list_empty(&bo->ddestroy))) {
diff --git a/drivers/gpu/drm/ttm/ttm_bo_manager.c b/drivers/gpu/drm/ttm/ttm_bo_manager.c
index 90a6c0b03afc..a7c232dc39cb 100644
--- a/drivers/gpu/drm/ttm/ttm_bo_manager.c
+++ b/drivers/gpu/drm/ttm/ttm_bo_manager.c
@@ -136,13 +136,12 @@ static int ttm_bo_man_takedown(struct ttm_mem_type_manager *man)
 }
 
 static void ttm_bo_man_debug(struct ttm_mem_type_manager *man,
-			     const char *prefix)
+			     struct drm_printer *printer)
 {
 	struct ttm_range_manager *rman = (struct ttm_range_manager *) man->priv;
-	struct drm_printer p = drm_debug_printer(prefix);
 
 	spin_lock(&rman->lock);
-	drm_mm_print(&rman->mm, &p);
+	drm_mm_print(&rman->mm, printer);
 	spin_unlock(&rman->lock);
 }
 
diff --git a/drivers/gpu/drm/ttm/ttm_page_alloc.c b/drivers/gpu/drm/ttm/ttm_page_alloc.c
index eeddc1e48409..871599826773 100644
--- a/drivers/gpu/drm/ttm/ttm_page_alloc.c
+++ b/drivers/gpu/drm/ttm/ttm_page_alloc.c
@@ -615,7 +615,7 @@ static void ttm_page_pool_fill_locked(struct ttm_page_pool *pool,
 		} else {
 			pr_err("Failed to fill pool (%p)\n", pool);
 			/* If we have any pages left put them to the pool. */
-			list_for_each_entry(p, &pool->list, lru) {
+			list_for_each_entry(p, &new_pages, lru) {
 				++cpages;
 			}
 			list_splice(&new_pages, &pool->list);
diff --git a/drivers/gpu/drm/udl/udl_connector.c b/drivers/gpu/drm/udl/udl_connector.c
index d2f57c52f7db..9f9a49748d17 100644
--- a/drivers/gpu/drm/udl/udl_connector.c
+++ b/drivers/gpu/drm/udl/udl_connector.c
@@ -96,7 +96,7 @@ static int udl_mode_valid(struct drm_connector *connector,
 static enum drm_connector_status
 udl_detect(struct drm_connector *connector, bool force)
 {
-	if (drm_device_is_unplugged(connector->dev))
+	if (drm_dev_is_unplugged(connector->dev))
 		return connector_status_disconnected;
 	return connector_status_connected;
 }
diff --git a/drivers/gpu/drm/udl/udl_drv.c b/drivers/gpu/drm/udl/udl_drv.c
index 0f02e1acf0ba..31421b6b586e 100644
--- a/drivers/gpu/drm/udl/udl_drv.c
+++ b/drivers/gpu/drm/udl/udl_drv.c
@@ -54,7 +54,6 @@ static struct drm_driver driver = {
 
 	.dumb_create = udl_dumb_create,
 	.dumb_map_offset = udl_gem_mmap,
-	.dumb_destroy = drm_gem_dumb_destroy,
 	.fops = &udl_driver_fops,
 
 	.prime_handle_to_fd = drm_gem_prime_handle_to_fd,
@@ -102,7 +101,7 @@ static void udl_usb_disconnect(struct usb_interface *interface)
 	drm_kms_helper_poll_disable(dev);
 	udl_fbdev_unplug(dev);
 	udl_drop_usb(dev);
-	drm_unplug_dev(dev);
+	drm_dev_unplug(dev);
 }
 
 /*
@@ -112,7 +111,7 @@ static void udl_usb_disconnect(struct usb_interface *interface)
  * which is compatible with all known USB 2.0 era graphics chips and firmware,
  * but allows DisplayLink to increment those for any future incompatible chips
  */
-static struct usb_device_id id_table[] = {
+static const struct usb_device_id id_table[] = {
 	{.idVendor = 0x17e9, .bInterfaceClass = 0xff,
 	 .bInterfaceSubClass = 0x00,
 	 .bInterfaceProtocol = 0x00,
diff --git a/drivers/gpu/drm/udl/udl_fb.c b/drivers/gpu/drm/udl/udl_fb.c
index a5c54dc60def..b7ca90db4e80 100644
--- a/drivers/gpu/drm/udl/udl_fb.c
+++ b/drivers/gpu/drm/udl/udl_fb.c
@@ -198,7 +198,7 @@ static int udl_fb_open(struct fb_info *info, int user)
 	struct udl_device *udl = dev->dev_private;
 
 	/* If the USB device is gone, we don't accept new opens */
-	if (drm_device_is_unplugged(udl->ddev))
+	if (drm_dev_is_unplugged(udl->ddev))
 		return -ENODEV;
 
 	ufbdev->fb_count++;
@@ -309,7 +309,7 @@ static void udl_user_framebuffer_destroy(struct drm_framebuffer *fb)
 	struct udl_framebuffer *ufb = to_udl_fb(fb);
 
 	if (ufb->obj)
-		drm_gem_object_unreference_unlocked(&ufb->obj->base);
+		drm_gem_object_put_unlocked(&ufb->obj->base);
 
 	drm_framebuffer_cleanup(fb);
 	kfree(ufb);
@@ -403,7 +403,7 @@ static int udlfb_create(struct drm_fb_helper *helper,
 
 	return ret;
 out_gfree:
-	drm_gem_object_unreference_unlocked(&ufbdev->ufb.obj->base);
+	drm_gem_object_put_unlocked(&ufbdev->ufb.obj->base);
 out:
 	return ret;
 }
@@ -419,7 +419,7 @@ static void udl_fbdev_destroy(struct drm_device *dev,
 	drm_fb_helper_fini(&ufbdev->helper);
 	drm_framebuffer_unregister_private(&ufbdev->ufb.base);
 	drm_framebuffer_cleanup(&ufbdev->ufb.base);
-	drm_gem_object_unreference_unlocked(&ufbdev->ufb.obj->base);
+	drm_gem_object_put_unlocked(&ufbdev->ufb.obj->base);
 }
 
 int udl_fbdev_init(struct drm_device *dev)
diff --git a/drivers/gpu/drm/udl/udl_gem.c b/drivers/gpu/drm/udl/udl_gem.c
index db9ceceba30e..dee6bd9a3dd1 100644
--- a/drivers/gpu/drm/udl/udl_gem.c
+++ b/drivers/gpu/drm/udl/udl_gem.c
@@ -52,7 +52,7 @@ udl_gem_create(struct drm_file *file,
 		return ret;
 	}
 
-	drm_gem_object_unreference_unlocked(&obj->base);
+	drm_gem_object_put_unlocked(&obj->base);
 	*handle_p = handle;
 	return 0;
 }
@@ -234,7 +234,7 @@ int udl_gem_mmap(struct drm_file *file, struct drm_device *dev,
 	*offset = drm_vma_node_offset_addr(&gobj->base.vma_node);
 
 out:
-	drm_gem_object_unreference(&gobj->base);
+	drm_gem_object_put(&gobj->base);
 unlock:
 	mutex_unlock(&dev->struct_mutex);
 	return ret;
diff --git a/drivers/gpu/drm/vc4/vc4_bo.c b/drivers/gpu/drm/vc4/vc4_bo.c
index b24dd8685590..3afdbf4bc10b 100644
--- a/drivers/gpu/drm/vc4/vc4_bo.c
+++ b/drivers/gpu/drm/vc4/vc4_bo.c
@@ -366,7 +366,7 @@ int vc4_dumb_create(struct drm_file *file_priv,
 		return PTR_ERR(bo);
 
 	ret = drm_gem_handle_create(file_priv, &bo->base.base, &args->handle);
-	drm_gem_object_unreference_unlocked(&bo->base.base);
+	drm_gem_object_put_unlocked(&bo->base.base);
 
 	return ret;
 }
@@ -482,7 +482,7 @@ vc4_prime_export(struct drm_device *dev, struct drm_gem_object *obj, int flags)
 	struct vc4_bo *bo = to_vc4_bo(obj);
 
 	if (bo->validated_shader) {
-		DRM_ERROR("Attempting to export shader BO\n");
+		DRM_DEBUG("Attempting to export shader BO\n");
 		return ERR_PTR(-EINVAL);
 	}
 
@@ -503,7 +503,7 @@ int vc4_mmap(struct file *filp, struct vm_area_struct *vma)
 	bo = to_vc4_bo(gem_obj);
 
 	if (bo->validated_shader && (vma->vm_flags & VM_WRITE)) {
-		DRM_ERROR("mmaping of shader BOs for writing not allowed.\n");
+		DRM_DEBUG("mmaping of shader BOs for writing not allowed.\n");
 		return -EINVAL;
 	}
 
@@ -528,7 +528,7 @@ int vc4_prime_mmap(struct drm_gem_object *obj, struct vm_area_struct *vma)
 	struct vc4_bo *bo = to_vc4_bo(obj);
 
 	if (bo->validated_shader && (vma->vm_flags & VM_WRITE)) {
-		DRM_ERROR("mmaping of shader BOs for writing not allowed.\n");
+		DRM_DEBUG("mmaping of shader BOs for writing not allowed.\n");
 		return -EINVAL;
 	}
 
@@ -540,7 +540,7 @@ void *vc4_prime_vmap(struct drm_gem_object *obj)
 	struct vc4_bo *bo = to_vc4_bo(obj);
 
 	if (bo->validated_shader) {
-		DRM_ERROR("mmaping of shader BOs not allowed.\n");
+		DRM_DEBUG("mmaping of shader BOs not allowed.\n");
 		return ERR_PTR(-EINVAL);
 	}
 
@@ -581,7 +581,7 @@ int vc4_create_bo_ioctl(struct drm_device *dev, void *data,
 		return PTR_ERR(bo);
 
 	ret = drm_gem_handle_create(file_priv, &bo->base.base, &args->handle);
-	drm_gem_object_unreference_unlocked(&bo->base.base);
+	drm_gem_object_put_unlocked(&bo->base.base);
 
 	return ret;
 }
@@ -594,14 +594,14 @@ int vc4_mmap_bo_ioctl(struct drm_device *dev, void *data,
 
 	gem_obj = drm_gem_object_lookup(file_priv, args->handle);
 	if (!gem_obj) {
-		DRM_ERROR("Failed to look up GEM BO %d\n", args->handle);
+		DRM_DEBUG("Failed to look up GEM BO %d\n", args->handle);
 		return -EINVAL;
 	}
 
 	/* The mmap offset was set up at BO allocation time. */
 	args->offset = drm_vma_node_offset_addr(&gem_obj->vma_node);
 
-	drm_gem_object_unreference_unlocked(gem_obj);
+	drm_gem_object_put_unlocked(gem_obj);
 	return 0;
 }
 
@@ -657,7 +657,7 @@ vc4_create_shader_bo_ioctl(struct drm_device *dev, void *data,
 	ret = drm_gem_handle_create(file_priv, &bo->base.base, &args->handle);
 
  fail:
-	drm_gem_object_unreference_unlocked(&bo->base.base);
+	drm_gem_object_put_unlocked(&bo->base.base);
 
 	return ret;
 }
@@ -698,13 +698,13 @@ int vc4_set_tiling_ioctl(struct drm_device *dev, void *data,
 
 	gem_obj = drm_gem_object_lookup(file_priv, args->handle);
 	if (!gem_obj) {
-		DRM_ERROR("Failed to look up GEM BO %d\n", args->handle);
+		DRM_DEBUG("Failed to look up GEM BO %d\n", args->handle);
 		return -ENOENT;
 	}
 	bo = to_vc4_bo(gem_obj);
 	bo->t_format = t_format;
 
-	drm_gem_object_unreference_unlocked(gem_obj);
+	drm_gem_object_put_unlocked(gem_obj);
 
 	return 0;
 }
@@ -729,7 +729,7 @@ int vc4_get_tiling_ioctl(struct drm_device *dev, void *data,
 
 	gem_obj = drm_gem_object_lookup(file_priv, args->handle);
 	if (!gem_obj) {
-		DRM_ERROR("Failed to look up GEM BO %d\n", args->handle);
+		DRM_DEBUG("Failed to look up GEM BO %d\n", args->handle);
 		return -ENOENT;
 	}
 	bo = to_vc4_bo(gem_obj);
@@ -739,7 +739,7 @@ int vc4_get_tiling_ioctl(struct drm_device *dev, void *data,
 	else
 		args->modifier = DRM_FORMAT_MOD_NONE;
 
-	drm_gem_object_unreference_unlocked(gem_obj);
+	drm_gem_object_put_unlocked(gem_obj);
 
 	return 0;
 }
@@ -830,7 +830,7 @@ int vc4_label_bo_ioctl(struct drm_device *dev, void *data,
 		ret = -ENOMEM;
 	mutex_unlock(&vc4->bo_lock);
 
-	drm_gem_object_unreference_unlocked(gem_obj);
+	drm_gem_object_put_unlocked(gem_obj);
 
 	return ret;
 }
diff --git a/drivers/gpu/drm/vc4/vc4_crtc.c b/drivers/gpu/drm/vc4/vc4_crtc.c
index 664a55b45af0..ce1e3b9e14c9 100644
--- a/drivers/gpu/drm/vc4/vc4_crtc.c
+++ b/drivers/gpu/drm/vc4/vc4_crtc.c
@@ -763,7 +763,7 @@ vc4_async_page_flip_complete(struct vc4_seqno_cb *cb)
 	}
 
 	drm_crtc_vblank_put(crtc);
-	drm_framebuffer_unreference(flip_state->fb);
+	drm_framebuffer_put(flip_state->fb);
 	kfree(flip_state);
 
 	up(&vc4->async_modeset);
@@ -792,7 +792,7 @@ static int vc4_async_page_flip(struct drm_crtc *crtc,
 	if (!flip_state)
 		return -ENOMEM;
 
-	drm_framebuffer_reference(fb);
+	drm_framebuffer_get(fb);
 	flip_state->fb = fb;
 	flip_state->crtc = crtc;
 	flip_state->event = event;
@@ -800,7 +800,7 @@ static int vc4_async_page_flip(struct drm_crtc *crtc,
 	/* Make sure all other async modesetes have landed. */
 	ret = down_interruptible(&vc4->async_modeset);
 	if (ret) {
-		drm_framebuffer_unreference(fb);
+		drm_framebuffer_put(fb);
 		kfree(flip_state);
 		return ret;
 	}
diff --git a/drivers/gpu/drm/vc4/vc4_drv.c b/drivers/gpu/drm/vc4/vc4_drv.c
index e8f0e1790d5e..1c96edcb302b 100644
--- a/drivers/gpu/drm/vc4/vc4_drv.c
+++ b/drivers/gpu/drm/vc4/vc4_drv.c
@@ -99,6 +99,7 @@ static int vc4_get_param_ioctl(struct drm_device *dev, void *data,
 	case DRM_VC4_PARAM_SUPPORTS_BRANCHES:
 	case DRM_VC4_PARAM_SUPPORTS_ETC1:
 	case DRM_VC4_PARAM_SUPPORTS_THREADED_FS:
+	case DRM_VC4_PARAM_SUPPORTS_FIXED_RCL_ORDER:
 		args->value = true;
 		break;
 	default:
diff --git a/drivers/gpu/drm/vc4/vc4_dsi.c b/drivers/gpu/drm/vc4/vc4_dsi.c
index 629d372633e6..d1e0dc908048 100644
--- a/drivers/gpu/drm/vc4/vc4_dsi.c
+++ b/drivers/gpu/drm/vc4/vc4_dsi.c
@@ -1636,14 +1636,10 @@ static void vc4_dsi_unbind(struct device *dev, struct device *master,
 
 	pm_runtime_disable(dev);
 
-	drm_bridge_remove(dsi->bridge);
 	vc4_dsi_encoder_destroy(dsi->encoder);
 
 	mipi_dsi_host_unregister(&dsi->dsi_host);
 
-	clk_disable_unprepare(dsi->pll_phy_clock);
-	clk_disable_unprepare(dsi->escape_clock);
-
 	if (dsi->port == 1)
 		vc4->dsi1 = NULL;
 }
diff --git a/drivers/gpu/drm/vc4/vc4_gem.c b/drivers/gpu/drm/vc4/vc4_gem.c
index 209fccd0d3b4..d0c6bfb68c4e 100644
--- a/drivers/gpu/drm/vc4/vc4_gem.c
+++ b/drivers/gpu/drm/vc4/vc4_gem.c
@@ -55,7 +55,7 @@ vc4_free_hang_state(struct drm_device *dev, struct vc4_hang_state *state)
 	unsigned int i;
 
 	for (i = 0; i < state->user_state.bo_count; i++)
-		drm_gem_object_unreference_unlocked(state->bo[i]);
+		drm_gem_object_put_unlocked(state->bo[i]);
 
 	kfree(state);
 }
@@ -188,12 +188,12 @@ vc4_save_hang_state(struct drm_device *dev)
 			continue;
 
 		for (j = 0; j < exec[i]->bo_count; j++) {
-			drm_gem_object_reference(&exec[i]->bo[j]->base);
+			drm_gem_object_get(&exec[i]->bo[j]->base);
 			kernel_state->bo[j + prev_idx] = &exec[i]->bo[j]->base;
 		}
 
 		list_for_each_entry(bo, &exec[i]->unref_list, unref_head) {
-			drm_gem_object_reference(&bo->base.base);
+			drm_gem_object_get(&bo->base.base);
 			kernel_state->bo[j + prev_idx] = &bo->base.base;
 			j++;
 		}
@@ -659,7 +659,7 @@ vc4_cl_lookup_bos(struct drm_device *dev,
 		/* See comment on bo_index for why we have to check
 		 * this.
 		 */
-		DRM_ERROR("Rendering requires BOs to validate\n");
+		DRM_DEBUG("Rendering requires BOs to validate\n");
 		return -EINVAL;
 	}
 
@@ -690,13 +690,13 @@ vc4_cl_lookup_bos(struct drm_device *dev,
 		struct drm_gem_object *bo = idr_find(&file_priv->object_idr,
 						     handles[i]);
 		if (!bo) {
-			DRM_ERROR("Failed to look up GEM BO %d: %d\n",
+			DRM_DEBUG("Failed to look up GEM BO %d: %d\n",
 				  i, handles[i]);
 			ret = -EINVAL;
 			spin_unlock(&file_priv->table_lock);
 			goto fail;
 		}
-		drm_gem_object_reference(bo);
+		drm_gem_object_get(bo);
 		exec->bo[i] = (struct drm_gem_cma_object *)bo;
 	}
 	spin_unlock(&file_priv->table_lock);
@@ -728,7 +728,7 @@ vc4_get_bcl(struct drm_device *dev, struct vc4_exec_info *exec)
 	    args->shader_rec_count >= (UINT_MAX /
 					  sizeof(struct vc4_shader_state)) ||
 	    temp_size < exec_size) {
-		DRM_ERROR("overflow in exec arguments\n");
+		DRM_DEBUG("overflow in exec arguments\n");
 		ret = -EINVAL;
 		goto fail;
 	}
@@ -834,7 +834,7 @@ vc4_complete_exec(struct drm_device *dev, struct vc4_exec_info *exec)
 
 	if (exec->bo) {
 		for (i = 0; i < exec->bo_count; i++)
-			drm_gem_object_unreference_unlocked(&exec->bo[i]->base);
+			drm_gem_object_put_unlocked(&exec->bo[i]->base);
 		kvfree(exec->bo);
 	}
 
@@ -842,7 +842,7 @@ vc4_complete_exec(struct drm_device *dev, struct vc4_exec_info *exec)
 		struct vc4_bo *bo = list_first_entry(&exec->unref_list,
 						     struct vc4_bo, unref_head);
 		list_del(&bo->unref_head);
-		drm_gem_object_unreference_unlocked(&bo->base.base);
+		drm_gem_object_put_unlocked(&bo->base.base);
 	}
 
 	/* Free up the allocation of any bin slots we used. */
@@ -973,7 +973,7 @@ vc4_wait_bo_ioctl(struct drm_device *dev, void *data,
 
 	gem_obj = drm_gem_object_lookup(file_priv, args->handle);
 	if (!gem_obj) {
-		DRM_ERROR("Failed to look up GEM BO %d\n", args->handle);
+		DRM_DEBUG("Failed to look up GEM BO %d\n", args->handle);
 		return -EINVAL;
 	}
 	bo = to_vc4_bo(gem_obj);
@@ -981,7 +981,7 @@ vc4_wait_bo_ioctl(struct drm_device *dev, void *data,
 	ret = vc4_wait_for_seqno_ioctl_helper(dev, bo->seqno,
 					      &args->timeout_ns);
 
-	drm_gem_object_unreference_unlocked(gem_obj);
+	drm_gem_object_put_unlocked(gem_obj);
 	return ret;
 }
 
@@ -1007,8 +1007,11 @@ vc4_submit_cl_ioctl(struct drm_device *dev, void *data,
 	struct ww_acquire_ctx acquire_ctx;
 	int ret = 0;
 
-	if ((args->flags & ~VC4_SUBMIT_CL_USE_CLEAR_COLOR) != 0) {
-		DRM_ERROR("Unknown flags: 0x%02x\n", args->flags);
+	if ((args->flags & ~(VC4_SUBMIT_CL_USE_CLEAR_COLOR |
+			     VC4_SUBMIT_CL_FIXED_RCL_ORDER |
+			     VC4_SUBMIT_CL_RCL_ORDER_INCREASING_X |
+			     VC4_SUBMIT_CL_RCL_ORDER_INCREASING_Y)) != 0) {
+		DRM_DEBUG("Unknown flags: 0x%02x\n", args->flags);
 		return -EINVAL;
 	}
 
@@ -1117,6 +1120,4 @@ vc4_gem_destroy(struct drm_device *dev)
 
 	if (vc4->hang_state)
 		vc4_free_hang_state(dev, vc4->hang_state);
-
-	vc4_bo_cache_destroy(dev);
 }
diff --git a/drivers/gpu/drm/vc4/vc4_hdmi.c b/drivers/gpu/drm/vc4/vc4_hdmi.c
index ff09b8e2f9ee..937da8dd65b8 100644
--- a/drivers/gpu/drm/vc4/vc4_hdmi.c
+++ b/drivers/gpu/drm/vc4/vc4_hdmi.c
@@ -288,6 +288,7 @@ static int vc4_hdmi_connector_get_modes(struct drm_connector *connector)
 	drm_mode_connector_update_edid_property(connector, edid);
 	ret = drm_add_edid_modes(connector, edid);
 	drm_edid_to_eld(connector, edid);
+	kfree(edid);
 
 	return ret;
 }
diff --git a/drivers/gpu/drm/vc4/vc4_kms.c b/drivers/gpu/drm/vc4/vc4_kms.c
index aeec6e8703d2..50c4959b5bd3 100644
--- a/drivers/gpu/drm/vc4/vc4_kms.c
+++ b/drivers/gpu/drm/vc4/vc4_kms.c
@@ -20,6 +20,7 @@
 #include <drm/drm_crtc_helper.h>
 #include <drm/drm_plane_helper.h>
 #include <drm/drm_fb_cma_helper.h>
+#include <drm/drm_gem_framebuffer_helper.h>
 #include "vc4_drv.h"
 
 static void vc4_output_poll_changed(struct drm_device *dev)
@@ -169,7 +170,7 @@ static struct drm_framebuffer *vc4_fb_create(struct drm_device *dev,
 		gem_obj = drm_gem_object_lookup(file_priv,
 						mode_cmd->handles[0]);
 		if (!gem_obj) {
-			DRM_ERROR("Failed to look up GEM BO %d\n",
+			DRM_DEBUG("Failed to look up GEM BO %d\n",
 				  mode_cmd->handles[0]);
 			return ERR_PTR(-ENOENT);
 		}
@@ -184,12 +185,12 @@ static struct drm_framebuffer *vc4_fb_create(struct drm_device *dev,
 			mode_cmd_local.modifier[0] = DRM_FORMAT_MOD_NONE;
 		}
 
-		drm_gem_object_unreference_unlocked(gem_obj);
+		drm_gem_object_put_unlocked(gem_obj);
 
 		mode_cmd = &mode_cmd_local;
 	}
 
-	return drm_fb_cma_create(dev, file_priv, mode_cmd);
+	return drm_gem_fb_create(dev, file_priv, mode_cmd);
 }
 
 static const struct drm_mode_config_funcs vc4_mode_funcs = {
diff --git a/drivers/gpu/drm/vc4/vc4_render_cl.c b/drivers/gpu/drm/vc4/vc4_render_cl.c
index 4a8051532f00..273984f71ae2 100644
--- a/drivers/gpu/drm/vc4/vc4_render_cl.c
+++ b/drivers/gpu/drm/vc4/vc4_render_cl.c
@@ -261,8 +261,17 @@ static int vc4_create_rcl_bo(struct drm_device *dev, struct vc4_exec_info *exec,
 	uint8_t max_y_tile = args->max_y_tile;
 	uint8_t xtiles = max_x_tile - min_x_tile + 1;
 	uint8_t ytiles = max_y_tile - min_y_tile + 1;
-	uint8_t x, y;
+	uint8_t xi, yi;
 	uint32_t size, loop_body_size;
+	bool positive_x = true;
+	bool positive_y = true;
+
+	if (args->flags & VC4_SUBMIT_CL_FIXED_RCL_ORDER) {
+		if (!(args->flags & VC4_SUBMIT_CL_RCL_ORDER_INCREASING_X))
+			positive_x = false;
+		if (!(args->flags & VC4_SUBMIT_CL_RCL_ORDER_INCREASING_Y))
+			positive_y = false;
+	}
 
 	size = VC4_PACKET_TILE_RENDERING_MODE_CONFIG_SIZE;
 	loop_body_size = VC4_PACKET_TILE_COORDINATES_SIZE;
@@ -354,10 +363,12 @@ static int vc4_create_rcl_bo(struct drm_device *dev, struct vc4_exec_info *exec,
 	rcl_u16(setup, args->height);
 	rcl_u16(setup, args->color_write.bits);
 
-	for (y = min_y_tile; y <= max_y_tile; y++) {
-		for (x = min_x_tile; x <= max_x_tile; x++) {
-			bool first = (x == min_x_tile && y == min_y_tile);
-			bool last = (x == max_x_tile && y == max_y_tile);
+	for (yi = 0; yi < ytiles; yi++) {
+		int y = positive_y ? min_y_tile + yi : max_y_tile - yi;
+		for (xi = 0; xi < xtiles; xi++) {
+			int x = positive_x ? min_x_tile + xi : max_x_tile - xi;
+			bool first = (xi == 0 && yi == 0);
+			bool last = (xi == xtiles - 1 && yi == ytiles - 1);
 
 			emit_tile(exec, setup, x, y, first, last);
 		}
@@ -378,14 +389,14 @@ static int vc4_full_res_bounds_check(struct vc4_exec_info *exec,
 	u32 render_tiles_stride = DIV_ROUND_UP(exec->args->width, 32);
 
 	if (surf->offset > obj->base.size) {
-		DRM_ERROR("surface offset %d > BO size %zd\n",
+		DRM_DEBUG("surface offset %d > BO size %zd\n",
 			  surf->offset, obj->base.size);
 		return -EINVAL;
 	}
 
 	if ((obj->base.size - surf->offset) / VC4_TILE_BUFFER_SIZE <
 	    render_tiles_stride * args->max_y_tile + args->max_x_tile) {
-		DRM_ERROR("MSAA tile %d, %d out of bounds "
+		DRM_DEBUG("MSAA tile %d, %d out of bounds "
 			  "(bo size %zd, offset %d).\n",
 			  args->max_x_tile, args->max_y_tile,
 			  obj->base.size,
@@ -401,7 +412,7 @@ static int vc4_rcl_msaa_surface_setup(struct vc4_exec_info *exec,
 				      struct drm_vc4_submit_rcl_surface *surf)
 {
 	if (surf->flags != 0 || surf->bits != 0) {
-		DRM_ERROR("MSAA surface had nonzero flags/bits\n");
+		DRM_DEBUG("MSAA surface had nonzero flags/bits\n");
 		return -EINVAL;
 	}
 
@@ -415,7 +426,7 @@ static int vc4_rcl_msaa_surface_setup(struct vc4_exec_info *exec,
 	exec->rcl_write_bo[exec->rcl_write_bo_count++] = *obj;
 
 	if (surf->offset & 0xf) {
-		DRM_ERROR("MSAA write must be 16b aligned.\n");
+		DRM_DEBUG("MSAA write must be 16b aligned.\n");
 		return -EINVAL;
 	}
 
@@ -437,7 +448,7 @@ static int vc4_rcl_surface_setup(struct vc4_exec_info *exec,
 	int ret;
 
 	if (surf->flags & ~VC4_SUBMIT_RCL_SURFACE_READ_IS_FULL_RES) {
-		DRM_ERROR("Extra flags set\n");
+		DRM_DEBUG("Extra flags set\n");
 		return -EINVAL;
 	}
 
@@ -453,12 +464,12 @@ static int vc4_rcl_surface_setup(struct vc4_exec_info *exec,
 
 	if (surf->flags & VC4_SUBMIT_RCL_SURFACE_READ_IS_FULL_RES) {
 		if (surf == &exec->args->zs_write) {
-			DRM_ERROR("general zs write may not be a full-res.\n");
+			DRM_DEBUG("general zs write may not be a full-res.\n");
 			return -EINVAL;
 		}
 
 		if (surf->bits != 0) {
-			DRM_ERROR("load/store general bits set with "
+			DRM_DEBUG("load/store general bits set with "
 				  "full res load/store.\n");
 			return -EINVAL;
 		}
@@ -473,19 +484,19 @@ static int vc4_rcl_surface_setup(struct vc4_exec_info *exec,
 	if (surf->bits & ~(VC4_LOADSTORE_TILE_BUFFER_TILING_MASK |
 			   VC4_LOADSTORE_TILE_BUFFER_BUFFER_MASK |
 			   VC4_LOADSTORE_TILE_BUFFER_FORMAT_MASK)) {
-		DRM_ERROR("Unknown bits in load/store: 0x%04x\n",
+		DRM_DEBUG("Unknown bits in load/store: 0x%04x\n",
 			  surf->bits);
 		return -EINVAL;
 	}
 
 	if (tiling > VC4_TILING_FORMAT_LT) {
-		DRM_ERROR("Bad tiling format\n");
+		DRM_DEBUG("Bad tiling format\n");
 		return -EINVAL;
 	}
 
 	if (buffer == VC4_LOADSTORE_TILE_BUFFER_ZS) {
 		if (format != 0) {
-			DRM_ERROR("No color format should be set for ZS\n");
+			DRM_DEBUG("No color format should be set for ZS\n");
 			return -EINVAL;
 		}
 		cpp = 4;
@@ -499,16 +510,16 @@ static int vc4_rcl_surface_setup(struct vc4_exec_info *exec,
 			cpp = 4;
 			break;
 		default:
-			DRM_ERROR("Bad tile buffer format\n");
+			DRM_DEBUG("Bad tile buffer format\n");
 			return -EINVAL;
 		}
 	} else {
-		DRM_ERROR("Bad load/store buffer %d.\n", buffer);
+		DRM_DEBUG("Bad load/store buffer %d.\n", buffer);
 		return -EINVAL;
 	}
 
 	if (surf->offset & 0xf) {
-		DRM_ERROR("load/store buffer must be 16b aligned.\n");
+		DRM_DEBUG("load/store buffer must be 16b aligned.\n");
 		return -EINVAL;
 	}
 
@@ -533,7 +544,7 @@ vc4_rcl_render_config_surface_setup(struct vc4_exec_info *exec,
 	int cpp;
 
 	if (surf->flags != 0) {
-		DRM_ERROR("No flags supported on render config.\n");
+		DRM_DEBUG("No flags supported on render config.\n");
 		return -EINVAL;
 	}
 
@@ -541,7 +552,7 @@ vc4_rcl_render_config_surface_setup(struct vc4_exec_info *exec,
 			   VC4_RENDER_CONFIG_FORMAT_MASK |
 			   VC4_RENDER_CONFIG_MS_MODE_4X |
 			   VC4_RENDER_CONFIG_DECIMATE_MODE_4X)) {
-		DRM_ERROR("Unknown bits in render config: 0x%04x\n",
+		DRM_DEBUG("Unknown bits in render config: 0x%04x\n",
 			  surf->bits);
 		return -EINVAL;
 	}
@@ -556,7 +567,7 @@ vc4_rcl_render_config_surface_setup(struct vc4_exec_info *exec,
 	exec->rcl_write_bo[exec->rcl_write_bo_count++] = *obj;
 
 	if (tiling > VC4_TILING_FORMAT_LT) {
-		DRM_ERROR("Bad tiling format\n");
+		DRM_DEBUG("Bad tiling format\n");
 		return -EINVAL;
 	}
 
@@ -569,7 +580,7 @@ vc4_rcl_render_config_surface_setup(struct vc4_exec_info *exec,
 		cpp = 4;
 		break;
 	default:
-		DRM_ERROR("Bad tile buffer format\n");
+		DRM_DEBUG("Bad tile buffer format\n");
 		return -EINVAL;
 	}
 
@@ -590,7 +601,7 @@ int vc4_get_rcl(struct drm_device *dev, struct vc4_exec_info *exec)
 
 	if (args->min_x_tile > args->max_x_tile ||
 	    args->min_y_tile > args->max_y_tile) {
-		DRM_ERROR("Bad render tile set (%d,%d)-(%d,%d)\n",
+		DRM_DEBUG("Bad render tile set (%d,%d)-(%d,%d)\n",
 			  args->min_x_tile, args->min_y_tile,
 			  args->max_x_tile, args->max_y_tile);
 		return -EINVAL;
@@ -599,7 +610,7 @@ int vc4_get_rcl(struct drm_device *dev, struct vc4_exec_info *exec)
 	if (has_bin &&
 	    (args->max_x_tile > exec->bin_tiles_x ||
 	     args->max_y_tile > exec->bin_tiles_y)) {
-		DRM_ERROR("Render tiles (%d,%d) outside of bin config "
+		DRM_DEBUG("Render tiles (%d,%d) outside of bin config "
 			  "(%d,%d)\n",
 			  args->max_x_tile, args->max_y_tile,
 			  exec->bin_tiles_x, exec->bin_tiles_y);
@@ -642,7 +653,7 @@ int vc4_get_rcl(struct drm_device *dev, struct vc4_exec_info *exec)
 	 */
 	if (!setup.color_write && !setup.zs_write &&
 	    !setup.msaa_color_write && !setup.msaa_zs_write) {
-		DRM_ERROR("RCL requires color or Z/S write\n");
+		DRM_DEBUG("RCL requires color or Z/S write\n");
 		return -EINVAL;
 	}
 
diff --git a/drivers/gpu/drm/vc4/vc4_validate.c b/drivers/gpu/drm/vc4/vc4_validate.c
index 814b512c6b9a..2db485abb186 100644
--- a/drivers/gpu/drm/vc4/vc4_validate.c
+++ b/drivers/gpu/drm/vc4/vc4_validate.c
@@ -109,7 +109,7 @@ vc4_use_bo(struct vc4_exec_info *exec, uint32_t hindex)
 	struct vc4_bo *bo;
 
 	if (hindex >= exec->bo_count) {
-		DRM_ERROR("BO index %d greater than BO count %d\n",
+		DRM_DEBUG("BO index %d greater than BO count %d\n",
 			  hindex, exec->bo_count);
 		return NULL;
 	}
@@ -117,7 +117,7 @@ vc4_use_bo(struct vc4_exec_info *exec, uint32_t hindex)
 	bo = to_vc4_bo(&obj->base);
 
 	if (bo->validated_shader) {
-		DRM_ERROR("Trying to use shader BO as something other than "
+		DRM_DEBUG("Trying to use shader BO as something other than "
 			  "a shader\n");
 		return NULL;
 	}
@@ -172,7 +172,7 @@ vc4_check_tex_size(struct vc4_exec_info *exec, struct drm_gem_cma_object *fbo,
 	 * our math.
 	 */
 	if (width > 4096 || height > 4096) {
-		DRM_ERROR("Surface dimensions (%d,%d) too large",
+		DRM_DEBUG("Surface dimensions (%d,%d) too large",
 			  width, height);
 		return false;
 	}
@@ -191,7 +191,7 @@ vc4_check_tex_size(struct vc4_exec_info *exec, struct drm_gem_cma_object *fbo,
 		aligned_height = round_up(height, utile_h);
 		break;
 	default:
-		DRM_ERROR("buffer tiling %d unsupported\n", tiling_format);
+		DRM_DEBUG("buffer tiling %d unsupported\n", tiling_format);
 		return false;
 	}
 
@@ -200,7 +200,7 @@ vc4_check_tex_size(struct vc4_exec_info *exec, struct drm_gem_cma_object *fbo,
 
 	if (size + offset < size ||
 	    size + offset > fbo->base.size) {
-		DRM_ERROR("Overflow in %dx%d (%dx%d) fbo size (%d + %d > %zd)\n",
+		DRM_DEBUG("Overflow in %dx%d (%dx%d) fbo size (%d + %d > %zd)\n",
 			  width, height,
 			  aligned_width, aligned_height,
 			  size, offset, fbo->base.size);
@@ -214,7 +214,7 @@ static int
 validate_flush(VALIDATE_ARGS)
 {
 	if (!validate_bin_pos(exec, untrusted, exec->args->bin_cl_size - 1)) {
-		DRM_ERROR("Bin CL must end with VC4_PACKET_FLUSH\n");
+		DRM_DEBUG("Bin CL must end with VC4_PACKET_FLUSH\n");
 		return -EINVAL;
 	}
 	exec->found_flush = true;
@@ -226,13 +226,13 @@ static int
 validate_start_tile_binning(VALIDATE_ARGS)
 {
 	if (exec->found_start_tile_binning_packet) {
-		DRM_ERROR("Duplicate VC4_PACKET_START_TILE_BINNING\n");
+		DRM_DEBUG("Duplicate VC4_PACKET_START_TILE_BINNING\n");
 		return -EINVAL;
 	}
 	exec->found_start_tile_binning_packet = true;
 
 	if (!exec->found_tile_binning_mode_config_packet) {
-		DRM_ERROR("missing VC4_PACKET_TILE_BINNING_MODE_CONFIG\n");
+		DRM_DEBUG("missing VC4_PACKET_TILE_BINNING_MODE_CONFIG\n");
 		return -EINVAL;
 	}
 
@@ -243,7 +243,7 @@ static int
 validate_increment_semaphore(VALIDATE_ARGS)
 {
 	if (!validate_bin_pos(exec, untrusted, exec->args->bin_cl_size - 2)) {
-		DRM_ERROR("Bin CL must end with "
+		DRM_DEBUG("Bin CL must end with "
 			  "VC4_PACKET_INCREMENT_SEMAPHORE\n");
 		return -EINVAL;
 	}
@@ -264,7 +264,7 @@ validate_indexed_prim_list(VALIDATE_ARGS)
 
 	/* Check overflow condition */
 	if (exec->shader_state_count == 0) {
-		DRM_ERROR("shader state must precede primitives\n");
+		DRM_DEBUG("shader state must precede primitives\n");
 		return -EINVAL;
 	}
 	shader_state = &exec->shader_state[exec->shader_state_count - 1];
@@ -281,7 +281,7 @@ validate_indexed_prim_list(VALIDATE_ARGS)
 
 	if (offset > ib->base.size ||
 	    (ib->base.size - offset) / index_size < length) {
-		DRM_ERROR("IB access overflow (%d + %d*%d > %zd)\n",
+		DRM_DEBUG("IB access overflow (%d + %d*%d > %zd)\n",
 			  offset, length, index_size, ib->base.size);
 		return -EINVAL;
 	}
@@ -301,13 +301,13 @@ validate_gl_array_primitive(VALIDATE_ARGS)
 
 	/* Check overflow condition */
 	if (exec->shader_state_count == 0) {
-		DRM_ERROR("shader state must precede primitives\n");
+		DRM_DEBUG("shader state must precede primitives\n");
 		return -EINVAL;
 	}
 	shader_state = &exec->shader_state[exec->shader_state_count - 1];
 
 	if (length + base_index < length) {
-		DRM_ERROR("primitive vertex count overflow\n");
+		DRM_DEBUG("primitive vertex count overflow\n");
 		return -EINVAL;
 	}
 	max_index = length + base_index - 1;
@@ -324,7 +324,7 @@ validate_gl_shader_state(VALIDATE_ARGS)
 	uint32_t i = exec->shader_state_count++;
 
 	if (i >= exec->shader_state_size) {
-		DRM_ERROR("More requests for shader states than declared\n");
+		DRM_DEBUG("More requests for shader states than declared\n");
 		return -EINVAL;
 	}
 
@@ -332,7 +332,7 @@ validate_gl_shader_state(VALIDATE_ARGS)
 	exec->shader_state[i].max_index = 0;
 
 	if (exec->shader_state[i].addr & ~0xf) {
-		DRM_ERROR("high bits set in GL shader rec reference\n");
+		DRM_DEBUG("high bits set in GL shader rec reference\n");
 		return -EINVAL;
 	}
 
@@ -356,7 +356,7 @@ validate_tile_binning_config(VALIDATE_ARGS)
 	int bin_slot;
 
 	if (exec->found_tile_binning_mode_config_packet) {
-		DRM_ERROR("Duplicate VC4_PACKET_TILE_BINNING_MODE_CONFIG\n");
+		DRM_DEBUG("Duplicate VC4_PACKET_TILE_BINNING_MODE_CONFIG\n");
 		return -EINVAL;
 	}
 	exec->found_tile_binning_mode_config_packet = true;
@@ -368,14 +368,14 @@ validate_tile_binning_config(VALIDATE_ARGS)
 
 	if (exec->bin_tiles_x == 0 ||
 	    exec->bin_tiles_y == 0) {
-		DRM_ERROR("Tile binning config of %dx%d too small\n",
+		DRM_DEBUG("Tile binning config of %dx%d too small\n",
 			  exec->bin_tiles_x, exec->bin_tiles_y);
 		return -EINVAL;
 	}
 
 	if (flags & (VC4_BIN_CONFIG_DB_NON_MS |
 		     VC4_BIN_CONFIG_TILE_BUFFER_64BIT)) {
-		DRM_ERROR("unsupported binning config flags 0x%02x\n", flags);
+		DRM_DEBUG("unsupported binning config flags 0x%02x\n", flags);
 		return -EINVAL;
 	}
 
@@ -493,20 +493,20 @@ vc4_validate_bin_cl(struct drm_device *dev,
 		const struct cmd_info *info;
 
 		if (cmd >= ARRAY_SIZE(cmd_info)) {
-			DRM_ERROR("0x%08x: packet %d out of bounds\n",
+			DRM_DEBUG("0x%08x: packet %d out of bounds\n",
 				  src_offset, cmd);
 			return -EINVAL;
 		}
 
 		info = &cmd_info[cmd];
 		if (!info->name) {
-			DRM_ERROR("0x%08x: packet %d invalid\n",
+			DRM_DEBUG("0x%08x: packet %d invalid\n",
 				  src_offset, cmd);
 			return -EINVAL;
 		}
 
 		if (src_offset + info->len > len) {
-			DRM_ERROR("0x%08x: packet %d (%s) length 0x%08x "
+			DRM_DEBUG("0x%08x: packet %d (%s) length 0x%08x "
 				  "exceeds bounds (0x%08x)\n",
 				  src_offset, cmd, info->name, info->len,
 				  src_offset + len);
@@ -519,7 +519,7 @@ vc4_validate_bin_cl(struct drm_device *dev,
 		if (info->func && info->func(exec,
 					     dst_pkt + 1,
 					     src_pkt + 1)) {
-			DRM_ERROR("0x%08x: packet %d (%s) failed to validate\n",
+			DRM_DEBUG("0x%08x: packet %d (%s) failed to validate\n",
 				  src_offset, cmd, info->name);
 			return -EINVAL;
 		}
@@ -537,7 +537,7 @@ vc4_validate_bin_cl(struct drm_device *dev,
 	exec->ct0ea = exec->ct0ca + dst_offset;
 
 	if (!exec->found_start_tile_binning_packet) {
-		DRM_ERROR("Bin CL missing VC4_PACKET_START_TILE_BINNING\n");
+		DRM_DEBUG("Bin CL missing VC4_PACKET_START_TILE_BINNING\n");
 		return -EINVAL;
 	}
 
@@ -549,7 +549,7 @@ vc4_validate_bin_cl(struct drm_device *dev,
 	 * semaphore increment.
 	 */
 	if (!exec->found_increment_semaphore_packet || !exec->found_flush) {
-		DRM_ERROR("Bin CL missing VC4_PACKET_INCREMENT_SEMAPHORE + "
+		DRM_DEBUG("Bin CL missing VC4_PACKET_INCREMENT_SEMAPHORE + "
 			  "VC4_PACKET_FLUSH\n");
 		return -EINVAL;
 	}
@@ -588,11 +588,11 @@ reloc_tex(struct vc4_exec_info *exec,
 		uint32_t remaining_size = tex->base.size - p0;
 
 		if (p0 > tex->base.size - 4) {
-			DRM_ERROR("UBO offset greater than UBO size\n");
+			DRM_DEBUG("UBO offset greater than UBO size\n");
 			goto fail;
 		}
 		if (p1 > remaining_size - 4) {
-			DRM_ERROR("UBO clamp would allow reads "
+			DRM_DEBUG("UBO clamp would allow reads "
 				  "outside of UBO\n");
 			goto fail;
 		}
@@ -612,14 +612,14 @@ reloc_tex(struct vc4_exec_info *exec,
 		if (VC4_GET_FIELD(p3, VC4_TEX_P2_PTYPE) ==
 		    VC4_TEX_P2_PTYPE_CUBE_MAP_STRIDE) {
 			if (cube_map_stride) {
-				DRM_ERROR("Cube map stride set twice\n");
+				DRM_DEBUG("Cube map stride set twice\n");
 				goto fail;
 			}
 
 			cube_map_stride = p3 & VC4_TEX_P2_CMST_MASK;
 		}
 		if (!cube_map_stride) {
-			DRM_ERROR("Cube map stride not set\n");
+			DRM_DEBUG("Cube map stride not set\n");
 			goto fail;
 		}
 	}
@@ -660,7 +660,7 @@ reloc_tex(struct vc4_exec_info *exec,
 	case VC4_TEXTURE_TYPE_RGBA64:
 	case VC4_TEXTURE_TYPE_YUV422R:
 	default:
-		DRM_ERROR("Texture format %d unsupported\n", type);
+		DRM_DEBUG("Texture format %d unsupported\n", type);
 		goto fail;
 	}
 	utile_w = utile_width(cpp);
@@ -713,7 +713,7 @@ reloc_tex(struct vc4_exec_info *exec,
 		level_size = aligned_width * cpp * aligned_height;
 
 		if (offset < level_size) {
-			DRM_ERROR("Level %d (%dx%d -> %dx%d) size %db "
+			DRM_DEBUG("Level %d (%dx%d -> %dx%d) size %db "
 				  "overflowed buffer bounds (offset %d)\n",
 				  i, level_width, level_height,
 				  aligned_width, aligned_height,
@@ -764,7 +764,7 @@ validate_gl_shader_rec(struct drm_device *dev,
 
 	nr_relocs = ARRAY_SIZE(shader_reloc_offsets) + nr_attributes;
 	if (nr_relocs * 4 > exec->shader_rec_size) {
-		DRM_ERROR("overflowed shader recs reading %d handles "
+		DRM_DEBUG("overflowed shader recs reading %d handles "
 			  "from %d bytes left\n",
 			  nr_relocs, exec->shader_rec_size);
 		return -EINVAL;
@@ -774,7 +774,7 @@ validate_gl_shader_rec(struct drm_device *dev,
 	exec->shader_rec_size -= nr_relocs * 4;
 
 	if (packet_size > exec->shader_rec_size) {
-		DRM_ERROR("overflowed shader recs copying %db packet "
+		DRM_DEBUG("overflowed shader recs copying %db packet "
 			  "from %d bytes left\n",
 			  packet_size, exec->shader_rec_size);
 		return -EINVAL;
@@ -794,7 +794,7 @@ validate_gl_shader_rec(struct drm_device *dev,
 
 	for (i = 0; i < shader_reloc_count; i++) {
 		if (src_handles[i] > exec->bo_count) {
-			DRM_ERROR("Shader handle %d too big\n", src_handles[i]);
+			DRM_DEBUG("Shader handle %d too big\n", src_handles[i]);
 			return -EINVAL;
 		}
 
@@ -810,13 +810,13 @@ validate_gl_shader_rec(struct drm_device *dev,
 
 	if (((*(uint16_t *)pkt_u & VC4_SHADER_FLAG_FS_SINGLE_THREAD) == 0) !=
 	    to_vc4_bo(&bo[0]->base)->validated_shader->is_threaded) {
-		DRM_ERROR("Thread mode of CL and FS do not match\n");
+		DRM_DEBUG("Thread mode of CL and FS do not match\n");
 		return -EINVAL;
 	}
 
 	if (to_vc4_bo(&bo[1]->base)->validated_shader->is_threaded ||
 	    to_vc4_bo(&bo[2]->base)->validated_shader->is_threaded) {
-		DRM_ERROR("cs and vs cannot be threaded\n");
+		DRM_DEBUG("cs and vs cannot be threaded\n");
 		return -EINVAL;
 	}
 
@@ -831,7 +831,7 @@ validate_gl_shader_rec(struct drm_device *dev,
 		*(uint32_t *)(pkt_v + o) = bo[i]->paddr + src_offset;
 
 		if (src_offset != 0) {
-			DRM_ERROR("Shaders must be at offset 0 of "
+			DRM_DEBUG("Shaders must be at offset 0 of "
 				  "the BO.\n");
 			return -EINVAL;
 		}
@@ -842,7 +842,7 @@ validate_gl_shader_rec(struct drm_device *dev,
 
 		if (validated_shader->uniforms_src_size >
 		    exec->uniforms_size) {
-			DRM_ERROR("Uniforms src buffer overflow\n");
+			DRM_DEBUG("Uniforms src buffer overflow\n");
 			return -EINVAL;
 		}
 
@@ -900,7 +900,7 @@ validate_gl_shader_rec(struct drm_device *dev,
 
 		if (vbo->base.size < offset ||
 		    vbo->base.size - offset < attr_size) {
-			DRM_ERROR("BO offset overflow (%d + %d > %zu)\n",
+			DRM_DEBUG("BO offset overflow (%d + %d > %zu)\n",
 				  offset, attr_size, vbo->base.size);
 			return -EINVAL;
 		}
@@ -909,7 +909,7 @@ validate_gl_shader_rec(struct drm_device *dev,
 			max_index = ((vbo->base.size - offset - attr_size) /
 				     stride);
 			if (state->max_index > max_index) {
-				DRM_ERROR("primitives use index %d out of "
+				DRM_DEBUG("primitives use index %d out of "
 					  "supplied %d\n",
 					  state->max_index, max_index);
 				return -EINVAL;
diff --git a/drivers/gpu/drm/vc4/vc4_validate_shaders.c b/drivers/gpu/drm/vc4/vc4_validate_shaders.c
index 0b2df5c6efb4..d3f15bf60900 100644
--- a/drivers/gpu/drm/vc4/vc4_validate_shaders.c
+++ b/drivers/gpu/drm/vc4/vc4_validate_shaders.c
@@ -200,7 +200,7 @@ check_tmu_write(struct vc4_validated_shader_info *validated_shader,
 		uint32_t clamp_reg, clamp_offset;
 
 		if (sig == QPU_SIG_SMALL_IMM) {
-			DRM_ERROR("direct TMU read used small immediate\n");
+			DRM_DEBUG("direct TMU read used small immediate\n");
 			return false;
 		}
 
@@ -209,7 +209,7 @@ check_tmu_write(struct vc4_validated_shader_info *validated_shader,
 		 */
 		if (is_mul ||
 		    QPU_GET_FIELD(inst, QPU_OP_ADD) != QPU_A_ADD) {
-			DRM_ERROR("direct TMU load wasn't an add\n");
+			DRM_DEBUG("direct TMU load wasn't an add\n");
 			return false;
 		}
 
@@ -220,13 +220,13 @@ check_tmu_write(struct vc4_validated_shader_info *validated_shader,
 		 */
 		clamp_reg = raddr_add_a_to_live_reg_index(inst);
 		if (clamp_reg == ~0) {
-			DRM_ERROR("direct TMU load wasn't clamped\n");
+			DRM_DEBUG("direct TMU load wasn't clamped\n");
 			return false;
 		}
 
 		clamp_offset = validation_state->live_min_clamp_offsets[clamp_reg];
 		if (clamp_offset == ~0) {
-			DRM_ERROR("direct TMU load wasn't clamped\n");
+			DRM_DEBUG("direct TMU load wasn't clamped\n");
 			return false;
 		}
 
@@ -238,7 +238,7 @@ check_tmu_write(struct vc4_validated_shader_info *validated_shader,
 
 		if (!(add_b == QPU_MUX_A && raddr_a == QPU_R_UNIF) &&
 		    !(add_b == QPU_MUX_B && raddr_b == QPU_R_UNIF)) {
-			DRM_ERROR("direct TMU load didn't add to a uniform\n");
+			DRM_DEBUG("direct TMU load didn't add to a uniform\n");
 			return false;
 		}
 
@@ -246,14 +246,14 @@ check_tmu_write(struct vc4_validated_shader_info *validated_shader,
 	} else {
 		if (raddr_a == QPU_R_UNIF || (sig != QPU_SIG_SMALL_IMM &&
 					      raddr_b == QPU_R_UNIF)) {
-			DRM_ERROR("uniform read in the same instruction as "
+			DRM_DEBUG("uniform read in the same instruction as "
 				  "texture setup.\n");
 			return false;
 		}
 	}
 
 	if (validation_state->tmu_write_count[tmu] >= 4) {
-		DRM_ERROR("TMU%d got too many parameters before dispatch\n",
+		DRM_DEBUG("TMU%d got too many parameters before dispatch\n",
 			  tmu);
 		return false;
 	}
@@ -265,7 +265,7 @@ check_tmu_write(struct vc4_validated_shader_info *validated_shader,
 	 */
 	if (!is_direct) {
 		if (validation_state->needs_uniform_address_update) {
-			DRM_ERROR("Texturing with undefined uniform address\n");
+			DRM_DEBUG("Texturing with undefined uniform address\n");
 			return false;
 		}
 
@@ -336,35 +336,35 @@ validate_uniform_address_write(struct vc4_validated_shader_info *validated_shade
 	case QPU_SIG_LOAD_TMU1:
 		break;
 	default:
-		DRM_ERROR("uniforms address change must be "
+		DRM_DEBUG("uniforms address change must be "
 			  "normal math\n");
 		return false;
 	}
 
 	if (is_mul || QPU_GET_FIELD(inst, QPU_OP_ADD) != QPU_A_ADD) {
-		DRM_ERROR("Uniform address reset must be an ADD.\n");
+		DRM_DEBUG("Uniform address reset must be an ADD.\n");
 		return false;
 	}
 
 	if (QPU_GET_FIELD(inst, QPU_COND_ADD) != QPU_COND_ALWAYS) {
-		DRM_ERROR("Uniform address reset must be unconditional.\n");
+		DRM_DEBUG("Uniform address reset must be unconditional.\n");
 		return false;
 	}
 
 	if (QPU_GET_FIELD(inst, QPU_PACK) != QPU_PACK_A_NOP &&
 	    !(inst & QPU_PM)) {
-		DRM_ERROR("No packing allowed on uniforms reset\n");
+		DRM_DEBUG("No packing allowed on uniforms reset\n");
 		return false;
 	}
 
 	if (add_lri == -1) {
-		DRM_ERROR("First argument of uniform address write must be "
+		DRM_DEBUG("First argument of uniform address write must be "
 			  "an immediate value.\n");
 		return false;
 	}
 
 	if (validation_state->live_immediates[add_lri] != expected_offset) {
-		DRM_ERROR("Resetting uniforms with offset %db instead of %db\n",
+		DRM_DEBUG("Resetting uniforms with offset %db instead of %db\n",
 			  validation_state->live_immediates[add_lri],
 			  expected_offset);
 		return false;
@@ -372,7 +372,7 @@ validate_uniform_address_write(struct vc4_validated_shader_info *validated_shade
 
 	if (!(add_b == QPU_MUX_A && raddr_a == QPU_R_UNIF) &&
 	    !(add_b == QPU_MUX_B && raddr_b == QPU_R_UNIF)) {
-		DRM_ERROR("Second argument of uniform address write must be "
+		DRM_DEBUG("Second argument of uniform address write must be "
 			  "a uniform.\n");
 		return false;
 	}
@@ -417,7 +417,7 @@ check_reg_write(struct vc4_validated_shader_info *validated_shader,
 	switch (waddr) {
 	case QPU_W_UNIFORMS_ADDRESS:
 		if (is_b) {
-			DRM_ERROR("relative uniforms address change "
+			DRM_DEBUG("relative uniforms address change "
 				  "unsupported\n");
 			return false;
 		}
@@ -452,11 +452,11 @@ check_reg_write(struct vc4_validated_shader_info *validated_shader,
 		/* XXX: I haven't thought about these, so don't support them
 		 * for now.
 		 */
-		DRM_ERROR("Unsupported waddr %d\n", waddr);
+		DRM_DEBUG("Unsupported waddr %d\n", waddr);
 		return false;
 
 	case QPU_W_VPM_ADDR:
-		DRM_ERROR("General VPM DMA unsupported\n");
+		DRM_DEBUG("General VPM DMA unsupported\n");
 		return false;
 
 	case QPU_W_VPM:
@@ -559,7 +559,7 @@ check_instruction_writes(struct vc4_validated_shader_info *validated_shader,
 	bool ok;
 
 	if (is_tmu_write(waddr_add) && is_tmu_write(waddr_mul)) {
-		DRM_ERROR("ADD and MUL both set up textures\n");
+		DRM_DEBUG("ADD and MUL both set up textures\n");
 		return false;
 	}
 
@@ -588,7 +588,7 @@ check_branch(uint64_t inst,
 	 * there's no need for it.
 	 */
 	if (waddr_add != QPU_W_NOP || waddr_mul != QPU_W_NOP) {
-		DRM_ERROR("branch instruction at %d wrote a register.\n",
+		DRM_DEBUG("branch instruction at %d wrote a register.\n",
 			  validation_state->ip);
 		return false;
 	}
@@ -614,7 +614,7 @@ check_instruction_reads(struct vc4_validated_shader_info *validated_shader,
 		validated_shader->uniforms_size += 4;
 
 		if (validation_state->needs_uniform_address_update) {
-			DRM_ERROR("Uniform read with undefined uniform "
+			DRM_DEBUG("Uniform read with undefined uniform "
 				  "address\n");
 			return false;
 		}
@@ -660,19 +660,19 @@ vc4_validate_branches(struct vc4_shader_validation_state *validation_state)
 			continue;
 
 		if (ip - last_branch < 4) {
-			DRM_ERROR("Branch at %d during delay slots\n", ip);
+			DRM_DEBUG("Branch at %d during delay slots\n", ip);
 			return false;
 		}
 		last_branch = ip;
 
 		if (inst & QPU_BRANCH_REG) {
-			DRM_ERROR("branching from register relative "
+			DRM_DEBUG("branching from register relative "
 				  "not supported\n");
 			return false;
 		}
 
 		if (!(inst & QPU_BRANCH_REL)) {
-			DRM_ERROR("relative branching required\n");
+			DRM_DEBUG("relative branching required\n");
 			return false;
 		}
 
@@ -682,13 +682,13 @@ vc4_validate_branches(struct vc4_shader_validation_state *validation_state)
 		 * end of the shader object.
 		 */
 		if (branch_imm % sizeof(inst) != 0) {
-			DRM_ERROR("branch target not aligned\n");
+			DRM_DEBUG("branch target not aligned\n");
 			return false;
 		}
 
 		branch_target_ip = after_delay_ip + (branch_imm >> 3);
 		if (branch_target_ip >= validation_state->max_ip) {
-			DRM_ERROR("Branch at %d outside of shader (ip %d/%d)\n",
+			DRM_DEBUG("Branch at %d outside of shader (ip %d/%d)\n",
 				  ip, branch_target_ip,
 				  validation_state->max_ip);
 			return false;
@@ -699,7 +699,7 @@ vc4_validate_branches(struct vc4_shader_validation_state *validation_state)
 		 * the shader.
 		 */
 		if (after_delay_ip >= validation_state->max_ip) {
-			DRM_ERROR("Branch at %d continues past shader end "
+			DRM_DEBUG("Branch at %d continues past shader end "
 				  "(%d/%d)\n",
 				  ip, after_delay_ip, validation_state->max_ip);
 			return false;
@@ -709,7 +709,7 @@ vc4_validate_branches(struct vc4_shader_validation_state *validation_state)
 	}
 
 	if (max_branch_target > validation_state->max_ip - 3) {
-		DRM_ERROR("Branch landed after QPU_SIG_PROG_END");
+		DRM_DEBUG("Branch landed after QPU_SIG_PROG_END");
 		return false;
 	}
 
@@ -750,7 +750,7 @@ vc4_handle_branch_target(struct vc4_shader_validation_state *validation_state)
 		return true;
 
 	if (texturing_in_progress(validation_state)) {
-		DRM_ERROR("Branch target landed during TMU setup\n");
+		DRM_DEBUG("Branch target landed during TMU setup\n");
 		return false;
 	}
 
@@ -837,7 +837,7 @@ vc4_validate_shader(struct drm_gem_cma_object *shader_obj)
 		case QPU_SIG_LAST_THREAD_SWITCH:
 			if (!check_instruction_writes(validated_shader,
 						      &validation_state)) {
-				DRM_ERROR("Bad write at ip %d\n", ip);
+				DRM_DEBUG("Bad write at ip %d\n", ip);
 				goto fail;
 			}
 
@@ -855,7 +855,7 @@ vc4_validate_shader(struct drm_gem_cma_object *shader_obj)
 				validated_shader->is_threaded = true;
 
 				if (ip < last_thread_switch_ip + 3) {
-					DRM_ERROR("Thread switch too soon after "
+					DRM_DEBUG("Thread switch too soon after "
 						  "last switch at ip %d\n", ip);
 					goto fail;
 				}
@@ -867,7 +867,7 @@ vc4_validate_shader(struct drm_gem_cma_object *shader_obj)
 		case QPU_SIG_LOAD_IMM:
 			if (!check_instruction_writes(validated_shader,
 						      &validation_state)) {
-				DRM_ERROR("Bad LOAD_IMM write at ip %d\n", ip);
+				DRM_DEBUG("Bad LOAD_IMM write at ip %d\n", ip);
 				goto fail;
 			}
 			break;
@@ -878,14 +878,14 @@ vc4_validate_shader(struct drm_gem_cma_object *shader_obj)
 				goto fail;
 
 			if (ip < last_thread_switch_ip + 3) {
-				DRM_ERROR("Branch in thread switch at ip %d",
+				DRM_DEBUG("Branch in thread switch at ip %d",
 					  ip);
 				goto fail;
 			}
 
 			break;
 		default:
-			DRM_ERROR("Unsupported QPU signal %d at "
+			DRM_DEBUG("Unsupported QPU signal %d at "
 				  "instruction %d\n", sig, ip);
 			goto fail;
 		}
@@ -898,7 +898,7 @@ vc4_validate_shader(struct drm_gem_cma_object *shader_obj)
 	}
 
 	if (ip == validation_state.max_ip) {
-		DRM_ERROR("shader failed to terminate before "
+		DRM_DEBUG("shader failed to terminate before "
 			  "shader BO end at %zd\n",
 			  shader_obj->base.size);
 		goto fail;
@@ -907,7 +907,7 @@ vc4_validate_shader(struct drm_gem_cma_object *shader_obj)
 	/* Might corrupt other thread */
 	if (validated_shader->is_threaded &&
 	    validation_state.all_registers_used) {
-		DRM_ERROR("Shader uses threading, but uses the upper "
+		DRM_DEBUG("Shader uses threading, but uses the upper "
 			  "half of the registers, too\n");
 		goto fail;
 	}
diff --git a/drivers/gpu/drm/vgem/vgem_drv.c b/drivers/gpu/drm/vgem/vgem_drv.c
index 12289673f457..2524ff116f00 100644
--- a/drivers/gpu/drm/vgem/vgem_drv.c
+++ b/drivers/gpu/drm/vgem/vgem_drv.c
@@ -190,7 +190,7 @@ static struct drm_gem_object *vgem_gem_create(struct drm_device *dev,
 		return ERR_CAST(obj);
 
 	ret = drm_gem_handle_create(file, &obj->base, handle);
-	drm_gem_object_unreference_unlocked(&obj->base);
+	drm_gem_object_put_unlocked(&obj->base);
 	if (ret)
 		goto err;
 
@@ -245,7 +245,7 @@ static int vgem_gem_dumb_map(struct drm_file *file, struct drm_device *dev,
 
 	*offset = drm_vma_node_offset_addr(&obj->vma_node);
 unref:
-	drm_gem_object_unreference_unlocked(obj);
+	drm_gem_object_put_unlocked(obj);
 
 	return ret;
 }
diff --git a/drivers/gpu/drm/vgem/vgem_fence.c b/drivers/gpu/drm/vgem/vgem_fence.c
index 3109c8308eb5..8fd52f211e9d 100644
--- a/drivers/gpu/drm/vgem/vgem_fence.c
+++ b/drivers/gpu/drm/vgem/vgem_fence.c
@@ -213,7 +213,7 @@ err_fence:
 		dma_fence_put(fence);
 	}
 err:
-	drm_gem_object_unreference_unlocked(obj);
+	drm_gem_object_put_unlocked(obj);
 	return ret;
 }
 
diff --git a/drivers/gpu/drm/virtio/virtgpu_drv.c b/drivers/gpu/drm/virtio/virtgpu_drv.c
index 63d35c7e416c..49a3d8d5a249 100644
--- a/drivers/gpu/drm/virtio/virtgpu_drv.c
+++ b/drivers/gpu/drm/virtio/virtgpu_drv.c
@@ -122,7 +122,6 @@ static struct drm_driver driver = {
 
 	.dumb_create = virtio_gpu_mode_dumb_create,
 	.dumb_map_offset = virtio_gpu_mode_dumb_mmap,
-	.dumb_destroy = virtio_gpu_mode_dumb_destroy,
 
 #if defined(CONFIG_DEBUG_FS)
 	.debugfs_init = virtio_gpu_debugfs_init,
diff --git a/drivers/gpu/drm/virtio/virtgpu_drv.h b/drivers/gpu/drm/virtio/virtgpu_drv.h
index 3a66abb8fd50..da2fb585fea4 100644
--- a/drivers/gpu/drm/virtio/virtgpu_drv.h
+++ b/drivers/gpu/drm/virtio/virtgpu_drv.h
@@ -236,9 +236,6 @@ struct virtio_gpu_object *virtio_gpu_alloc_object(struct drm_device *dev,
 int virtio_gpu_mode_dumb_create(struct drm_file *file_priv,
 				struct drm_device *dev,
 				struct drm_mode_create_dumb *args);
-int virtio_gpu_mode_dumb_destroy(struct drm_file *file_priv,
-				 struct drm_device *dev,
-				 uint32_t handle);
 int virtio_gpu_mode_dumb_mmap(struct drm_file *file_priv,
 			      struct drm_device *dev,
 			      uint32_t handle, uint64_t *offset_p);
diff --git a/drivers/gpu/drm/virtio/virtgpu_fb.c b/drivers/gpu/drm/virtio/virtgpu_fb.c
index 046e28b69d99..15d18fd0c64b 100644
--- a/drivers/gpu/drm/virtio/virtgpu_fb.c
+++ b/drivers/gpu/drm/virtio/virtgpu_fb.c
@@ -308,7 +308,7 @@ static int virtio_gpu_fbdev_destroy(struct drm_device *dev,
 
 	return 0;
 }
-static struct drm_fb_helper_funcs virtio_gpu_fb_helper_funcs = {
+static const struct drm_fb_helper_funcs virtio_gpu_fb_helper_funcs = {
 	.fb_probe = virtio_gpufb_create,
 };
 
diff --git a/drivers/gpu/drm/virtio/virtgpu_gem.c b/drivers/gpu/drm/virtio/virtgpu_gem.c
index cc025d8fbe19..72ad7b103448 100644
--- a/drivers/gpu/drm/virtio/virtgpu_gem.c
+++ b/drivers/gpu/drm/virtio/virtgpu_gem.c
@@ -118,13 +118,6 @@ fail:
 	return ret;
 }
 
-int virtio_gpu_mode_dumb_destroy(struct drm_file *file_priv,
-				 struct drm_device *dev,
-				 uint32_t handle)
-{
-	return drm_gem_handle_delete(file_priv, handle);
-}
-
 int virtio_gpu_mode_dumb_mmap(struct drm_file *file_priv,
 			      struct drm_device *dev,
 			      uint32_t handle, uint64_t *offset_p)
diff --git a/drivers/gpu/drm/virtio/virtgpu_ttm.c b/drivers/gpu/drm/virtio/virtgpu_ttm.c
index e695d74eaa9f..cd389c5eaef5 100644
--- a/drivers/gpu/drm/virtio/virtgpu_ttm.c
+++ b/drivers/gpu/drm/virtio/virtgpu_ttm.c
@@ -192,7 +192,7 @@ static int ttm_bo_man_takedown(struct ttm_mem_type_manager *man)
 }
 
 static void ttm_bo_man_debug(struct ttm_mem_type_manager *man,
-			     const char *prefix)
+			     struct drm_printer *printer)
 {
 }
 
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_buffer.c b/drivers/gpu/drm/vmwgfx/vmwgfx_buffer.c
index 35bf781e418e..c7056322211c 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_buffer.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_buffer.c
@@ -30,49 +30,49 @@
 #include <drm/ttm/ttm_placement.h>
 #include <drm/ttm/ttm_page_alloc.h>
 
-static struct ttm_place vram_placement_flags = {
+static const struct ttm_place vram_placement_flags = {
 	.fpfn = 0,
 	.lpfn = 0,
 	.flags = TTM_PL_FLAG_VRAM | TTM_PL_FLAG_CACHED
 };
 
-static struct ttm_place vram_ne_placement_flags = {
+static const struct ttm_place vram_ne_placement_flags = {
 	.fpfn = 0,
 	.lpfn = 0,
 	.flags = TTM_PL_FLAG_VRAM | TTM_PL_FLAG_CACHED | TTM_PL_FLAG_NO_EVICT
 };
 
-static struct ttm_place sys_placement_flags = {
+static const struct ttm_place sys_placement_flags = {
 	.fpfn = 0,
 	.lpfn = 0,
 	.flags = TTM_PL_FLAG_SYSTEM | TTM_PL_FLAG_CACHED
 };
 
-static struct ttm_place sys_ne_placement_flags = {
+static const struct ttm_place sys_ne_placement_flags = {
 	.fpfn = 0,
 	.lpfn = 0,
 	.flags = TTM_PL_FLAG_SYSTEM | TTM_PL_FLAG_CACHED | TTM_PL_FLAG_NO_EVICT
 };
 
-static struct ttm_place gmr_placement_flags = {
+static const struct ttm_place gmr_placement_flags = {
 	.fpfn = 0,
 	.lpfn = 0,
 	.flags = VMW_PL_FLAG_GMR | TTM_PL_FLAG_CACHED
 };
 
-static struct ttm_place gmr_ne_placement_flags = {
+static const struct ttm_place gmr_ne_placement_flags = {
 	.fpfn = 0,
 	.lpfn = 0,
 	.flags = VMW_PL_FLAG_GMR | TTM_PL_FLAG_CACHED | TTM_PL_FLAG_NO_EVICT
 };
 
-static struct ttm_place mob_placement_flags = {
+static const struct ttm_place mob_placement_flags = {
 	.fpfn = 0,
 	.lpfn = 0,
 	.flags = VMW_PL_FLAG_MOB | TTM_PL_FLAG_CACHED
 };
 
-static struct ttm_place mob_ne_placement_flags = {
+static const struct ttm_place mob_ne_placement_flags = {
 	.fpfn = 0,
 	.lpfn = 0,
 	.flags = VMW_PL_FLAG_MOB | TTM_PL_FLAG_CACHED | TTM_PL_FLAG_NO_EVICT
@@ -85,7 +85,7 @@ struct ttm_placement vmw_vram_placement = {
 	.busy_placement = &vram_placement_flags
 };
 
-static struct ttm_place vram_gmr_placement_flags[] = {
+static const struct ttm_place vram_gmr_placement_flags[] = {
 	{
 		.fpfn = 0,
 		.lpfn = 0,
@@ -97,7 +97,7 @@ static struct ttm_place vram_gmr_placement_flags[] = {
 	}
 };
 
-static struct ttm_place gmr_vram_placement_flags[] = {
+static const struct ttm_place gmr_vram_placement_flags[] = {
 	{
 		.fpfn = 0,
 		.lpfn = 0,
@@ -116,7 +116,7 @@ struct ttm_placement vmw_vram_gmr_placement = {
 	.busy_placement = &gmr_placement_flags
 };
 
-static struct ttm_place vram_gmr_ne_placement_flags[] = {
+static const struct ttm_place vram_gmr_ne_placement_flags[] = {
 	{
 		.fpfn = 0,
 		.lpfn = 0,
@@ -165,7 +165,7 @@ struct ttm_placement vmw_sys_ne_placement = {
 	.busy_placement = &sys_ne_placement_flags
 };
 
-static struct ttm_place evictable_placement_flags[] = {
+static const struct ttm_place evictable_placement_flags[] = {
 	{
 		.fpfn = 0,
 		.lpfn = 0,
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_cmdbuf.c b/drivers/gpu/drm/vmwgfx/vmwgfx_cmdbuf.c
index 99a7f4ab7d97..c706ad30411b 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_cmdbuf.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_cmdbuf.c
@@ -51,6 +51,7 @@ struct vmw_cmdbuf_context {
 	struct list_head hw_submitted;
 	struct list_head preempted;
 	unsigned num_hw_submitted;
+	bool block_submission;
 };
 
 /**
@@ -60,6 +61,9 @@ struct vmw_cmdbuf_context {
  * kernel command submissions, @cur.
  * @space_mutex: Mutex to protect against starvation when we allocate
  * main pool buffer space.
+ * @error_mutex: Mutex to serialize the work queue error handling.
+ * Note this is not needed if the same workqueue handler
+ * can't race with itself...
  * @work: A struct work_struct implementeing command buffer error handling.
  * Immutable.
  * @dev_priv: Pointer to the device private struct. Immutable.
@@ -85,7 +89,6 @@ struct vmw_cmdbuf_context {
  * Internal protection.
  * @dheaders: Pool of DMA memory for device command buffer headers with trailing
  * space for inline data. Internal protection.
- * @tasklet: Tasklet struct for irq processing. Immutable.
  * @alloc_queue: Wait queue for processes waiting to allocate command buffer
  * space.
  * @idle_queue: Wait queue for processes waiting for command buffer idle.
@@ -102,6 +105,7 @@ struct vmw_cmdbuf_context {
 struct vmw_cmdbuf_man {
 	struct mutex cur_mutex;
 	struct mutex space_mutex;
+	struct mutex error_mutex;
 	struct work_struct work;
 	struct vmw_private *dev_priv;
 	struct vmw_cmdbuf_context ctx[SVGA_CB_CONTEXT_MAX];
@@ -117,7 +121,6 @@ struct vmw_cmdbuf_man {
 	spinlock_t lock;
 	struct dma_pool *headers;
 	struct dma_pool *dheaders;
-	struct tasklet_struct tasklet;
 	wait_queue_head_t alloc_queue;
 	wait_queue_head_t idle_queue;
 	bool irq_on;
@@ -181,12 +184,13 @@ struct vmw_cmdbuf_alloc_info {
 };
 
 /* Loop over each context in the command buffer manager. */
-#define for_each_cmdbuf_ctx(_man, _i, _ctx) \
+#define for_each_cmdbuf_ctx(_man, _i, _ctx)				\
 	for (_i = 0, _ctx = &(_man)->ctx[0]; (_i) < SVGA_CB_CONTEXT_MAX; \
 	     ++(_i), ++(_ctx))
 
-static int vmw_cmdbuf_startstop(struct vmw_cmdbuf_man *man, bool enable);
-
+static int vmw_cmdbuf_startstop(struct vmw_cmdbuf_man *man, u32 context,
+				bool enable);
+static int vmw_cmdbuf_preempt(struct vmw_cmdbuf_man *man, u32 context);
 
 /**
  * vmw_cmdbuf_cur_lock - Helper to lock the cur_mutex.
@@ -278,9 +282,9 @@ void vmw_cmdbuf_header_free(struct vmw_cmdbuf_header *header)
 		vmw_cmdbuf_header_inline_free(header);
 		return;
 	}
-	spin_lock_bh(&man->lock);
+	spin_lock(&man->lock);
 	__vmw_cmdbuf_header_free(header);
-	spin_unlock_bh(&man->lock);
+	spin_unlock(&man->lock);
 }
 
 
@@ -331,7 +335,8 @@ static void vmw_cmdbuf_ctx_submit(struct vmw_cmdbuf_man *man,
 				  struct vmw_cmdbuf_context *ctx)
 {
 	while (ctx->num_hw_submitted < man->max_hw_submitted &&
-	      !list_empty(&ctx->submitted)) {
+	       !list_empty(&ctx->submitted) &&
+	       !ctx->block_submission) {
 		struct vmw_cmdbuf_header *entry;
 		SVGACBStatus status;
 
@@ -386,12 +391,17 @@ static void vmw_cmdbuf_ctx_process(struct vmw_cmdbuf_man *man,
 			__vmw_cmdbuf_header_free(entry);
 			break;
 		case SVGA_CB_STATUS_COMMAND_ERROR:
-		case SVGA_CB_STATUS_CB_HEADER_ERROR:
+			entry->cb_header->status = SVGA_CB_STATUS_NONE;
 			list_add_tail(&entry->list, &man->error);
 			schedule_work(&man->work);
 			break;
 		case SVGA_CB_STATUS_PREEMPTED:
-			list_add(&entry->list, &ctx->preempted);
+			entry->cb_header->status = SVGA_CB_STATUS_NONE;
+			list_add_tail(&entry->list, &ctx->preempted);
+			break;
+		case SVGA_CB_STATUS_CB_HEADER_ERROR:
+			WARN_ONCE(true, "Command buffer header error.\n");
+			__vmw_cmdbuf_header_free(entry);
 			break;
 		default:
 			WARN_ONCE(true, "Undefined command buffer status.\n");
@@ -468,20 +478,17 @@ static void vmw_cmdbuf_ctx_add(struct vmw_cmdbuf_man *man,
 }
 
 /**
- * vmw_cmdbuf_man_tasklet - The main part of the command buffer interrupt
- * handler implemented as a tasklet.
+ * vmw_cmdbuf_irqthread - The main part of the command buffer interrupt
+ * handler implemented as a threaded irq task.
  *
- * @data: Tasklet closure. A pointer to the command buffer manager cast to
- * an unsigned long.
+ * @man: Pointer to the command buffer manager.
  *
- * The bottom half (tasklet) of the interrupt handler simply calls into the
+ * The bottom half of the interrupt handler simply calls into the
  * command buffer processor to free finished buffers and submit any
  * queued buffers to hardware.
  */
-static void vmw_cmdbuf_man_tasklet(unsigned long data)
+void vmw_cmdbuf_irqthread(struct vmw_cmdbuf_man *man)
 {
-	struct vmw_cmdbuf_man *man = (struct vmw_cmdbuf_man *) data;
-
 	spin_lock(&man->lock);
 	vmw_cmdbuf_man_process(man);
 	spin_unlock(&man->lock);
@@ -502,24 +509,112 @@ static void vmw_cmdbuf_work_func(struct work_struct *work)
 		container_of(work, struct vmw_cmdbuf_man, work);
 	struct vmw_cmdbuf_header *entry, *next;
 	uint32_t dummy;
-	bool restart = false;
+	bool restart[SVGA_CB_CONTEXT_MAX];
+	bool send_fence = false;
+	struct list_head restart_head[SVGA_CB_CONTEXT_MAX];
+	int i;
+	struct vmw_cmdbuf_context *ctx;
 
-	spin_lock_bh(&man->lock);
+	for_each_cmdbuf_ctx(man, i, ctx) {
+		INIT_LIST_HEAD(&restart_head[i]);
+		restart[i] = false;
+	}
+
+	mutex_lock(&man->error_mutex);
+	spin_lock(&man->lock);
 	list_for_each_entry_safe(entry, next, &man->error, list) {
-		restart = true;
-		DRM_ERROR("Command buffer error.\n");
+		SVGACBHeader *cb_hdr = entry->cb_header;
+		SVGA3dCmdHeader *header = (SVGA3dCmdHeader *)
+			(entry->cmd + cb_hdr->errorOffset);
+		u32 error_cmd_size, new_start_offset;
+		const char *cmd_name;
+
+		list_del_init(&entry->list);
+		restart[entry->cb_context] = true;
+
+		if (!vmw_cmd_describe(header, &error_cmd_size, &cmd_name)) {
+			DRM_ERROR("Unknown command causing device error.\n");
+			DRM_ERROR("Command buffer offset is %lu\n",
+				  (unsigned long) cb_hdr->errorOffset);
+			__vmw_cmdbuf_header_free(entry);
+			send_fence = true;
+			continue;
+		}
 
-		list_del(&entry->list);
-		__vmw_cmdbuf_header_free(entry);
-		wake_up_all(&man->idle_queue);
+		DRM_ERROR("Command \"%s\" causing device error.\n", cmd_name);
+		DRM_ERROR("Command buffer offset is %lu\n",
+			  (unsigned long) cb_hdr->errorOffset);
+		DRM_ERROR("Command size is %lu\n",
+			  (unsigned long) error_cmd_size);
+
+		new_start_offset = cb_hdr->errorOffset + error_cmd_size;
+
+		if (new_start_offset >= cb_hdr->length) {
+			__vmw_cmdbuf_header_free(entry);
+			send_fence = true;
+			continue;
+		}
+
+		if (man->using_mob)
+			cb_hdr->ptr.mob.mobOffset += new_start_offset;
+		else
+			cb_hdr->ptr.pa += (u64) new_start_offset;
+
+		entry->cmd += new_start_offset;
+		cb_hdr->length -= new_start_offset;
+		cb_hdr->errorOffset = 0;
+		cb_hdr->offset = 0;
+		list_add_tail(&entry->list, &restart_head[entry->cb_context]);
+		man->ctx[entry->cb_context].block_submission = true;
+	}
+	spin_unlock(&man->lock);
+
+	/* Preempt all contexts with errors */
+	for_each_cmdbuf_ctx(man, i, ctx) {
+		if (ctx->block_submission && vmw_cmdbuf_preempt(man, i))
+			DRM_ERROR("Failed preempting command buffer "
+				  "context %u.\n", i);
+	}
+
+	spin_lock(&man->lock);
+	for_each_cmdbuf_ctx(man, i, ctx) {
+		if (!ctx->block_submission)
+			continue;
+
+		/* Move preempted command buffers to the preempted queue. */
+		vmw_cmdbuf_ctx_process(man, ctx, &dummy);
+
+		/*
+		 * Add the preempted queue after the command buffer
+		 * that caused an error.
+		 */
+		list_splice_init(&ctx->preempted, restart_head[i].prev);
+
+		/*
+		 * Finally add all command buffers first in the submitted
+		 * queue, to rerun them.
+		 */
+		list_splice_init(&restart_head[i], &ctx->submitted);
+
+		ctx->block_submission = false;
 	}
-	spin_unlock_bh(&man->lock);
 
-	if (restart && vmw_cmdbuf_startstop(man, true))
-		DRM_ERROR("Failed restarting command buffer context 0.\n");
+	vmw_cmdbuf_man_process(man);
+	spin_unlock(&man->lock);
+
+	for_each_cmdbuf_ctx(man, i, ctx) {
+		if (restart[i] && vmw_cmdbuf_startstop(man, i, true))
+			DRM_ERROR("Failed restarting command buffer "
+				  "context %u.\n", i);
+	}
 
 	/* Send a new fence in case one was removed */
-	vmw_fifo_send_fence(man->dev_priv, &dummy);
+	if (send_fence) {
+		vmw_fifo_send_fence(man->dev_priv, &dummy);
+		wake_up_all(&man->idle_queue);
+	}
+
+	mutex_unlock(&man->error_mutex);
 }
 
 /**
@@ -536,7 +631,7 @@ static bool vmw_cmdbuf_man_idle(struct vmw_cmdbuf_man *man,
 	bool idle = false;
 	int i;
 
-	spin_lock_bh(&man->lock);
+	spin_lock(&man->lock);
 	vmw_cmdbuf_man_process(man);
 	for_each_cmdbuf_ctx(man, i, ctx) {
 		if (!list_empty(&ctx->submitted) ||
@@ -548,7 +643,7 @@ static bool vmw_cmdbuf_man_idle(struct vmw_cmdbuf_man *man,
 	idle = list_empty(&man->error);
 
 out_unlock:
-	spin_unlock_bh(&man->lock);
+	spin_unlock(&man->lock);
 
 	return idle;
 }
@@ -571,7 +666,7 @@ static void __vmw_cmdbuf_cur_flush(struct vmw_cmdbuf_man *man)
 	if (!cur)
 		return;
 
-	spin_lock_bh(&man->lock);
+	spin_lock(&man->lock);
 	if (man->cur_pos == 0) {
 		__vmw_cmdbuf_header_free(cur);
 		goto out_unlock;
@@ -580,7 +675,7 @@ static void __vmw_cmdbuf_cur_flush(struct vmw_cmdbuf_man *man)
 	man->cur->cb_header->length = man->cur_pos;
 	vmw_cmdbuf_ctx_add(man, man->cur, SVGA_CB_CONTEXT_0);
 out_unlock:
-	spin_unlock_bh(&man->lock);
+	spin_unlock(&man->lock);
 	man->cur = NULL;
 	man->cur_pos = 0;
 }
@@ -673,14 +768,14 @@ static bool vmw_cmdbuf_try_alloc(struct vmw_cmdbuf_man *man,
 		return true;
  
 	memset(info->node, 0, sizeof(*info->node));
-	spin_lock_bh(&man->lock);
+	spin_lock(&man->lock);
 	ret = drm_mm_insert_node(&man->mm, info->node, info->page_size);
 	if (ret) {
 		vmw_cmdbuf_man_process(man);
 		ret = drm_mm_insert_node(&man->mm, info->node, info->page_size);
 	}
 
-	spin_unlock_bh(&man->lock);
+	spin_unlock(&man->lock);
 	info->done = !ret;
 
 	return info->done;
@@ -779,8 +874,8 @@ static int vmw_cmdbuf_space_pool(struct vmw_cmdbuf_man *man,
 	if (ret)
 		return ret;
 
-	header->cb_header = dma_pool_alloc(man->headers, GFP_KERNEL,
-					   &header->handle);
+	header->cb_header = dma_pool_zalloc(man->headers, GFP_KERNEL,
+					    &header->handle);
 	if (!header->cb_header) {
 		ret = -ENOMEM;
 		goto out_no_cb_header;
@@ -790,7 +885,6 @@ static int vmw_cmdbuf_space_pool(struct vmw_cmdbuf_man *man,
 	cb_hdr = header->cb_header;
 	offset = header->node.start << PAGE_SHIFT;
 	header->cmd = man->map + offset;
-	memset(cb_hdr, 0, sizeof(*cb_hdr));
 	if (man->using_mob) {
 		cb_hdr->flags = SVGA_CB_FLAG_MOB;
 		cb_hdr->ptr.mob.mobid = man->cmd_space->mem.start;
@@ -802,9 +896,9 @@ static int vmw_cmdbuf_space_pool(struct vmw_cmdbuf_man *man,
 	return 0;
 
 out_no_cb_header:
-	spin_lock_bh(&man->lock);
+	spin_lock(&man->lock);
 	drm_mm_remove_node(&header->node);
-	spin_unlock_bh(&man->lock);
+	spin_unlock(&man->lock);
 
 	return ret;
 }
@@ -827,8 +921,8 @@ static int vmw_cmdbuf_space_inline(struct vmw_cmdbuf_man *man,
 	if (WARN_ON_ONCE(size > VMW_CMDBUF_INLINE_SIZE))
 		return -ENOMEM;
 
-	dheader = dma_pool_alloc(man->dheaders, GFP_KERNEL,
-				 &header->handle);
+	dheader = dma_pool_zalloc(man->dheaders, GFP_KERNEL,
+				  &header->handle);
 	if (!dheader)
 		return -ENOMEM;
 
@@ -837,7 +931,6 @@ static int vmw_cmdbuf_space_inline(struct vmw_cmdbuf_man *man,
 	cb_hdr = &dheader->cb_header;
 	header->cb_header = cb_hdr;
 	header->cmd = dheader->cmd;
-	memset(dheader, 0, sizeof(*dheader));
 	cb_hdr->status = SVGA_CB_STATUS_NONE;
 	cb_hdr->flags = SVGA_CB_FLAG_NONE;
 	cb_hdr->ptr.pa = (u64)header->handle +
@@ -1025,18 +1118,6 @@ void vmw_cmdbuf_commit(struct vmw_cmdbuf_man *man, size_t size,
 	vmw_cmdbuf_cur_unlock(man);
 }
 
-/**
- * vmw_cmdbuf_tasklet_schedule - Schedule the interrupt handler bottom half.
- *
- * @man: The command buffer manager.
- */
-void vmw_cmdbuf_tasklet_schedule(struct vmw_cmdbuf_man *man)
-{
-	if (!man)
-		return;
-
-	tasklet_schedule(&man->tasklet);
-}
 
 /**
  * vmw_cmdbuf_send_device_command - Send a command through the device context.
@@ -1061,9 +1142,9 @@ static int vmw_cmdbuf_send_device_command(struct vmw_cmdbuf_man *man,
 	memcpy(cmd, command, size);
 	header->cb_header->length = size;
 	header->cb_context = SVGA_CB_CONTEXT_DEVICE;
-	spin_lock_bh(&man->lock);
+	spin_lock(&man->lock);
 	status = vmw_cmdbuf_header_submit(header);
-	spin_unlock_bh(&man->lock);
+	spin_unlock(&man->lock);
 	vmw_cmdbuf_header_free(header);
 
 	if (status != SVGA_CB_STATUS_COMPLETED) {
@@ -1076,6 +1157,29 @@ static int vmw_cmdbuf_send_device_command(struct vmw_cmdbuf_man *man,
 }
 
 /**
+ * vmw_cmdbuf_preempt - Send a preempt command through the device
+ * context.
+ *
+ * @man: The command buffer manager.
+ *
+ * Synchronously sends a preempt command.
+ */
+static int vmw_cmdbuf_preempt(struct vmw_cmdbuf_man *man, u32 context)
+{
+	struct {
+		uint32 id;
+		SVGADCCmdPreempt body;
+	} __packed cmd;
+
+	cmd.id = SVGA_DC_CMD_PREEMPT;
+	cmd.body.context = SVGA_CB_CONTEXT_0 + context;
+	cmd.body.ignoreIDZero = 0;
+
+	return vmw_cmdbuf_send_device_command(man, &cmd, sizeof(cmd));
+}
+
+
+/**
  * vmw_cmdbuf_startstop - Send a start / stop command through the device
  * context.
  *
@@ -1084,7 +1188,7 @@ static int vmw_cmdbuf_send_device_command(struct vmw_cmdbuf_man *man,
  *
  * Synchronously sends a device start / stop context command.
  */
-static int vmw_cmdbuf_startstop(struct vmw_cmdbuf_man *man,
+static int vmw_cmdbuf_startstop(struct vmw_cmdbuf_man *man, u32 context,
 				bool enable)
 {
 	struct {
@@ -1094,7 +1198,7 @@ static int vmw_cmdbuf_startstop(struct vmw_cmdbuf_man *man,
 
 	cmd.id = SVGA_DC_CMD_START_STOP_CONTEXT;
 	cmd.body.enable = (enable) ? 1 : 0;
-	cmd.body.context = SVGA_CB_CONTEXT_0;
+	cmd.body.context = SVGA_CB_CONTEXT_0 + context;
 
 	return vmw_cmdbuf_send_device_command(man, &cmd, sizeof(cmd));
 }
@@ -1193,7 +1297,7 @@ struct vmw_cmdbuf_man *vmw_cmdbuf_man_create(struct vmw_private *dev_priv)
 {
 	struct vmw_cmdbuf_man *man;
 	struct vmw_cmdbuf_context *ctx;
-	int i;
+	unsigned int i;
 	int ret;
 
 	if (!(dev_priv->capabilities & SVGA_CAP_COMMAND_BUFFERS))
@@ -1228,8 +1332,7 @@ struct vmw_cmdbuf_man *vmw_cmdbuf_man_create(struct vmw_private *dev_priv)
 	spin_lock_init(&man->lock);
 	mutex_init(&man->cur_mutex);
 	mutex_init(&man->space_mutex);
-	tasklet_init(&man->tasklet, vmw_cmdbuf_man_tasklet,
-		     (unsigned long) man);
+	mutex_init(&man->error_mutex);
 	man->default_size = VMW_CMDBUF_INLINE_SIZE;
 	init_waitqueue_head(&man->alloc_queue);
 	init_waitqueue_head(&man->idle_queue);
@@ -1238,11 +1341,14 @@ struct vmw_cmdbuf_man *vmw_cmdbuf_man_create(struct vmw_private *dev_priv)
 	INIT_WORK(&man->work, &vmw_cmdbuf_work_func);
 	vmw_generic_waiter_add(dev_priv, SVGA_IRQFLAG_ERROR,
 			       &dev_priv->error_waiters);
-	ret = vmw_cmdbuf_startstop(man, true);
-	if (ret) {
-		DRM_ERROR("Failed starting command buffer context 0.\n");
-		vmw_cmdbuf_man_destroy(man);
-		return ERR_PTR(ret);
+	for_each_cmdbuf_ctx(man, i, ctx) {
+		ret = vmw_cmdbuf_startstop(man, i, true);
+		if (ret) {
+			DRM_ERROR("Failed starting command buffer "
+				  "context %u.\n", i);
+			vmw_cmdbuf_man_destroy(man);
+			return ERR_PTR(ret);
+		}
 	}
 
 	return man;
@@ -1292,18 +1398,24 @@ void vmw_cmdbuf_remove_pool(struct vmw_cmdbuf_man *man)
  */
 void vmw_cmdbuf_man_destroy(struct vmw_cmdbuf_man *man)
 {
+	struct vmw_cmdbuf_context *ctx;
+	unsigned int i;
+
 	WARN_ON_ONCE(man->has_pool);
 	(void) vmw_cmdbuf_idle(man, false, 10*HZ);
-	if (vmw_cmdbuf_startstop(man, false))
-		DRM_ERROR("Failed stopping command buffer context 0.\n");
+
+	for_each_cmdbuf_ctx(man, i, ctx)
+		if (vmw_cmdbuf_startstop(man, i, false))
+			DRM_ERROR("Failed stopping command buffer "
+				  "context %u.\n", i);
 
 	vmw_generic_waiter_remove(man->dev_priv, SVGA_IRQFLAG_ERROR,
 				  &man->dev_priv->error_waiters);
-	tasklet_kill(&man->tasklet);
 	(void) cancel_work_sync(&man->work);
 	dma_pool_destroy(man->dheaders);
 	dma_pool_destroy(man->headers);
 	mutex_destroy(&man->cur_mutex);
 	mutex_destroy(&man->space_mutex);
+	mutex_destroy(&man->error_mutex);
 	kfree(man);
 }
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_cmdbuf_res.c b/drivers/gpu/drm/vmwgfx/vmwgfx_cmdbuf_res.c
index 1f013d45c9e9..36c7b6c839c0 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_cmdbuf_res.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_cmdbuf_res.c
@@ -205,7 +205,7 @@ int vmw_cmdbuf_res_add(struct vmw_cmdbuf_res_manager *man,
 	int ret;
 
 	cres = kzalloc(sizeof(*cres), GFP_KERNEL);
-	if (unlikely(cres == NULL))
+	if (unlikely(!cres))
 		return -ENOMEM;
 
 	cres->hash.key = user_key | (res_type << 24);
@@ -291,7 +291,7 @@ vmw_cmdbuf_res_man_create(struct vmw_private *dev_priv)
 	int ret;
 
 	man = kzalloc(sizeof(*man), GFP_KERNEL);
-	if (man == NULL)
+	if (!man)
 		return ERR_PTR(-ENOMEM);
 
 	man->dev_priv = dev_priv;
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_context.c b/drivers/gpu/drm/vmwgfx/vmwgfx_context.c
index bcc6d4136c87..4212b3e673bc 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_context.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_context.c
@@ -210,8 +210,8 @@ static int vmw_gb_context_init(struct vmw_private *dev_priv,
 		for (i = 0; i < SVGA_COTABLE_DX10_MAX; ++i) {
 			uctx->cotables[i] = vmw_cotable_alloc(dev_priv,
 							      &uctx->res, i);
-			if (unlikely(uctx->cotables[i] == NULL)) {
-				ret = -ENOMEM;
+			if (unlikely(IS_ERR(uctx->cotables[i]))) {
+				ret = PTR_ERR(uctx->cotables[i]);
 				goto out_cotables;
 			}
 		}
@@ -777,7 +777,7 @@ static int vmw_context_define(struct drm_device *dev, void *data,
 	}
 
 	ctx = kzalloc(sizeof(*ctx), GFP_KERNEL);
-	if (unlikely(ctx == NULL)) {
+	if (unlikely(!ctx)) {
 		ttm_mem_global_free(vmw_mem_glob(dev_priv),
 				    vmw_user_context_size);
 		ret = -ENOMEM;
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_cotable.c b/drivers/gpu/drm/vmwgfx/vmwgfx_cotable.c
index 6c026d75c180..d87861bbe971 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_cotable.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_cotable.c
@@ -584,7 +584,7 @@ struct vmw_resource *vmw_cotable_alloc(struct vmw_private *dev_priv,
 		return ERR_PTR(ret);
 
 	vcotbl = kzalloc(sizeof(*vcotbl), GFP_KERNEL);
-	if (unlikely(vcotbl == NULL)) {
+	if (unlikely(!vcotbl)) {
 		ret = -ENOMEM;
 		goto out_no_alloc;
 	}
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
index 204bf181b69e..e84fee3ec4f3 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
@@ -36,7 +36,6 @@
 #include <drm/ttm/ttm_module.h>
 #include <linux/dma_remapping.h>
 
-#define VMWGFX_DRIVER_NAME "vmwgfx"
 #define VMWGFX_DRIVER_DESC "Linux drm driver for VMware graphics devices"
 #define VMWGFX_CHIP_SVGAII 0
 #define VMW_FB_RESERVATION 0
@@ -227,7 +226,7 @@ static const struct drm_ioctl_desc vmw_ioctls[] = {
 		      DRM_AUTH | DRM_RENDER_ALLOW),
 };
 
-static struct pci_device_id vmw_pci_id_list[] = {
+static const struct pci_device_id vmw_pci_id_list[] = {
 	{0x15ad, 0x0405, PCI_ANY_ID, PCI_ANY_ID, 0, 0, VMWGFX_CHIP_SVGAII},
 	{0, 0, 0}
 };
@@ -630,7 +629,7 @@ static int vmw_driver_load(struct drm_device *dev, unsigned long chipset)
 	char host_log[100] = {0};
 
 	dev_priv = kzalloc(sizeof(*dev_priv), GFP_KERNEL);
-	if (unlikely(dev_priv == NULL)) {
+	if (unlikely(!dev_priv)) {
 		DRM_ERROR("Failed allocating a device private struct.\n");
 		return -ENOMEM;
 	}
@@ -825,7 +824,7 @@ static int vmw_driver_load(struct drm_device *dev, unsigned long chipset)
 	}
 
 	if (dev_priv->capabilities & SVGA_CAP_IRQMASK) {
-		ret = drm_irq_install(dev, dev->pdev->irq);
+		ret = vmw_irq_install(dev, dev->pdev->irq);
 		if (ret != 0) {
 			DRM_ERROR("Failed installing irq: %d\n", ret);
 			goto out_no_irq;
@@ -937,7 +936,7 @@ out_no_bdev:
 	vmw_fence_manager_takedown(dev_priv->fman);
 out_no_fman:
 	if (dev_priv->capabilities & SVGA_CAP_IRQMASK)
-		drm_irq_uninstall(dev_priv->dev);
+		vmw_irq_uninstall(dev_priv->dev);
 out_no_irq:
 	if (dev_priv->stealth)
 		pci_release_region(dev->pdev, 2);
@@ -990,7 +989,7 @@ static void vmw_driver_unload(struct drm_device *dev)
 	vmw_release_device_late(dev_priv);
 	vmw_fence_manager_takedown(dev_priv->fman);
 	if (dev_priv->capabilities & SVGA_CAP_IRQMASK)
-		drm_irq_uninstall(dev_priv->dev);
+		vmw_irq_uninstall(dev_priv->dev);
 	if (dev_priv->stealth)
 		pci_release_region(dev->pdev, 2);
 	else
@@ -1035,7 +1034,7 @@ static int vmw_driver_open(struct drm_device *dev, struct drm_file *file_priv)
 	int ret = -ENOMEM;
 
 	vmw_fp = kzalloc(sizeof(*vmw_fp), GFP_KERNEL);
-	if (unlikely(vmw_fp == NULL))
+	if (unlikely(!vmw_fp))
 		return ret;
 
 	vmw_fp->tfile = ttm_object_file_init(dev_priv->tdev, 10);
@@ -1196,7 +1195,7 @@ static int vmw_master_create(struct drm_device *dev,
 	struct vmw_master *vmaster;
 
 	vmaster = kzalloc(sizeof(*vmaster), GFP_KERNEL);
-	if (unlikely(vmaster == NULL))
+	if (unlikely(!vmaster))
 		return -ENOMEM;
 
 	vmw_master_init(vmaster);
@@ -1516,10 +1515,6 @@ static struct drm_driver driver = {
 	.load = vmw_driver_load,
 	.unload = vmw_driver_unload,
 	.lastclose = vmw_lastclose,
-	.irq_preinstall = vmw_irq_preinstall,
-	.irq_postinstall = vmw_irq_postinstall,
-	.irq_uninstall = vmw_irq_uninstall,
-	.irq_handler = vmw_irq_handler,
 	.get_vblank_counter = vmw_get_vblank_counter,
 	.enable_vblank = vmw_enable_vblank,
 	.disable_vblank = vmw_disable_vblank,
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
index 4b948fba9eec..7e5f30e234b1 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
@@ -40,10 +40,12 @@
 #include <drm/ttm/ttm_execbuf_util.h>
 #include <drm/ttm/ttm_module.h>
 #include "vmwgfx_fence.h"
+#include <linux/sync_file.h>
 
-#define VMWGFX_DRIVER_DATE "20170607"
+#define VMWGFX_DRIVER_NAME "vmwgfx"
+#define VMWGFX_DRIVER_DATE "20170612"
 #define VMWGFX_DRIVER_MAJOR 2
-#define VMWGFX_DRIVER_MINOR 13
+#define VMWGFX_DRIVER_MINOR 14
 #define VMWGFX_DRIVER_PATCHLEVEL 0
 #define VMWGFX_FILE_PAGE_OFFSET 0x00100000
 #define VMWGFX_FIFO_STATIC_SIZE (1024*1024)
@@ -351,6 +353,12 @@ struct vmw_otable_batch {
 	struct ttm_buffer_object *otable_bo;
 };
 
+enum {
+	VMW_IRQTHREAD_FENCE,
+	VMW_IRQTHREAD_CMDBUF,
+	VMW_IRQTHREAD_MAX
+};
+
 struct vmw_private {
 	struct ttm_bo_device bdev;
 	struct ttm_bo_global_ref bo_global_ref;
@@ -529,6 +537,7 @@ struct vmw_private {
 	struct vmw_otable_batch otable_batch;
 
 	struct vmw_cmdbuf_man *cman;
+	DECLARE_BITMAP(irqthread_pending, VMW_IRQTHREAD_MAX);
 };
 
 static inline struct vmw_surface *vmw_res_to_srf(struct vmw_resource *res)
@@ -561,24 +570,21 @@ static inline struct vmw_master *vmw_master(struct drm_master *master)
 static inline void vmw_write(struct vmw_private *dev_priv,
 			     unsigned int offset, uint32_t value)
 {
-	unsigned long irq_flags;
-
-	spin_lock_irqsave(&dev_priv->hw_lock, irq_flags);
+	spin_lock(&dev_priv->hw_lock);
 	outl(offset, dev_priv->io_start + VMWGFX_INDEX_PORT);
 	outl(value, dev_priv->io_start + VMWGFX_VALUE_PORT);
-	spin_unlock_irqrestore(&dev_priv->hw_lock, irq_flags);
+	spin_unlock(&dev_priv->hw_lock);
 }
 
 static inline uint32_t vmw_read(struct vmw_private *dev_priv,
 				unsigned int offset)
 {
-	unsigned long irq_flags;
 	u32 val;
 
-	spin_lock_irqsave(&dev_priv->hw_lock, irq_flags);
+	spin_lock(&dev_priv->hw_lock);
 	outl(offset, dev_priv->io_start + VMWGFX_INDEX_PORT);
 	val = inl(dev_priv->io_start + VMWGFX_VALUE_PORT);
-	spin_unlock_irqrestore(&dev_priv->hw_lock, irq_flags);
+	spin_unlock(&dev_priv->hw_lock);
 
 	return val;
 }
@@ -821,7 +827,8 @@ extern int vmw_execbuf_process(struct drm_file *file_priv,
 			       uint32_t dx_context_handle,
 			       struct drm_vmw_fence_rep __user
 			       *user_fence_rep,
-			       struct vmw_fence_obj **out_fence);
+			       struct vmw_fence_obj **out_fence,
+			       uint32_t flags);
 extern void __vmw_execbuf_release_pinned_bo(struct vmw_private *dev_priv,
 					    struct vmw_fence_obj *fence);
 extern void vmw_execbuf_release_pinned_bo(struct vmw_private *dev_priv);
@@ -836,23 +843,23 @@ extern void vmw_execbuf_copy_fence_user(struct vmw_private *dev_priv,
 					struct drm_vmw_fence_rep __user
 					*user_fence_rep,
 					struct vmw_fence_obj *fence,
-					uint32_t fence_handle);
+					uint32_t fence_handle,
+					int32_t out_fence_fd,
+					struct sync_file *sync_file);
 extern int vmw_validate_single_buffer(struct vmw_private *dev_priv,
 				      struct ttm_buffer_object *bo,
 				      bool interruptible,
 				      bool validate_as_mob);
-
+bool vmw_cmd_describe(const void *buf, u32 *size, char const **cmd);
 
 /**
  * IRQs and wating - vmwgfx_irq.c
  */
 
-extern irqreturn_t vmw_irq_handler(int irq, void *arg);
 extern int vmw_wait_seqno(struct vmw_private *dev_priv, bool lazy,
 			  uint32_t seqno, bool interruptible,
 			  unsigned long timeout);
-extern void vmw_irq_preinstall(struct drm_device *dev);
-extern int vmw_irq_postinstall(struct drm_device *dev);
+extern int vmw_irq_install(struct drm_device *dev, int irq);
 extern void vmw_irq_uninstall(struct drm_device *dev);
 extern bool vmw_seqno_passed(struct vmw_private *dev_priv,
 				uint32_t seqno);
@@ -1150,13 +1157,13 @@ extern void *vmw_cmdbuf_reserve(struct vmw_cmdbuf_man *man, size_t size,
 extern void vmw_cmdbuf_commit(struct vmw_cmdbuf_man *man, size_t size,
 			      struct vmw_cmdbuf_header *header,
 			      bool flush);
-extern void vmw_cmdbuf_tasklet_schedule(struct vmw_cmdbuf_man *man);
 extern void *vmw_cmdbuf_alloc(struct vmw_cmdbuf_man *man,
 			      size_t size, bool interruptible,
 			      struct vmw_cmdbuf_header **p_header);
 extern void vmw_cmdbuf_header_free(struct vmw_cmdbuf_header *header);
 extern int vmw_cmdbuf_cur_flush(struct vmw_cmdbuf_man *man,
 				bool interruptible);
+extern void vmw_cmdbuf_irqthread(struct vmw_cmdbuf_man *man);
 
 
 /**
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c
index c7b53d987f06..21c62a34e558 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c
@@ -24,6 +24,7 @@
  * USE OR OTHER DEALINGS IN THE SOFTWARE.
  *
  **************************************************************************/
+#include <linux/sync_file.h>
 
 #include "vmwgfx_drv.h"
 #include "vmwgfx_reg.h"
@@ -112,11 +113,12 @@ struct vmw_cmd_entry {
 	bool user_allow;
 	bool gb_disable;
 	bool gb_enable;
+	const char *cmd_name;
 };
 
 #define VMW_CMD_DEF(_cmd, _func, _user_allow, _gb_disable, _gb_enable)	\
 	[(_cmd) - SVGA_3D_CMD_BASE] = {(_func), (_user_allow),\
-				       (_gb_disable), (_gb_enable)}
+				       (_gb_disable), (_gb_enable), #_cmd}
 
 static int vmw_resource_context_res_add(struct vmw_private *dev_priv,
 					struct vmw_sw_context *sw_context,
@@ -264,7 +266,7 @@ static int vmw_resource_val_add(struct vmw_sw_context *sw_context,
 	}
 
 	node = kzalloc(sizeof(*node), GFP_KERNEL);
-	if (unlikely(node == NULL)) {
+	if (unlikely(!node)) {
 		DRM_ERROR("Failed to allocate a resource validation "
 			  "entry.\n");
 		return -ENOMEM;
@@ -452,7 +454,7 @@ static int vmw_resource_relocation_add(struct list_head *list,
 	struct vmw_resource_relocation *rel;
 
 	rel = kmalloc(sizeof(*rel), GFP_KERNEL);
-	if (unlikely(rel == NULL)) {
+	if (unlikely(!rel)) {
 		DRM_ERROR("Failed to allocate a resource relocation.\n");
 		return -ENOMEM;
 	}
@@ -519,7 +521,7 @@ static int vmw_cmd_invalid(struct vmw_private *dev_priv,
 			   struct vmw_sw_context *sw_context,
 			   SVGA3dCmdHeader *header)
 {
-	return capable(CAP_SYS_ADMIN) ? : -EINVAL;
+	return -EINVAL;
 }
 
 static int vmw_cmd_ok(struct vmw_private *dev_priv,
@@ -2584,7 +2586,7 @@ static int vmw_cmd_dx_set_vertex_buffers(struct vmw_private *dev_priv,
 
 /**
  * vmw_cmd_dx_ia_set_vertex_buffers - Validate an
- * SVGA_3D_CMD_DX_IA_SET_VERTEX_BUFFERS command.
+ * SVGA_3D_CMD_DX_IA_SET_INDEX_BUFFER command.
  *
  * @dev_priv: Pointer to a device private struct.
  * @sw_context: The software context being used for this batch.
@@ -3302,6 +3304,8 @@ static const struct vmw_cmd_entry vmw_cmd_entries[SVGA_3D_CMD_MAX] = {
 		    true, false, true),
 	VMW_CMD_DEF(SVGA_3D_CMD_NOP, &vmw_cmd_ok,
 		    true, false, true),
+	VMW_CMD_DEF(SVGA_3D_CMD_NOP_ERROR, &vmw_cmd_ok,
+		    true, false, true),
 	VMW_CMD_DEF(SVGA_3D_CMD_ENABLE_GART, &vmw_cmd_invalid,
 		    false, false, true),
 	VMW_CMD_DEF(SVGA_3D_CMD_DISABLE_GART, &vmw_cmd_invalid,
@@ -3469,6 +3473,51 @@ static const struct vmw_cmd_entry vmw_cmd_entries[SVGA_3D_CMD_MAX] = {
 		    true, false, true),
 };
 
+bool vmw_cmd_describe(const void *buf, u32 *size, char const **cmd)
+{
+	u32 cmd_id = ((u32 *) buf)[0];
+
+	if (cmd_id >= SVGA_CMD_MAX) {
+		SVGA3dCmdHeader *header = (SVGA3dCmdHeader *) buf;
+		const struct vmw_cmd_entry *entry;
+
+		*size = header->size + sizeof(SVGA3dCmdHeader);
+		cmd_id = header->id;
+		if (cmd_id >= SVGA_3D_CMD_MAX)
+			return false;
+
+		cmd_id -= SVGA_3D_CMD_BASE;
+		entry = &vmw_cmd_entries[cmd_id];
+		*cmd = entry->cmd_name;
+		return true;
+	}
+
+	switch (cmd_id) {
+	case SVGA_CMD_UPDATE:
+		*cmd = "SVGA_CMD_UPDATE";
+		*size = sizeof(u32) + sizeof(SVGAFifoCmdUpdate);
+		break;
+	case SVGA_CMD_DEFINE_GMRFB:
+		*cmd = "SVGA_CMD_DEFINE_GMRFB";
+		*size = sizeof(u32) + sizeof(SVGAFifoCmdDefineGMRFB);
+		break;
+	case SVGA_CMD_BLIT_GMRFB_TO_SCREEN:
+		*cmd = "SVGA_CMD_BLIT_GMRFB_TO_SCREEN";
+		*size = sizeof(u32) + sizeof(SVGAFifoCmdBlitGMRFBToScreen);
+		break;
+	case SVGA_CMD_BLIT_SCREEN_TO_GMRFB:
+		*cmd = "SVGA_CMD_BLIT_SCREEN_TO_GMRFB";
+		*size = sizeof(u32) + sizeof(SVGAFifoCmdBlitGMRFBToScreen);
+		break;
+	default:
+		*cmd = "UNKNOWN";
+		*size = 0;
+		return false;
+	}
+
+	return true;
+}
+
 static int vmw_cmd_check(struct vmw_private *dev_priv,
 			 struct vmw_sw_context *sw_context,
 			 void *buf, uint32_t *size)
@@ -3781,6 +3830,8 @@ int vmw_execbuf_fence_commands(struct drm_file *file_priv,
  * which the information should be copied.
  * @fence: Pointer to the fenc object.
  * @fence_handle: User-space fence handle.
+ * @out_fence_fd: exported file descriptor for the fence.  -1 if not used
+ * @sync_file:  Only used to clean up in case of an error in this function.
  *
  * This function copies fence information to user-space. If copying fails,
  * The user-space struct drm_vmw_fence_rep::error member is hopefully
@@ -3796,7 +3847,9 @@ vmw_execbuf_copy_fence_user(struct vmw_private *dev_priv,
 			    int ret,
 			    struct drm_vmw_fence_rep __user *user_fence_rep,
 			    struct vmw_fence_obj *fence,
-			    uint32_t fence_handle)
+			    uint32_t fence_handle,
+			    int32_t out_fence_fd,
+			    struct sync_file *sync_file)
 {
 	struct drm_vmw_fence_rep fence_rep;
 
@@ -3806,6 +3859,7 @@ vmw_execbuf_copy_fence_user(struct vmw_private *dev_priv,
 	memset(&fence_rep, 0, sizeof(fence_rep));
 
 	fence_rep.error = ret;
+	fence_rep.fd = out_fence_fd;
 	if (ret == 0) {
 		BUG_ON(fence == NULL);
 
@@ -3828,6 +3882,14 @@ vmw_execbuf_copy_fence_user(struct vmw_private *dev_priv,
 	 * and unreference the handle.
 	 */
 	if (unlikely(ret != 0) && (fence_rep.error == 0)) {
+		if (sync_file)
+			fput(sync_file->file);
+
+		if (fence_rep.fd != -1) {
+			put_unused_fd(fence_rep.fd);
+			fence_rep.fd = -1;
+		}
+
 		ttm_ref_object_base_unref(vmw_fp->tfile,
 					  fence_handle, TTM_REF_USAGE);
 		DRM_ERROR("Fence copy error. Syncing.\n");
@@ -4003,7 +4065,8 @@ int vmw_execbuf_process(struct drm_file *file_priv,
 			uint64_t throttle_us,
 			uint32_t dx_context_handle,
 			struct drm_vmw_fence_rep __user *user_fence_rep,
-			struct vmw_fence_obj **out_fence)
+			struct vmw_fence_obj **out_fence,
+			uint32_t flags)
 {
 	struct vmw_sw_context *sw_context = &dev_priv->ctx;
 	struct vmw_fence_obj *fence = NULL;
@@ -4013,20 +4076,33 @@ int vmw_execbuf_process(struct drm_file *file_priv,
 	struct ww_acquire_ctx ticket;
 	uint32_t handle;
 	int ret;
+	int32_t out_fence_fd = -1;
+	struct sync_file *sync_file = NULL;
+
+
+	if (flags & DRM_VMW_EXECBUF_FLAG_EXPORT_FENCE_FD) {
+		out_fence_fd = get_unused_fd_flags(O_CLOEXEC);
+		if (out_fence_fd < 0) {
+			DRM_ERROR("Failed to get a fence file descriptor.\n");
+			return out_fence_fd;
+		}
+	}
 
 	if (throttle_us) {
 		ret = vmw_wait_lag(dev_priv, &dev_priv->fifo.marker_queue,
 				   throttle_us);
 
 		if (ret)
-			return ret;
+			goto out_free_fence_fd;
 	}
 
 	kernel_commands = vmw_execbuf_cmdbuf(dev_priv, user_commands,
 					     kernel_commands, command_size,
 					     &header);
-	if (IS_ERR(kernel_commands))
-		return PTR_ERR(kernel_commands);
+	if (IS_ERR(kernel_commands)) {
+		ret = PTR_ERR(kernel_commands);
+		goto out_free_fence_fd;
+	}
 
 	ret = mutex_lock_interruptible(&dev_priv->cmdbuf_mutex);
 	if (ret) {
@@ -4162,8 +4238,32 @@ int vmw_execbuf_process(struct drm_file *file_priv,
 		__vmw_execbuf_release_pinned_bo(dev_priv, fence);
 
 	vmw_clear_validations(sw_context);
+
+	/*
+	 * If anything fails here, give up trying to export the fence
+	 * and do a sync since the user mode will not be able to sync
+	 * the fence itself.  This ensures we are still functionally
+	 * correct.
+	 */
+	if (flags & DRM_VMW_EXECBUF_FLAG_EXPORT_FENCE_FD) {
+
+		sync_file = sync_file_create(&fence->base);
+		if (!sync_file) {
+			DRM_ERROR("Unable to create sync file for fence\n");
+			put_unused_fd(out_fence_fd);
+			out_fence_fd = -1;
+
+			(void) vmw_fence_obj_wait(fence, false, false,
+						  VMW_FENCE_WAIT_TIMEOUT);
+		} else {
+			/* Link the fence with the FD created earlier */
+			fd_install(out_fence_fd, sync_file->file);
+		}
+	}
+
 	vmw_execbuf_copy_fence_user(dev_priv, vmw_fpriv(file_priv), ret,
-				    user_fence_rep, fence, handle);
+				    user_fence_rep, fence, handle,
+				    out_fence_fd, sync_file);
 
 	/* Don't unreference when handing fence out */
 	if (unlikely(out_fence != NULL)) {
@@ -4214,6 +4314,9 @@ out_unlock:
 out_free_header:
 	if (header)
 		vmw_cmdbuf_header_free(header);
+out_free_fence_fd:
+	if (out_fence_fd >= 0)
+		put_unused_fd(out_fence_fd);
 
 	return ret;
 }
@@ -4366,6 +4469,7 @@ int vmw_execbuf_ioctl(struct drm_device *dev, unsigned long data,
 	static const size_t copy_offset[] = {
 		offsetof(struct drm_vmw_execbuf_arg, context_handle),
 		sizeof(struct drm_vmw_execbuf_arg)};
+	struct dma_fence *in_fence = NULL;
 
 	if (unlikely(size < copy_offset[0])) {
 		DRM_ERROR("Invalid command size, ioctl %d\n",
@@ -4401,15 +4505,25 @@ int vmw_execbuf_ioctl(struct drm_device *dev, unsigned long data,
 		arg.context_handle = (uint32_t) -1;
 		break;
 	case 2:
-		if (arg.pad64 != 0) {
-			DRM_ERROR("Unused IOCTL data not set to zero.\n");
-			return -EINVAL;
-		}
-		break;
 	default:
 		break;
 	}
 
+
+	/* If imported a fence FD from elsewhere, then wait on it */
+	if (arg.flags & DRM_VMW_EXECBUF_FLAG_IMPORT_FENCE_FD) {
+		in_fence = sync_file_get_fence(arg.imported_fence_fd);
+
+		if (!in_fence) {
+			DRM_ERROR("Cannot get imported fence\n");
+			return -EINVAL;
+		}
+
+		ret = vmw_wait_dma_fence(dev_priv->fman, in_fence);
+		if (ret)
+			goto out;
+	}
+
 	ret = ttm_read_lock(&dev_priv->reservation_sem, true);
 	if (unlikely(ret != 0))
 		return ret;
@@ -4419,12 +4533,16 @@ int vmw_execbuf_ioctl(struct drm_device *dev, unsigned long data,
 				  NULL, arg.command_size, arg.throttle_us,
 				  arg.context_handle,
 				  (void __user *)(unsigned long)arg.fence_rep,
-				  NULL);
+				  NULL,
+				  arg.flags);
 	ttm_read_unlock(&dev_priv->reservation_sem);
 	if (unlikely(ret != 0))
-		return ret;
+		goto out;
 
 	vmw_kms_cursor_post_execbuf(dev_priv);
 
-	return 0;
+out:
+	if (in_fence)
+		dma_fence_put(in_fence);
+	return ret;
 }
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c b/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c
index 6b2708b4eafe..3bbad22b3748 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c
@@ -114,12 +114,11 @@ static void vmw_fence_obj_destroy(struct dma_fence *f)
 		container_of(f, struct vmw_fence_obj, base);
 
 	struct vmw_fence_manager *fman = fman_from_fence(fence);
-	unsigned long irq_flags;
 
-	spin_lock_irqsave(&fman->lock, irq_flags);
+	spin_lock(&fman->lock);
 	list_del_init(&fence->head);
 	--fman->num_fence_objects;
-	spin_unlock_irqrestore(&fman->lock, irq_flags);
+	spin_unlock(&fman->lock);
 	fence->destroy(fence);
 }
 
@@ -252,10 +251,10 @@ static void vmw_fence_work_func(struct work_struct *work)
 		INIT_LIST_HEAD(&list);
 		mutex_lock(&fman->goal_irq_mutex);
 
-		spin_lock_irq(&fman->lock);
+		spin_lock(&fman->lock);
 		list_splice_init(&fman->cleanup_list, &list);
 		seqno_valid = fman->seqno_valid;
-		spin_unlock_irq(&fman->lock);
+		spin_unlock(&fman->lock);
 
 		if (!seqno_valid && fman->goal_irq_on) {
 			fman->goal_irq_on = false;
@@ -284,7 +283,7 @@ struct vmw_fence_manager *vmw_fence_manager_init(struct vmw_private *dev_priv)
 {
 	struct vmw_fence_manager *fman = kzalloc(sizeof(*fman), GFP_KERNEL);
 
-	if (unlikely(fman == NULL))
+	if (unlikely(!fman))
 		return NULL;
 
 	fman->dev_priv = dev_priv;
@@ -305,15 +304,14 @@ struct vmw_fence_manager *vmw_fence_manager_init(struct vmw_private *dev_priv)
 
 void vmw_fence_manager_takedown(struct vmw_fence_manager *fman)
 {
-	unsigned long irq_flags;
 	bool lists_empty;
 
 	(void) cancel_work_sync(&fman->work);
 
-	spin_lock_irqsave(&fman->lock, irq_flags);
+	spin_lock(&fman->lock);
 	lists_empty = list_empty(&fman->fence_list) &&
 		list_empty(&fman->cleanup_list);
-	spin_unlock_irqrestore(&fman->lock, irq_flags);
+	spin_unlock(&fman->lock);
 
 	BUG_ON(!lists_empty);
 	kfree(fman);
@@ -323,7 +321,6 @@ static int vmw_fence_obj_init(struct vmw_fence_manager *fman,
 			      struct vmw_fence_obj *fence, u32 seqno,
 			      void (*destroy) (struct vmw_fence_obj *fence))
 {
-	unsigned long irq_flags;
 	int ret = 0;
 
 	dma_fence_init(&fence->base, &vmw_fence_ops, &fman->lock,
@@ -331,7 +328,7 @@ static int vmw_fence_obj_init(struct vmw_fence_manager *fman,
 	INIT_LIST_HEAD(&fence->seq_passed_actions);
 	fence->destroy = destroy;
 
-	spin_lock_irqsave(&fman->lock, irq_flags);
+	spin_lock(&fman->lock);
 	if (unlikely(fman->fifo_down)) {
 		ret = -EBUSY;
 		goto out_unlock;
@@ -340,7 +337,7 @@ static int vmw_fence_obj_init(struct vmw_fence_manager *fman,
 	++fman->num_fence_objects;
 
 out_unlock:
-	spin_unlock_irqrestore(&fman->lock, irq_flags);
+	spin_unlock(&fman->lock);
 	return ret;
 
 }
@@ -489,11 +486,9 @@ rerun:
 
 void vmw_fences_update(struct vmw_fence_manager *fman)
 {
-	unsigned long irq_flags;
-
-	spin_lock_irqsave(&fman->lock, irq_flags);
+	spin_lock(&fman->lock);
 	__vmw_fences_update(fman);
-	spin_unlock_irqrestore(&fman->lock, irq_flags);
+	spin_unlock(&fman->lock);
 }
 
 bool vmw_fence_obj_signaled(struct vmw_fence_obj *fence)
@@ -541,7 +536,7 @@ int vmw_fence_create(struct vmw_fence_manager *fman,
  	int ret;
 
 	fence = kzalloc(sizeof(*fence), GFP_KERNEL);
-	if (unlikely(fence == NULL))
+	if (unlikely(!fence))
 		return -ENOMEM;
 
 	ret = vmw_fence_obj_init(fman, fence, seqno,
@@ -606,7 +601,7 @@ int vmw_user_fence_create(struct drm_file *file_priv,
 		return ret;
 
 	ufence = kzalloc(sizeof(*ufence), GFP_KERNEL);
-	if (unlikely(ufence == NULL)) {
+	if (unlikely(!ufence)) {
 		ret = -ENOMEM;
 		goto out_no_object;
 	}
@@ -650,6 +645,51 @@ out_no_object:
 
 
 /**
+ * vmw_wait_dma_fence - Wait for a dma fence
+ *
+ * @fman: pointer to a fence manager
+ * @fence: DMA fence to wait on
+ *
+ * This function handles the case when the fence is actually a fence
+ * array.  If that's the case, it'll wait on each of the child fence
+ */
+int vmw_wait_dma_fence(struct vmw_fence_manager *fman,
+		       struct dma_fence *fence)
+{
+	struct dma_fence_array *fence_array;
+	int ret = 0;
+	int i;
+
+
+	if (dma_fence_is_signaled(fence))
+		return 0;
+
+	if (!dma_fence_is_array(fence))
+		return dma_fence_wait(fence, true);
+
+	/* From i915: Note that if the fence-array was created in
+	 * signal-on-any mode, we should *not* decompose it into its individual
+	 * fences. However, we don't currently store which mode the fence-array
+	 * is operating in. Fortunately, the only user of signal-on-any is
+	 * private to amdgpu and we should not see any incoming fence-array
+	 * from sync-file being in signal-on-any mode.
+	 */
+
+	fence_array = to_dma_fence_array(fence);
+	for (i = 0; i < fence_array->num_fences; i++) {
+		struct dma_fence *child = fence_array->fences[i];
+
+		ret = dma_fence_wait(child, true);
+
+		if (ret < 0)
+			return ret;
+	}
+
+	return 0;
+}
+
+
+/**
  * vmw_fence_fifo_down - signal all unsignaled fence objects.
  */
 
@@ -663,14 +703,14 @@ void vmw_fence_fifo_down(struct vmw_fence_manager *fman)
 	 * restart when we've released the fman->lock.
 	 */
 
-	spin_lock_irq(&fman->lock);
+	spin_lock(&fman->lock);
 	fman->fifo_down = true;
 	while (!list_empty(&fman->fence_list)) {
 		struct vmw_fence_obj *fence =
 			list_entry(fman->fence_list.prev, struct vmw_fence_obj,
 				   head);
 		dma_fence_get(&fence->base);
-		spin_unlock_irq(&fman->lock);
+		spin_unlock(&fman->lock);
 
 		ret = vmw_fence_obj_wait(fence, false, false,
 					 VMW_FENCE_WAIT_TIMEOUT);
@@ -686,18 +726,16 @@ void vmw_fence_fifo_down(struct vmw_fence_manager *fman)
 
 		BUG_ON(!list_empty(&fence->head));
 		dma_fence_put(&fence->base);
-		spin_lock_irq(&fman->lock);
+		spin_lock(&fman->lock);
 	}
-	spin_unlock_irq(&fman->lock);
+	spin_unlock(&fman->lock);
 }
 
 void vmw_fence_fifo_up(struct vmw_fence_manager *fman)
 {
-	unsigned long irq_flags;
-
-	spin_lock_irqsave(&fman->lock, irq_flags);
+	spin_lock(&fman->lock);
 	fman->fifo_down = false;
-	spin_unlock_irqrestore(&fman->lock, irq_flags);
+	spin_unlock(&fman->lock);
 }
 
 
@@ -812,9 +850,9 @@ int vmw_fence_obj_signaled_ioctl(struct drm_device *dev, void *data,
 	arg->signaled = vmw_fence_obj_signaled(fence);
 
 	arg->signaled_flags = arg->flags;
-	spin_lock_irq(&fman->lock);
+	spin_lock(&fman->lock);
 	arg->passed_seqno = dev_priv->last_read_seqno;
-	spin_unlock_irq(&fman->lock);
+	spin_unlock(&fman->lock);
 
 	ttm_base_object_unref(&base);
 
@@ -841,8 +879,7 @@ int vmw_fence_obj_unref_ioctl(struct drm_device *dev, void *data,
  *
  * This function is called when the seqno of the fence where @action is
  * attached has passed. It queues the event on the submitter's event list.
- * This function is always called from atomic context, and may be called
- * from irq context.
+ * This function is always called from atomic context.
  */
 static void vmw_event_fence_action_seq_passed(struct vmw_fence_action *action)
 {
@@ -851,13 +888,13 @@ static void vmw_event_fence_action_seq_passed(struct vmw_fence_action *action)
 	struct drm_device *dev = eaction->dev;
 	struct drm_pending_event *event = eaction->event;
 	struct drm_file *file_priv;
-	unsigned long irq_flags;
+
 
 	if (unlikely(event == NULL))
 		return;
 
 	file_priv = event->file_priv;
-	spin_lock_irqsave(&dev->event_lock, irq_flags);
+	spin_lock_irq(&dev->event_lock);
 
 	if (likely(eaction->tv_sec != NULL)) {
 		struct timeval tv;
@@ -869,7 +906,7 @@ static void vmw_event_fence_action_seq_passed(struct vmw_fence_action *action)
 
 	drm_send_event_locked(dev, eaction->event);
 	eaction->event = NULL;
-	spin_unlock_irqrestore(&dev->event_lock, irq_flags);
+	spin_unlock_irq(&dev->event_lock);
 }
 
 /**
@@ -904,11 +941,10 @@ static void vmw_fence_obj_add_action(struct vmw_fence_obj *fence,
 			      struct vmw_fence_action *action)
 {
 	struct vmw_fence_manager *fman = fman_from_fence(fence);
-	unsigned long irq_flags;
 	bool run_update = false;
 
 	mutex_lock(&fman->goal_irq_mutex);
-	spin_lock_irqsave(&fman->lock, irq_flags);
+	spin_lock(&fman->lock);
 
 	fman->pending_actions[action->type]++;
 	if (dma_fence_is_signaled_locked(&fence->base)) {
@@ -927,7 +963,7 @@ static void vmw_fence_obj_add_action(struct vmw_fence_obj *fence,
 		run_update = vmw_fence_goal_check_locked(fence);
 	}
 
-	spin_unlock_irqrestore(&fman->lock, irq_flags);
+	spin_unlock(&fman->lock);
 
 	if (run_update) {
 		if (!fman->goal_irq_on) {
@@ -966,7 +1002,7 @@ int vmw_event_fence_action_queue(struct drm_file *file_priv,
 	struct vmw_fence_manager *fman = fman_from_fence(fence);
 
 	eaction = kzalloc(sizeof(*eaction), GFP_KERNEL);
-	if (unlikely(eaction == NULL))
+	if (unlikely(!eaction))
 		return -ENOMEM;
 
 	eaction->event = event;
@@ -1002,7 +1038,7 @@ static int vmw_event_fence_action_create(struct drm_file *file_priv,
 	int ret;
 
 	event = kzalloc(sizeof(*event), GFP_KERNEL);
-	if (unlikely(event == NULL)) {
+	if (unlikely(!event)) {
 		DRM_ERROR("Failed to allocate an event.\n");
 		ret = -ENOMEM;
 		goto out_no_space;
@@ -1114,7 +1150,7 @@ int vmw_fence_event_ioctl(struct drm_device *dev, void *data,
 	}
 
 	vmw_execbuf_copy_fence_user(dev_priv, vmw_fp, 0, user_fence_rep, fence,
-				    handle);
+				    handle, -1, NULL);
 	vmw_fence_obj_unreference(&fence);
 	return 0;
 out_no_create:
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_fence.h b/drivers/gpu/drm/vmwgfx/vmwgfx_fence.h
index d9d85aa6ed20..20224dba9d8e 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_fence.h
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_fence.h
@@ -28,6 +28,7 @@
 #ifndef _VMWGFX_FENCE_H_
 
 #include <linux/dma-fence.h>
+#include <linux/dma-fence-array.h>
 
 #define VMW_FENCE_WAIT_TIMEOUT (5*HZ)
 
@@ -102,6 +103,9 @@ extern int vmw_user_fence_create(struct drm_file *file_priv,
 				 struct vmw_fence_obj **p_fence,
 				 uint32_t *p_handle);
 
+extern int vmw_wait_dma_fence(struct vmw_fence_manager *fman,
+			      struct dma_fence *fence);
+
 extern void vmw_fence_fifo_up(struct vmw_fence_manager *fman);
 
 extern void vmw_fence_fifo_down(struct vmw_fence_manager *fman);
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_gmrid_manager.c b/drivers/gpu/drm/vmwgfx/vmwgfx_gmrid_manager.c
index c1900f4390a4..f2f9d88131f2 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_gmrid_manager.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_gmrid_manager.c
@@ -121,7 +121,7 @@ static int vmw_gmrid_man_init(struct ttm_mem_type_manager *man,
 	struct vmwgfx_gmrid_man *gman =
 		kzalloc(sizeof(*gman), GFP_KERNEL);
 
-	if (unlikely(gman == NULL))
+	if (unlikely(!gman))
 		return -ENOMEM;
 
 	spin_lock_init(&gman->lock);
@@ -157,9 +157,9 @@ static int vmw_gmrid_man_takedown(struct ttm_mem_type_manager *man)
 }
 
 static void vmw_gmrid_man_debug(struct ttm_mem_type_manager *man,
-				const char *prefix)
+				struct drm_printer *printer)
 {
-	pr_info("%s: No debug info available for the GMR id manager\n", prefix);
+	drm_printf(printer, "No debug info available for the GMR id manager\n");
 }
 
 const struct ttm_mem_type_manager_func vmw_gmrid_manager_func = {
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_irq.c b/drivers/gpu/drm/vmwgfx/vmwgfx_irq.c
index 0c7e1723292c..b9239ba067c4 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_irq.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_irq.c
@@ -30,11 +30,56 @@
 
 #define VMW_FENCE_WRAP (1 << 24)
 
-irqreturn_t vmw_irq_handler(int irq, void *arg)
+/**
+ * vmw_thread_fn - Deferred (process context) irq handler
+ *
+ * @irq: irq number
+ * @arg: Closure argument. Pointer to a struct drm_device cast to void *
+ *
+ * This function implements the deferred part of irq processing.
+ * The function is guaranteed to run at least once after the
+ * vmw_irq_handler has returned with IRQ_WAKE_THREAD.
+ *
+ */
+static irqreturn_t vmw_thread_fn(int irq, void *arg)
+{
+	struct drm_device *dev = (struct drm_device *)arg;
+	struct vmw_private *dev_priv = vmw_priv(dev);
+	irqreturn_t ret = IRQ_NONE;
+
+	if (test_and_clear_bit(VMW_IRQTHREAD_FENCE,
+			       dev_priv->irqthread_pending)) {
+		vmw_fences_update(dev_priv->fman);
+		wake_up_all(&dev_priv->fence_queue);
+		ret = IRQ_HANDLED;
+	}
+
+	if (test_and_clear_bit(VMW_IRQTHREAD_CMDBUF,
+			       dev_priv->irqthread_pending)) {
+		vmw_cmdbuf_irqthread(dev_priv->cman);
+		ret = IRQ_HANDLED;
+	}
+
+	return ret;
+}
+
+/**
+ * vmw_irq_handler irq handler
+ *
+ * @irq: irq number
+ * @arg: Closure argument. Pointer to a struct drm_device cast to void *
+ *
+ * This function implements the quick part of irq processing.
+ * The function performs fast actions like clearing the device interrupt
+ * flags and also reasonably quick actions like waking processes waiting for
+ * FIFO space. Other IRQ actions are deferred to the IRQ thread.
+ */
+static irqreturn_t vmw_irq_handler(int irq, void *arg)
 {
 	struct drm_device *dev = (struct drm_device *)arg;
 	struct vmw_private *dev_priv = vmw_priv(dev);
 	uint32_t status, masked_status;
+	irqreturn_t ret = IRQ_HANDLED;
 
 	status = inl(dev_priv->io_start + VMWGFX_IRQSTATUS_PORT);
 	masked_status = status & READ_ONCE(dev_priv->irq_mask);
@@ -45,20 +90,21 @@ irqreturn_t vmw_irq_handler(int irq, void *arg)
 	if (!status)
 		return IRQ_NONE;
 
-	if (masked_status & (SVGA_IRQFLAG_ANY_FENCE |
-			     SVGA_IRQFLAG_FENCE_GOAL)) {
-		vmw_fences_update(dev_priv->fman);
-		wake_up_all(&dev_priv->fence_queue);
-	}
-
 	if (masked_status & SVGA_IRQFLAG_FIFO_PROGRESS)
 		wake_up_all(&dev_priv->fifo_queue);
 
-	if (masked_status & (SVGA_IRQFLAG_COMMAND_BUFFER |
-			     SVGA_IRQFLAG_ERROR))
-		vmw_cmdbuf_tasklet_schedule(dev_priv->cman);
+	if ((masked_status & (SVGA_IRQFLAG_ANY_FENCE |
+			      SVGA_IRQFLAG_FENCE_GOAL)) &&
+	    !test_and_set_bit(VMW_IRQTHREAD_FENCE, dev_priv->irqthread_pending))
+		ret = IRQ_WAKE_THREAD;
 
-	return IRQ_HANDLED;
+	if ((masked_status & (SVGA_IRQFLAG_COMMAND_BUFFER |
+			      SVGA_IRQFLAG_ERROR)) &&
+	    !test_and_set_bit(VMW_IRQTHREAD_CMDBUF,
+			      dev_priv->irqthread_pending))
+		ret = IRQ_WAKE_THREAD;
+
+	return ret;
 }
 
 static bool vmw_fifo_idle(struct vmw_private *dev_priv, uint32_t seqno)
@@ -281,23 +327,15 @@ int vmw_wait_seqno(struct vmw_private *dev_priv,
 	return ret;
 }
 
-void vmw_irq_preinstall(struct drm_device *dev)
+static void vmw_irq_preinstall(struct drm_device *dev)
 {
 	struct vmw_private *dev_priv = vmw_priv(dev);
 	uint32_t status;
 
-	if (!(dev_priv->capabilities & SVGA_CAP_IRQMASK))
-		return;
-
 	status = inl(dev_priv->io_start + VMWGFX_IRQSTATUS_PORT);
 	outl(status, dev_priv->io_start + VMWGFX_IRQSTATUS_PORT);
 }
 
-int vmw_irq_postinstall(struct drm_device *dev)
-{
-	return 0;
-}
-
 void vmw_irq_uninstall(struct drm_device *dev)
 {
 	struct vmw_private *dev_priv = vmw_priv(dev);
@@ -306,8 +344,41 @@ void vmw_irq_uninstall(struct drm_device *dev)
 	if (!(dev_priv->capabilities & SVGA_CAP_IRQMASK))
 		return;
 
+	if (!dev->irq_enabled)
+		return;
+
 	vmw_write(dev_priv, SVGA_REG_IRQMASK, 0);
 
 	status = inl(dev_priv->io_start + VMWGFX_IRQSTATUS_PORT);
 	outl(status, dev_priv->io_start + VMWGFX_IRQSTATUS_PORT);
+
+	dev->irq_enabled = false;
+	free_irq(dev->irq, dev);
+}
+
+/**
+ * vmw_irq_install - Install the irq handlers
+ *
+ * @dev:  Pointer to the drm device.
+ * @irq:  The irq number.
+ * Return:  Zero if successful. Negative number otherwise.
+ */
+int vmw_irq_install(struct drm_device *dev, int irq)
+{
+	int ret;
+
+	if (dev->irq_enabled)
+		return -EBUSY;
+
+	vmw_irq_preinstall(dev);
+
+	ret = request_threaded_irq(irq, vmw_irq_handler, vmw_thread_fn,
+				   IRQF_SHARED, VMWGFX_DRIVER_NAME, dev);
+	if (ret < 0)
+		return ret;
+
+	dev->irq_enabled = true;
+	dev->irq = irq;
+
+	return ret;
 }
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
index 620180df1303..5d50e45ae274 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
@@ -384,6 +384,12 @@ vmw_du_cursor_plane_atomic_update(struct drm_plane *plane,
 
 	hotspot_x = du->hotspot_x;
 	hotspot_y = du->hotspot_y;
+
+	if (plane->fb) {
+		hotspot_x += plane->fb->hot_x;
+		hotspot_y += plane->fb->hot_y;
+	}
+
 	du->cursor_surface = vps->surf;
 	du->cursor_dmabuf = vps->dmabuf;
 
@@ -411,6 +417,9 @@ vmw_du_cursor_plane_atomic_update(struct drm_plane *plane,
 		vmw_cursor_update_position(dev_priv, true,
 					   du->cursor_x + hotspot_x,
 					   du->cursor_y + hotspot_y);
+
+		du->core_hotspot_x = hotspot_x - du->hotspot_x;
+		du->core_hotspot_y = hotspot_y - du->hotspot_y;
 	} else {
 		DRM_ERROR("Failed to update cursor image\n");
 	}
@@ -2485,7 +2494,7 @@ void vmw_kms_helper_buffer_finish(struct vmw_private *dev_priv,
 	if (file_priv)
 		vmw_execbuf_copy_fence_user(dev_priv, vmw_fpriv(file_priv),
 					    ret, user_fence_rep, fence,
-					    handle);
+					    handle, -1, NULL);
 	if (out_fence)
 		*out_fence = fence;
 	else
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_mob.c b/drivers/gpu/drm/vmwgfx/vmwgfx_mob.c
index 941bcfd131ff..b17f08fc50d3 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_mob.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_mob.c
@@ -320,14 +320,14 @@ int vmw_otables_setup(struct vmw_private *dev_priv)
 
 	if (dev_priv->has_dx) {
 		*otables = kmemdup(dx_tables, sizeof(dx_tables), GFP_KERNEL);
-		if (*otables == NULL)
+		if (!(*otables))
 			return -ENOMEM;
 
 		dev_priv->otable_batch.num_otables = ARRAY_SIZE(dx_tables);
 	} else {
 		*otables = kmemdup(pre_dx_tables, sizeof(pre_dx_tables),
 				   GFP_KERNEL);
-		if (*otables == NULL)
+		if (!(*otables))
 			return -ENOMEM;
 
 		dev_priv->otable_batch.num_otables = ARRAY_SIZE(pre_dx_tables);
@@ -407,7 +407,7 @@ struct vmw_mob *vmw_mob_create(unsigned long data_pages)
 {
 	struct vmw_mob *mob = kzalloc(sizeof(*mob), GFP_KERNEL);
 
-	if (unlikely(mob == NULL))
+	if (unlikely(!mob))
 		return NULL;
 
 	mob->num_pages = vmw_mob_calculate_pt_pages(data_pages);
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_msg.c b/drivers/gpu/drm/vmwgfx/vmwgfx_msg.c
index 6063c9636d4a..97000996b8dc 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_msg.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_msg.c
@@ -244,7 +244,7 @@ static int vmw_recv_msg(struct rpc_channel *channel, void **msg,
 
 		reply_len = ebx;
 		reply     = kzalloc(reply_len + 1, GFP_KERNEL);
-		if (reply == NULL) {
+		if (!reply) {
 			DRM_ERROR("Cannot allocate memory for reply\n");
 			return -ENOMEM;
 		}
@@ -340,7 +340,7 @@ int vmw_host_get_guestinfo(const char *guest_info_param,
 
 	msg_len = strlen(guest_info_param) + strlen("info-get ") + 1;
 	msg = kzalloc(msg_len, GFP_KERNEL);
-	if (msg == NULL) {
+	if (!msg) {
 		DRM_ERROR("Cannot allocate memory to get %s", guest_info_param);
 		return -ENOMEM;
 	}
@@ -400,7 +400,7 @@ int vmw_host_log(const char *log)
 
 	msg_len = strlen(log) + strlen("log ") + 1;
 	msg = kzalloc(msg_len, GFP_KERNEL);
-	if (msg == NULL) {
+	if (!msg) {
 		DRM_ERROR("Cannot allocate memory for log message\n");
 		return -ENOMEM;
 	}
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c b/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c
index 7d591f653dfa..a96f90f017d1 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c
@@ -446,7 +446,7 @@ int vmw_user_dmabuf_alloc(struct vmw_private *dev_priv,
 	int ret;
 
 	user_bo = kzalloc(sizeof(*user_bo), GFP_KERNEL);
-	if (unlikely(user_bo == NULL)) {
+	if (unlikely(!user_bo)) {
 		DRM_ERROR("Failed to allocate a buffer.\n");
 		return -ENOMEM;
 	}
@@ -836,7 +836,7 @@ static int vmw_resource_buf_alloc(struct vmw_resource *res,
 	}
 
 	backup = kzalloc(sizeof(*backup), GFP_KERNEL);
-	if (unlikely(backup == NULL))
+	if (unlikely(!backup))
 		return -ENOMEM;
 
 	ret = vmw_dmabuf_init(res->dev_priv, backup, res->backup_size,
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_shader.c b/drivers/gpu/drm/vmwgfx/vmwgfx_shader.c
index 68f135c5b0d8..9b832f136813 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_shader.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_shader.c
@@ -751,7 +751,7 @@ static int vmw_user_shader_alloc(struct vmw_private *dev_priv,
 	}
 
 	ushader = kzalloc(sizeof(*ushader), GFP_KERNEL);
-	if (unlikely(ushader == NULL)) {
+	if (unlikely(!ushader)) {
 		ttm_mem_global_free(vmw_mem_glob(dev_priv),
 				    vmw_user_shader_size);
 		ret = -ENOMEM;
@@ -821,7 +821,7 @@ static struct vmw_resource *vmw_shader_alloc(struct vmw_private *dev_priv,
 	}
 
 	shader = kzalloc(sizeof(*shader), GFP_KERNEL);
-	if (unlikely(shader == NULL)) {
+	if (unlikely(!shader)) {
 		ttm_mem_global_free(vmw_mem_glob(dev_priv),
 				    vmw_shader_size);
 		ret = -ENOMEM;
@@ -981,7 +981,7 @@ int vmw_compat_shader_add(struct vmw_private *dev_priv,
 
 	/* Allocate and pin a DMA buffer */
 	buf = kzalloc(sizeof(*buf), GFP_KERNEL);
-	if (unlikely(buf == NULL))
+	if (unlikely(!buf))
 		return -ENOMEM;
 
 	ret = vmw_dmabuf_init(dev_priv, buf, size, &vmw_sys_ne_placement,
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_stdu.c b/drivers/gpu/drm/vmwgfx/vmwgfx_stdu.c
index c4de4ad0543b..ca3afae2db1f 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_stdu.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_stdu.c
@@ -1642,8 +1642,8 @@ int vmw_kms_stdu_init_display(struct vmw_private *dev_priv)
 		 * something arbitrarily large and we will reject any layout
 		 * that doesn't fit prim_bb_mem later
 		 */
-		dev->mode_config.max_width = 16384;
-		dev->mode_config.max_height = 16384;
+		dev->mode_config.max_width = 8192;
+		dev->mode_config.max_height = 8192;
 	}
 
 	vmw_kms_create_implicit_placement_property(dev_priv, false);
diff --git a/drivers/gpu/host1x/bus.c b/drivers/gpu/host1x/bus.c
index 7ece0e9058c6..f9cde03030fd 100644
--- a/drivers/gpu/host1x/bus.c
+++ b/drivers/gpu/host1x/bus.c
@@ -44,9 +44,12 @@ struct host1x_subdev {
  * @np: device node
  */
 static int host1x_subdev_add(struct host1x_device *device,
+			     struct host1x_driver *driver,
 			     struct device_node *np)
 {
 	struct host1x_subdev *subdev;
+	struct device_node *child;
+	int err;
 
 	subdev = kzalloc(sizeof(*subdev), GFP_KERNEL);
 	if (!subdev)
@@ -59,6 +62,19 @@ static int host1x_subdev_add(struct host1x_device *device,
 	list_add_tail(&subdev->list, &device->subdevs);
 	mutex_unlock(&device->subdevs_lock);
 
+	/* recursively add children */
+	for_each_child_of_node(np, child) {
+		if (of_match_node(driver->subdevs, child) &&
+		    of_device_is_available(child)) {
+			err = host1x_subdev_add(device, driver, child);
+			if (err < 0) {
+				/* XXX cleanup? */
+				of_node_put(child);
+				return err;
+			}
+		}
+	}
+
 	return 0;
 }
 
@@ -87,7 +103,7 @@ static int host1x_device_parse_dt(struct host1x_device *device,
 	for_each_child_of_node(device->dev.parent->of_node, np) {
 		if (of_match_node(driver->subdevs, np) &&
 		    of_device_is_available(np)) {
-			err = host1x_subdev_add(device, np);
+			err = host1x_subdev_add(device, driver, np);
 			if (err < 0) {
 				of_node_put(np);
 				return err;
diff --git a/drivers/gpu/host1x/dev.c b/drivers/gpu/host1x/dev.c
index 2c58a390123a..7f22c5c37660 100644
--- a/drivers/gpu/host1x/dev.c
+++ b/drivers/gpu/host1x/dev.c
@@ -134,8 +134,8 @@ static int host1x_probe(struct platform_device *pdev)
 
 	syncpt_irq = platform_get_irq(pdev, 0);
 	if (syncpt_irq < 0) {
-		dev_err(&pdev->dev, "failed to get IRQ\n");
-		return -ENXIO;
+		dev_err(&pdev->dev, "failed to get IRQ: %d\n", syncpt_irq);
+		return syncpt_irq;
 	}
 
 	host = devm_kzalloc(&pdev->dev, sizeof(*host), GFP_KERNEL);
@@ -186,8 +186,13 @@ static int host1x_probe(struct platform_device *pdev)
 			return -ENOMEM;
 
 		err = iommu_attach_device(host->domain, &pdev->dev);
-		if (err)
+		if (err == -ENODEV) {
+			iommu_domain_free(host->domain);
+			host->domain = NULL;
+			goto skip_iommu;
+		} else if (err) {
 			goto fail_free_domain;
+		}
 
 		geometry = &host->domain->geometry;
 
@@ -198,6 +203,7 @@ static int host1x_probe(struct platform_device *pdev)
 		host->iova_end = geometry->aperture_end;
 	}
 
+skip_iommu:
 	err = host1x_channel_list_init(&host->channel_list,
 				       host->info->nb_channels);
 	if (err) {
diff --git a/drivers/gpu/host1x/hw/intr_hw.c b/drivers/gpu/host1x/hw/intr_hw.c
index dacb8009a605..37ebb51703fa 100644
--- a/drivers/gpu/host1x/hw/intr_hw.c
+++ b/drivers/gpu/host1x/hw/intr_hw.c
@@ -33,10 +33,10 @@ static void host1x_intr_syncpt_handle(struct host1x_syncpt *syncpt)
 	unsigned int id = syncpt->id;
 	struct host1x *host = syncpt->host;
 
-	host1x_sync_writel(host, BIT_MASK(id),
-		HOST1X_SYNC_SYNCPT_THRESH_INT_DISABLE(BIT_WORD(id)));
-	host1x_sync_writel(host, BIT_MASK(id),
-		HOST1X_SYNC_SYNCPT_THRESH_CPU0_INT_STATUS(BIT_WORD(id)));
+	host1x_sync_writel(host, BIT(id % 32),
+		HOST1X_SYNC_SYNCPT_THRESH_INT_DISABLE(id / 32));
+	host1x_sync_writel(host, BIT(id % 32),
+		HOST1X_SYNC_SYNCPT_THRESH_CPU0_INT_STATUS(id / 32));
 
 	schedule_work(&syncpt->intr.work);
 }
@@ -50,9 +50,9 @@ static irqreturn_t syncpt_thresh_isr(int irq, void *dev_id)
 	for (i = 0; i < DIV_ROUND_UP(host->info->nb_pts, 32); i++) {
 		reg = host1x_sync_readl(host,
 			HOST1X_SYNC_SYNCPT_THRESH_CPU0_INT_STATUS(i));
-		for_each_set_bit(id, &reg, BITS_PER_LONG) {
+		for_each_set_bit(id, &reg, 32) {
 			struct host1x_syncpt *syncpt =
-				host->syncpt + (i * BITS_PER_LONG + id);
+				host->syncpt + (i * 32 + id);
 			host1x_intr_syncpt_handle(syncpt);
 		}
 	}
@@ -117,17 +117,17 @@ static void _host1x_intr_set_syncpt_threshold(struct host1x *host,
 static void _host1x_intr_enable_syncpt_intr(struct host1x *host,
 					    unsigned int id)
 {
-	host1x_sync_writel(host, BIT_MASK(id),
-		HOST1X_SYNC_SYNCPT_THRESH_INT_ENABLE_CPU0(BIT_WORD(id)));
+	host1x_sync_writel(host, BIT(id % 32),
+		HOST1X_SYNC_SYNCPT_THRESH_INT_ENABLE_CPU0(id / 32));
 }
 
 static void _host1x_intr_disable_syncpt_intr(struct host1x *host,
 					     unsigned int id)
 {
-	host1x_sync_writel(host, BIT_MASK(id),
-		HOST1X_SYNC_SYNCPT_THRESH_INT_DISABLE(BIT_WORD(id)));
-	host1x_sync_writel(host, BIT_MASK(id),
-		HOST1X_SYNC_SYNCPT_THRESH_CPU0_INT_STATUS(BIT_WORD(id)));
+	host1x_sync_writel(host, BIT(id % 32),
+		HOST1X_SYNC_SYNCPT_THRESH_INT_DISABLE(id / 32));
+	host1x_sync_writel(host, BIT(id % 32),
+		HOST1X_SYNC_SYNCPT_THRESH_CPU0_INT_STATUS(id / 32));
 }
 
 static int _host1x_free_syncpt_irq(struct host1x *host)
diff --git a/drivers/gpu/host1x/hw/syncpt_hw.c b/drivers/gpu/host1x/hw/syncpt_hw.c
index c93f74fcce72..7b0270d60742 100644
--- a/drivers/gpu/host1x/hw/syncpt_hw.c
+++ b/drivers/gpu/host1x/hw/syncpt_hw.c
@@ -89,7 +89,7 @@ static int syncpt_cpu_incr(struct host1x_syncpt *sp)
 	    host1x_syncpt_idle(sp))
 		return -EINVAL;
 
-	host1x_sync_writel(host, BIT_MASK(sp->id),
+	host1x_sync_writel(host, BIT(sp->id % 32),
 			   HOST1X_SYNC_SYNCPT_CPU_INCR(reg_offset));
 	wmb();
 
diff --git a/drivers/gpu/host1x/job.c b/drivers/gpu/host1x/job.c
index bee504406cfc..db509ab8874e 100644
--- a/drivers/gpu/host1x/job.c
+++ b/drivers/gpu/host1x/job.c
@@ -197,10 +197,6 @@ static unsigned int pin_job(struct host1x *host, struct host1x_job *job)
 		}
 
 		phys_addr = host1x_bo_pin(reloc->target.bo, &sgt);
-		if (!phys_addr) {
-			err = -EINVAL;
-			goto unpin;
-		}
 
 		job->addr_phys[job->num_unpins] = phys_addr;
 		job->unpins[job->num_unpins].bo = reloc->target.bo;
@@ -225,10 +221,6 @@ static unsigned int pin_job(struct host1x *host, struct host1x_job *job)
 		}
 
 		phys_addr = host1x_bo_pin(g->bo, &sgt);
-		if (!phys_addr) {
-			err = -EINVAL;
-			goto unpin;
-		}
 
 		if (!IS_ENABLED(CONFIG_TEGRA_HOST1X_FIREWALL) && host->domain) {
 			for_each_sg(sgt->sgl, sg, sgt->nents, j)